Tumgik
#if you’ve compared them I’m not mad I promise! I totally get why they’re grouped together when you haven’t played them
Text
There’s always been something that bothered me about whenever I try to explain the concept of Hatoful Boyfriend and most people’s immediate reaction is “oh! Like Doki Doki!”. My response is always just “…I guess?” because I could never quite explain why that felt wrong to me but I think I finally can put a finger on it now.
(To be clear, I’ve played and enjoyed both games. They’re both fun - this isn’t a criticism at all! I just think they’re, at their cores, two very different things. Also I don’t spoil anything for either game here beyond the obvious; that is, that these games are not what they seem to be at first glance, which I’m pretty sure most people know going in. If that amount of spoiler still bothers you, then stop reading here and play the games first!!!)
When people compare the two games, I 100% understand where they’re coming from. It’s the same gimmick I suppose: cutesy looking or joke visual novel actually turns out to be a smokescreen to hide the horrors within. But. Like. I feel like that’s where the similarities end.
See, DDLC’s whole point is to be shocking. It’s intended to make you jump and disturb you. It’s primarily a horror game with you, the player, as its focal point, and as a result the characters are only really there to be used by the story. The majority of the characters, with the exception of one, are pretty straightforward and lean heavily on the tropes commonly used in dating sims so the game can later distort them in increasingly horrific ways - with the intent of shocking and scaring you. The characters are more tools to be used to tell the story and create the experience than fully fleshed out people, which I think works out well thematically for DDLC. It’s effective in the sense where the characters aren’t really what draws you in, but rather the premise and the anticipation of getting a good scare out of it, with some existential stuff mixed in there for a pretty cool experience. You came to get scared right? Well that’s what you got.
But Hatoful is very different because its intent is not to shock or scare you necessarily. Hatoful, since its initial development, has been designed to make you care. Hato Moa made the game intending for people to start it because of its ridiculous concept, but then to discover it’s depth along the way. Hatoful doesn’t want to jump scare you. It wants to tell you a story. It delves into horror, that’s true, but the horror is much more psychological and it stems from the character’s actions. It’s a character driven story at its core, and it does this by taking itself seriously. It knows it’s completely absurd and leans into the absurdity, but the character interactions are real, so that when actual plot begins to happen, you’ll wonder when you started getting so attached. The characters twist typical dating sim tropes again, but this time to ask “hey why is this character like this? maybe it’s not so straightforward as you think”. And then the game goes a step further and reveals that there is an actual in-universe reason for its insane setup of “a human girl goes to bird school which is for birds and dates pigeons”. Hatoful is less tongue in cheek and more completely unabashed about it’s ridiculous premise and plays it straight - which in turn makes its serious moments shockingly genuine. That, to me, is why it works. The point is to delve into the characters. The plot is driven by the characters. It’s horrifying at times because the characters will sometimes make or be forced to make horrific choices. It’s not a gimmick. It’s a cohesive story. It’s just asking you to dig into it a little further to get there.
Tldr; DDLC is a cutesy facade and Hatoful is unabashedly absurd. DDLC wants to scare you and Hatoful wants you to care. DDLC uses its characters as tools for the experience while Hatoful’s whole point is the characters.
Editing based off a reblog that made me aware I was not completely fair with DDLC here (again, still trying not to spoil too much): The experience in DDLC is heavily based on its thematic content. When I say “the characters serve the experience” and “it wants to scare you” I don’t mean this shallowly, and I think it can and has been frequently misinterpreted in this way. In DDLC, the horror comes from the situation. In that sense, it’s thematically appropriate that the characters are tools in that way - that’s actually part of it and you are meant to feel bad for them - their situations, issues, and personalities are thematically relevant and sympathetic. I shouldn’t have juxtaposed “to scare you” and “to make you care”; that was far too simplistic a summary to the point where it is inaccurate. They both want you to care - a story only has power if you care about it. But I do still feel strongly that DDLC’s draw is the terrifying situation it impresses on the player, and the self-aware questions it asks (in which characters are put through the existential oppressive wringer to emphasize these themes), while Hatoful primarily relies on you slowly but surely acclimating to the absurdity so it can tell what is simultaneously an off the walls story that is also genuine and believable for the characters you’ve grown to know.
75 notes · View notes
imagineaworlds · 3 years
Text
I Love You (Part Forty-Seven) -- Aaron Hotchner
Written By: @desperately-bisexual
Request: None.
Warnings: Cursing. Mentions of PTSD, Dom/sub relationship, death, etc-- everything Criminal Minds.. I think that’s it???
Pairing: Aaron Hotchner x Greenaway!Reader
Word Count: 12575
Timeline: Season 7 Episode 01. Three months after part forty-six.
Tumblr media
For the past three months, the team had worked dozens of different cases across the country, but none of them mattered except for the one that Morgan, Garcia, and I had been working on privately. Since I came back to work and was deemed the temporary unit chief, I decided to reopen the Ian Doyle case. At first, it was just me. When I wasn’t at work, I was at home, hiding in the upstairs office, working on trying to piece all of the evidence together to figure out where the hell Doyle could have possibly gone. But then it was just too hard to keep doing it on my own, and all of my resources were wasted. There was only one person who had access to more information than I did. Garcia. So, I asked if she would be willing to help me, but it was impossible for her to keep a secret, so we invited Morgan to help us. From there, we managed to find Declan, but we didn’t tell anyone about it outside of our small group of three.
Finding Declan was an interesting process. I had hit a wall with it, hence why I acquired Garcia’s help, but she managed to find him almost immediately after she told Derek what was going on and he helped profile Emily in order to profile Declan. He figured that she had been the one to move him considering how she died protecting his identity and location. Before joining the team, apparently she lived in Reston, Virginia, something I didn’t know about her, but he managed to know. From there, we decided that it was probably to be close to Declan. So, we started narrowing it down. We knew that she liked cul-de-sacs, so as Garcia was going through the school system to find kids that looked like Declan, she compared if their guardian looked like Louise, his nanny, and if they lived in a cul-de-sac. Then, bam. We found them.
Morgan had been the one to offer up the idea of looking for Declan before Ian because if we found Declan, then so would his father eventually, which made our job easier. After that, it was just a waiting game. Unfortunately, there wasn’t much we could do beyond searching all of the footage to see if Ian ever showed up, which he never did. When Morgan and I were away on cases, Penelope kept an eye on him; and when we were in town but not working on anything, Morgan was driving baby Emily out there and taking her on walks around Declan’s neighborhood to scout the place. I think it was also just an excuse for him to have Emily more.
On our way to work one morning, Morgan and I stopped to get sandwiches for lunch, as well as some donuts from Leonard’s for the office since we were feeling generous, I suppose. At work, we split off so that he could sit in the bullpen while I dragged my feet up to my office. I didn’t understand how Hotch did it. Being away from the team all day, every day was so… isolating and exhausting. There was no joking with Morgan, teasing Reid, or… well, when Emily was around, there was no more chatting with her and gossiping about Hotch. I missed the way she used to slide in veiled threats into the conversation about how she would hurt him if he ever hurt me. So, then again, maybe it was a good thing that I wasn’t down in the bullpen anymore. Sitting at my old desk would have just given me a reason to sit there and stare at her desk all day, reflecting on what it was like to have her around, wishing that I could bring her back.
By lunch time, Morgan rescued our sandwiches from the fridge and brought them up to my office. He closed the door behind him and sat down across from me. I caught my sandwich as he threw it at me, then tossed him a water bottle from Hotch’s secret stash in his bookshelf behind the desk. I hadn’t really changed anything. For the most part, the office was still exactly how he left it. From the artwork to the family photos, to the stacks of case files that he practically left for me on his desk to work through, to his computer wallpaper of us, it was all still Hotch. The only difference was that I was finding all of his secrets now, like the water bottles, and totally taking advantage of them.
Morgan kicked his feet up on the desk, something Hotch would’ve never let him do in a thousand years, and dug into his lunch. I picked at mine for a moment while thinking. I was doing too much of that nowadays. Without my “other half” (or some cliché shit like that) around, I didn’t really know who to talk to 24/7 about what I was thinking and feeling. Sure, I could’ve talked to Morgan, but it was different. When I talked to Morgan, it was like we were brainstorming for a solution; but with Hotch, it was just like he was there to listen, and I could go on and on for hours, and he would hear me out, and he would only offer up a solution if he felt it were necessary. Sometimes I just liked ranting my thoughts to him. I liked watching how he would slowly get lost in the details of my face as I would be talking, and how he it looked like he was falling in love with me over and over again with every minute that passed. Skype and calling on the phone every day didn’t do that look justice anymore.
“Oh, my god—Oh, my god—Oh, my god!” Garcia came running into the office with a laptop balancing on one of her arms. Morgan and I turned to face her. “It’s him! It’s him!” She quickly set her laptop down on my desk and pressed play on a video. Morgan and I both leaned forward in our seats to watch the security footage from Declan’s house. “It’s Doyle—” She paused and pointed to a driver in a car passing by. It was kind of hard to see, but when she zoomed the image in and enhanced it, it was undeniably Ian Doyle.
“When was this?” I asked, jumping into action by standing from my seat and started collecting everything we had on this case thus far.
“Two hours ago.”
“He could have Declan by now. Fuck.” I looked at the two of them to say, “Okay, I think it’s time. I’m going to call Hotch to let him know, meanwhile, I need you guys to get the team together and start explaining everything. I want Reid and Rossi to go retrieve Declan and Louise after they’re briefed.” I grabbed my phone from my desk. “Go.” Just as they were leaving, I dialed Hotch’s satellite phone out in the Middle East.
Hotch and I had still been calling each other every day and Skyping as often as we could. The longer he was there, however, the busier he and I got with our two different jobs, which made it increasingly complicated to keep up with each other. But we made the effort. We knew that if at least one day passed where we wouldn’t talk, one day would inevitably turn into two, then three, and so on. I didn’t want to stop talking to him. As boring as the conversations could get at times, at least I got to see him. He was growing out a beard that I happened to love, but Jack hated. He claimed that it made his dad look old. When we could talk in private a little bit later, I told Hotch that I thought it was hot, and that I couldn’t wait to pull at it while he towered over me. He disliked when I teased him like that, though.
When I called, he picked up almost immediately.
“You’re calling awfully early,” Hotch said over the phone, “are you okay?”
“I’m fine. Do you have a minute?”
“For you? I have all the time in the world.”
I blushed to myself and looked down to watch the toes of my shoe dig into the ground over and over again. “There’s something you need to know, baby, and you have to promise to not get mad.” He didn’t say anything. “Morgan, Garcia, and I have been looking for Doyle while you’ve been gone.” Still, nothing. “We found Declan.”
“What?!”
“Hotch—”
“Why didn’t you tell me about this, Y/N? Do you know how stupid that was? Your whole career could—”
“I don’t care about that, Hotch. I really don’t. I knew that finding Declan was the best way of finding Ian, so I asked Garcia to help; and you know her, she couldn’t keep it to herself, so she told Morgan, and he insisted that he help us. He’s basically the one who found Declan. Without the two of them, I wouldn’t have any leads.”
Hotch sighed quietly. “Is Declan safe?”
“Yes.”
“How can you be sure?”
“Morgan’s still surveilling him—But, Hotch, listen, we were right. By watching Declan all this time, we got eyes on Ian.”
He hesitated for a moment. “You’re sure?” He sounded more scared than angry now. I hummed a “yes” in response. “Baby, listen to me right now.”
I bit my lip. “I’m listening, Sir.”
He chuckled slightly. “If you get eyes on Doyle, you need to take the shot. Got it? Don’t let him get away again.” He sounded out of breath now like he was running around. “I’m coming back to help.”
I felt my heart skip a beat and the air leave my lungs with one shocked gasp. He was coming back. After three months, I was finally going to get to hold him in my arms again, and I was going to get to kiss his lips again. Holy shit. He was coming home. My husband was coming back to me. I almost wanted to dance and cheer, scream from the top of my lungs how happy I was and that I couldn’t wait to see him; but I had to show a modicum of respect still, so I buried that feeling until I would get to see him again.
“You can do that?” I questioned cautiously. As happy as I was, could he really just abandon his post to come back like that? It sounded almost too good to be true.
“I put in the request to transfer back to Quantico about two weeks ago.” And he didn’t tell me? “I didn’t tell you because I didn’t want to get your hopes up on anything; but they don’t need me here anymore, so it’s not a big deal if they find a replacement.” I hopped slightly with joy and giggled. Hotch chuckled at me. “I can’t wait to see you, baby.”
Even considering the circumstances, I couldn’t wait to see him either. “Call me when you’re landing.”
“I will. I love you.”
“I love you, too.”
----
By the time Garcia followed traffic camera footage across the state just to see where Doyle was hiding out, we had a SWAT team with snipers on call, and the team was ready to head out there. Just as I had ordered, Rossi and Reid headed to get Declan, but the drive was pretty far, so I didn’t expect to hear from them for a while, especially while we were in the early stages of staking out Doyle’s place. Morgan, JJ, and I were trapped in a van across the street from his apartment while SWAT was just down the road, and they had two snipers up on separate roofs. All of it was my call, though. The second we had reasonable cause to believe that it was actually Ian Doyle up there, I would be the one to dictate if the snipers or Morgan would be taking the shot, like Hotch said.
“That’s all Hotch said? ‘Take the shot’?” JJ questioned warily. I nodded and shrugged. “You’re sure?”
“I’m sure. Why?”
She shook her head. “Nothing…”
I eyed her suspiciously for a moment, taking into account how her gaze wasn’t resting on one thing, but instead a thousand different things, like she was too anxious to focus. It was odd, but I shrugged it off with the excuse that finding Doyle was a big deal. Since he killed Prentiss, we had all been itching to find him and avenge her. Out of everyone, I knew that JJ would want to know that Doyle was dead because of us. She wanted that justice. I couldn’t exactly tell why, but she had always been like that, especially since getting promoted to being a profiler.
And then my phone started ringing with a call from Spencer who had just arrived at Declan’s house. I figured it was just an update. I mean, him and Rossi were there to simply pick up Declan and Louise, that was it. But then something worrying happened when I picked up the call: Reid sounded concerned.
“Reid?”
“Declan’s missing,” he answered quickly.
“What?”
“We just searched the house, and he’s gone, so we called the school to see if he even made it home, and they told us that his mom picked him up early.”
“Louise?” I speculated.
“I doubt it.” He sounded annoyed now. “She’s dead.”
“Fuck,” I cursed under my breath, then looked up at Morgan and JJ to tell them what was going on. They also cursed, and he hit the side of the car. “Okay. Reid, you and Rossi should meet us here in case we end up seeing—” The sound of a police car’s siren flew past our van, catching our attention. It was just a squad car. No, no, no, no. Fuck. That was going to scare Doyle off. Fuck.
“He’s inside!” Morgan exclaimed. We all turned to face the window to see the drape in Doyle’s apartment window shaking like it had just been disturbed. “I saw him!” Morgan was already fiddling with the door handle and pushing his way out of the car.
“Shit…” This wasn’t at all how I imagined this going down. We were supposed to just monitor the situation, get legitimate confirmation that Doyle was in there, not just Morgan claiming he saw him— But what choice did we have now? “Go, go, go,” I made the call, following Morgan out of the van and hanging up the call with Reid.
JJ, Morgan, and I raced inside first, the SWAT van pulling up behind us and the team moving in, too. When we reached the apartment number, JJ and I stood our ground in the hallway with our weapons raised, and I gave Morgan the signal to kick the door down, which he did without hesitating, and then he dashed inside. The three of us took turned towards different sides of the apartment in order to clear it. Morgan went straight into the living room/bedroom, JJ went into the bathroom on the left, and I stepped into the tiny kitchen on the right. He wasn’t there. Neither was Declan. Reid said that they couldn’t find Declan, so I only assumed that Doyle got to him first. Fuck! How the fuck did he—
“I’ve got something!” Morgan called out. I hurried into the living room/bedroom to see him peeking up at the ceiling inside of the closet. I cocked a brow and carefully approached. “He’s climbing to the roof.” There was a hole in the ceiling for some kind of secret tunnel that led upwards.
“Follow him,” I ordered. “I’ll take the steps. JJ, watch the door downstairs to make sure he doesn��t get out, and coordinate with SWAT to make sure that the block is locked down just in case. Get a crime scene team here to see if Declan was ever here.”
Morgan holstered his weapon then quickly pulled himself up into the tunnel to follow Doyle, meanwhile I hurried out into the hallway to start sprinting up the rest of the stairs, and JJ ran down them. How the fuck did this happen? How did we lose Declan like this, and how was Doyle actually planning on getting out of this?
And then I burst out onto the roof to see Morgan chasing after Doyle who was running for the fire escape on the side of the building. Morgan aimed his weapon, but didn’t shoot. Instead, he tried commanding Doyle to stop. Like that would help.
“I’ve got the shot,” one of the snipers said over the comms.
“Don’t shoot!” I exclaimed. “Don’t!” I knew what Hotch said about taking the shot if given the chance, but things were different now that we couldn’t find Declan in the apartment. Our only way of getting him back now was by keeping his father alive. Asshole. “Hold your fire!” When nothing came, Morgan tackled Doyle and spun him onto his stomach so that he could be handcuffed. “You got it?” I asked him. He nodded while slapping the cuffs on Doyle. “I’ll have JJ take him back to the office while we start going through everything downstairs.”
Morgan groaned as he yanked Doyle up onto his feet. “You’re not gonna let me take a crack at him?”
“Not yet.”
Doyle chuckled. I squinted at him while searching his eyes for a reason that he could have possibly found this funny. He noticed my curiosity. “Lauren was right about you.”
Morgan tugged on Doyle’s restraints to make them hurt. “Shut up.”
Doyle kept laughing as Morgan pushed him towards the stairwell so that JJ could take him to the office like I had ordered. My only question, however, was, what the hell did he mean by that? Lauren was right about me? Right about what? The question lingered in my head as I sighed and moved to walk downstairs, too.
Afterwards, hours passed while we were combing through Doyle’s apartment, trying to find clues about where he must have taken Declan, as well as compiling evidence connecting him to Valhalla and Prentiss’s death. For someone in hiding, there was a lot he had with him. Usually, people in his situation would stay minimalist and wouldn’t have all of their condemning evidence in one spot, but Doyle was so cocky that he wouldn’t get caught that he didn’t bother to keep any of his current business elsewhere. There were so many guns in that one apartment that we could have started another small army. I couldn’t understand how on Earth someone would need all of that. Then again, Doyle already had a small army, and the reason they needed to smuggle all of those weapons out of the state was so that he could maintain power in Europe while he was underground. Well, looked like all of that was worthless now.
I sighed as I stepped outside of that tiny, dirty, smelly, cramped apartment for the first time in over ten hours. It really took that long. Doyle had been sitting at the office the entire time, brewing in his own thoughts, which was our strategy to break him. If he was left in darkness and silence long enough, he would open up. Hopefully. At least, Morgan and I would give it a shot when we would head back to the office. For now, we were down tearing the entire apartment to shreds. The bastard even had papers hiding behind the wallpaper.
When my phone started buzzing in my pocket, I sighed again and dug it out. “Hotchner,” I answered.
“Mmm… I could get used to that,” Hotch’s voice echoed over the phone, making me melt.
I perked up within an instant while looking down at my watch, spotting the time and how long it had been since he last called me, and how he promised he would call when he was close to landing in Virginia. Finally, I thought. “How close are you?”
“Ten, fifteen minutes.”
It was almost morning already. He was probably going to be landing just as the sun would start to peek over the tall buildings of Quantico. “I’m coming to pick you up.”
“Y/N, you should stay—”
“Nope,” I interrupted adamantly.
Hotch chuckled. “You’re really going to be a brat about this right now?”
“Mhm. You can scold me for it when you see me.”
“God, I love you.”
“I love you, too.”
“Hey, sunshine,” Morgan said, sneaking up behind me, “you ready to go?”
“I’ll see you in a few minutes, baby,” I told Hotch. Before he could say anything in return, though, I hung up on him. I hid my grin as I buried my phone in my back pocket, then turned to face Morgan. “Can you handle the interrogation for a bit while I head to the airport?”
“He’s back already?”
“Yeah.”
Morgan nodded. “Go get him.” We smiled at each other. “No detours, though, sunshine. There and back.”
“Yeah, yeah,” I responded half-heartedly while already walking away from him. “Just remember, when this is over with, you and Clooney are going home!”
----
To say I was nervous would have been a drastic understatement. I had gone four months without seeing him. Four months without his touch, his kiss, his hold, his love. A lot could change in four months. People could grow apart in four months. I was absolutely terrified that even though we called each other every day, we might not have been the same Hotchners who said goodbye to each other at the airport a few months ago. I knew that I still loved him endlessly— maybe even more now that I had been starved of him for so long. But I didn’t know what happened to him out there. I didn’t know what new experiences he had that could have possibly changed him. For all I knew, he could walk off that plane in a few minutes and walk right past me without a care. I just prayed that he was still Aaron. That the man I fell in love with would walk off that plane and know exactly how to hold me again. It was like I could already feel his touch, even though the plane’s engines hadn’t died out yet. He was so far, but he was so damn close. This was the closest we had been in four fucking months, and I was just so ready to see him already.
The plane came to a slow and steady stop on the runway. With every second that passed, I felt a growing need to throw up everywhere. He was probably sitting in there, watching me through the window, grinning ear to ear because I looked so nervous and sick. He could likely see me, yet I couldn’t see anything but the sun in my eyes. He was always an asshole, but he even had to beat me to the first look. Asshole x2.
The door of the jet popped open, and it started slowly falling down towards the concrete, the stairs on the inside making a slow appearance. I thought about what was going to happen. How was I going to react? How was he going to react? Was it better to stay where I was or was it better to meet him at the bottom of the steps? I felt like even if I wanted to move towards the plane, I wouldn’t have been able to. I was frozen in place as my mind raced with questions and endless possibilities. But there was one thing that was consistent: I was going to hold my husband in my arms, and I was going to tell him that I loved him over and over again until I passed out.
The second I saw a foot step out of the jet, I started sprinting as fast as I could without stopping to think. The rest of him emerged, and Hotch looked around for me. After spotting me from the top of the stairs, Hotch jumped down the steps, dropped his bag on the runway, and started running towards me, too. He held his arms out, an invitation for me to jump and hold him tight, so that was exactly what I did. Our bodies clashed, but he still managed to catch me, even though the force sent him stumbling back a few unwanted steps. My arms flew around his shoulders, my legs wrapped around his hips, my face buried in the crook of his neck. It didn’t matter that he smelled like sweat, sand, and cheap deodorant, because he was finally home with me and that was all that mattered. Hell, I even enjoyed the smell since it just meant that I had him in my arms again.
I squeezed him tight with my arms and legs after he nuzzled his face— beard, sunglasses, and all— into the crook of my neck. The aching in my chest, my broken heart that had been falling apart during every second that he was away from me, slowly started to mend with every breath we took together. I couldn’t tell if I wanted to cry, scream, cheer, laugh, or all of the above. Every single emotion imaginable was flooding my body and the only way I could cope with it was by nearly squeezing the life out of him while repeating: “I love you, I love you, I love you” over and over in his ear. I never wanted to stop saying it. I prayed that he would never let me go again and that we could say those three words until there was no more breath in our lungs.
“I love you, I love you, I love you,” we kept whispering like we were records stuck on repeat. Fuck, I missed the way he said it to me. I missed the way he said my name and the way he called me “baby”. Nothing felt more right than how he spoke to me and how he kissed my neck in between words.
When his beard started tickling my neck, I finally paused to let out a giggle. I leaned back, his arms still around my torso to hold me for support, and I got a good look at him for the first time in months. There was no screen between us, no shitty computer cameras, no god awful Skype buffer, no speakers that cut out every other word. He looked so different than how I remembered him. His face was slimmed behind the scratchy beard he had grown, and his eyes were more sunken with exhaustion— if that were even possible. Aaron Hotchner always had dark circles under his eyes, that was no secret. Between work, the kids, and wrangling me, Hotch never had time to sleep; but Pakistan kicked his ass, and I could see it in every little detail about him.
He watched me with the brightest smile I had ever seen on his face as I brought my left hand up to his face and scratched my nails gently down his beard. “This is even hotter in person,” I smirked.
“It has to go the second I get an electric shaver,” he laughed.
My smirk fell and I pouted, “Are you sure?” He nodded. “But I have literally never wanted to fuck you harder in my entire life.”
He paused and gulped. “Okay, so maybe it can stay for a few days.”
I perked up again at my little victory and scratched it again. “Please don’t ever leave me again. Please.”
“I’d rather die than go away again.”
“Don’t be hyperbolic, Agent Hotchner.” I hit his shoulder playfully. He grinned before tightening his arms around me again to make me lean in for a kiss. I grabbed his face, the hairs on his cheeks tickling my palms as I did so, and I moaned into his mouth. “I love you so much.”
His eyes screwed shut as he slid his tongue into the kiss and claimed dominance. I missed that feeling so much. I missed how he tasted, how he smelled, how he felt, how he had to dominate me every second of every day. I missed getting lost in his eyes and staring at his lips all day until I could finally kiss them until our lips were chapped. If there weren’t a million and one things going on outside of that airport, I would have begged him to take me inside and fuck me in the bathroom. It should have been impossible to need him that much considering everything that was going on with Ian Doyle, yet there I was, only thinking about showing Aaron Hotchner how much I loved him.
Hotch pulled away from my lips, but he kept his nose pressed to mine. “Emily?”
“At home with Jessica.”
“Jack?”
“At school.”
“You?”
“The most relieved I have ever been in my life.” I leaned in and kissed him again. I thought to myself: “Please, never let this end. Let us stay trapped in each other’s arms forever. Never let us get tired of kissing each other, of holding each other, of saying ‘I love you’. Never let the desperation for passion die. Never let me miss him ever again. Keep him by my side until we die of old age. Please.”
And just like he could read my mind, Hotch loosened his hold on me ever so slightly, making me drop down to my feet, and he said, “We should get back to the office.”
I shook my head. “Just… One more minute… Please.”
“One more minute will turn into an hour at least, you know that.”
“Is that so bad?”
“No, it isn’t.” Hotch cupped my cheeks and kissed me as hard as he could. After a short minute of him kissing me like he had been starved of it for years, Hotch pushed me away. Our hands fell to our sides and I bit my lip while I tried to focus on not jumping on him again. “We have to go.”
I tried to catch my breath, my chest rising and falling at a dramatic pace. It took every ounce of strength and self-restraint I had stored away to not jump back on him and kiss him again. I felt like I was going to cry because he wasn’t in my arms again. I felt like screaming and kicking— throwing a tantrum like Jack would. I felt like at any moment, the two of us would break, and I would tackle him to the ground before showering him in kisses and pinning him down so that he couldn’t leave me again.
Hotch stumbled away from me to grab his go-bag from where he dropped it because he had the same look I had, and I knew that he was sharing the same thoughts. When he came back, he took my hand, and we walked towards the car. I started catching him up on everything we knew about the case thus far. Ian and Declan, all of Ian’s business papers that he had in his apartment. I realized then just how little we actually knew. Hopefully by the time we got back to the office, Morgan would have something out of Doyle that would help.
I drove the whole way to Quantico because he was too tired from the flight and the time zones. He had one of my hands trapped in his instead of on the steering wheel, and he kept kissing my knuckles again and again. He wasn’t listening to me. Not that I blamed him. If he wanted to just talk shop and I was sitting there, not distracted by driving, I would’ve been antsy to shower him in love, too. But Morgan said no detours. As much as it pained me, he was right that there really wasn’t any time to drive home, or even pull to the side of the road to fuck Hotch. I really, really wanted to, though; so, I figured the longer I distracted myself with work, the better off we would be. After this, though… Once this was all resolved… I wasn’t letting him go.
At the office, Hotch and I headed up to the sixth floor, somehow managing to go that entire time without touching each other. I wanted to kiss him again. I wanted to hold him again. While we were standing in the elevator, I stared at him and thought about pushing him against the wall, but it wasn’t worth it right now. I hated this inner battle I was having to fight between wanting him all to myself in that very moment and just focusing wholeheartedly on the case. The worst thought hit me suddenly… Maybe it would have been better if he didn’t come back so soon… I mean, he was just so distracting, I felt like I couldn’t celebrate this win as much as I should’ve been.
And then we stepped onto the floor.
“Hey.” Hotch grabbed my hand, tugging me back a few steps towards him. “You trust me, right?”
I furrowed my brows as I turned to face him. Of course I trusted him. Why wouldn’t I? He was my husband, and we made a promise to each other a long time ago to not keep any secrets from one another. I had no reason to doubt him. I wouldn’t have met him at the airport like that if I didn’t trust and love him wholeheartedly.
“You know that I do everything for a reason?”
What was he on about? Was there something I missed between meeting him at the airport and walking into the building with him? Why was he suddenly acting so weird?
I put my hands on his biceps. “Listen. I know that you told me you were in an all guys unit. If there’s something you want to say about it, I’m all ears. And I’m going to support you no matter what.” I bit back a smile.
Hotch cracked a smile. “No,” he shook his head, “that’s not what I mean.”
“Phew.” I rolled my eyes playfully. “I thought I was going to have to give you the birds and the bees talk, and how love is love—”
“Shut up.” He laughed before kissing me.
I pushed him back gently. “I’m just saying, if you’re going to cheat on me, it should at least be with a guy so that you get to experiment a bit.” I patted his chest.
He was still laughing quietly. “What am I going to do with you, Agent Hotchner?”
“I don’t know, Agent Hotchner,” I responded as I let him pull me back in for another kiss. “I’m sure you’ll think of something before we get home after this case, though.” I pecked his lips again, then escaped into the board room.
Everyone was standing behind the round table, huddled together as they discussed something about the case. It seemed as though Morgan had just finished his first attempt at getting information out of Doyle, to no avail. They were discussing how he was going to approach the second try, but no one’s ideas seemed to stick since Morgan had an excuse for why none of it would work. It was like we had hit a dead end, even though Doyle was the end, technically. He was the one who took Declan, and he was the one who took our sister from us. Prentiss was dead because of him, and we weren’t going to let him get away this time.
When Hotch walked into the room, everyone fell silent and turned to get a look at him. Hotch really didn’t seem to care about our no PDA at work rule now, because as everyone’s eyes lingered on him and his beard, Hotch came over to me, took my hand in his, and kissed my cheek. A lot changed for him in the Middle East, I guess. He missed me and he didn’t care what rules we had. He wanted to just have me around all the time, and I couldn’t blame him. But, technically, I was still the unit chief until Hotch could officially come back, and I couldn’t let us break the rules. Now I understood why Hotch was always so adamant about me behaving while at work. It was excruciatingly painful to not touch and kiss him just because I was the boss and had to set an example.
I moved away from Hotch, taking my seat at the round table. Everyone followed my lead warily, their eyes still trained on Hotch. Him and JJ didn’t sit, though. They were the only ones who stayed on their feet at the front of the room. Hotch had his arms crossed, a frown hiding under his facial hair, his eyes wandering around the room as he thought nervously about something. This had to do with him asking about my trust, but I didn’t understand how. Him worrying about something and not telling me almost immediately was only making me worry, too. He needed to spit it out fast or I would go insane.
“Everything alright, Hotch?” Morgan asked, also taking notice of Hotch’s clear unease.
Hotch didn’t nod or give any kind of reassurance that everything was alright. Instead, he shifted his weight on the balls of his feet— a silent tell that something was definitely wrong. He kept his head down now, only glancing up through his lashes occasionally to look at me. A thought struck me that maybe my joke wasn’t really a joke in the hallway. Four months was a long time to go without me, and it probably didn’t help that all he got was shitty phone sex. Maybe he really did change out there in the desert. Maybe he was showering me in love because he was going to drop a huge ass bomb that said: “I’m leaving you” or “I’m quitting the BAU and running away forever” or “I’m taking the kids and you’ll never see them again.” Every shit scenario possible was racing through my head. Like I said, if he wasn’t going to spit it out soon, I was going to start screaming for answers.
“Nine months ago,” Hotch began, “I made a decision that affected this team. As you all know, Prentiss lost a lot of blood after her fight with Doyle. We told you that she succumbed to her wounds…” He hesitated a beat. “That’s not really the case.”
Shock filled the room as it dawned on each of us what Hotch was really saying. I didn’t want to jump to any conclusions because I really, really didn’t want to get my hopes up… but it sounded like he was telling us that Prentiss was alive. That over the past seven months, we have been led to believe that she died in that hospital.
“The truth is, the doctors were able to stabilize her, and she was airlifted to Bethesda under a covert exfiltration.”
My shoulders fell. So, it was true. Emily Prentiss was alive. The woman we mourned the loss of for seven months was… still around all along? That was why Hotch asked if I trusted him. That was why he was being so handsy. He knew that I hated it when he kept things from me. He knew that I hated it when he would betray my trust like this. I yelled at him for weeks the last time he did it— which was our suspension over two years ago. He was asking about trust and touching me as often as he could because he didn’t know if this admission of the truth was going to break us apart. This wasn’t like him lying and going to ask for a transfer out of the BAU. This wasn’t like him lying to me about his health. This was Emily Prentiss. This was our sister we lost in the field. We buried her. Why did Hotch… What… I— My thoughts were too scrambled to form another cohesive thought other than: “How could he do this to us?”
“After she got better, she was reassigned to Paris, where she was given multiple fake identities so that she could be safe,” Hotch continued.
“She’s alive?” Spencer finally asked the clarifying question we were all dying to know the answer to.
Hotch nodded shortly, keeping his eyes lowered. He wasn’t even looking at me now because of the shame and guilt he likely felt.
“But we buried her,” Spencer croaked.
Hotch nodded again. “If anyone has any issues with the executive decision that I made, then they can take it up with me.”
“Issues?” Morgan asked angrily. “Issues, Hotch? Yeah, I’ve got a few issues, but why don’t you start with the fact that you let your own wife believe that Emily was dead and let them name your daughter after her because of it, hmm? Don’t you think that you owe Y/N an explanation?”
“Morgan,” I hissed quietly. It was neither the time nor the place to discuss my daughter. I had a few choice words for Hotch running through my head, but I fully intended on keeping them to myself until Hotch and I could speak privately.
“Emily’s alive, Y/N. How are you going to explain that to your daughter as she grows up?”
“Stop it,” I demanded more harshly.
There wasn’t anything Morgan could say to me that I didn’t already know. Hotch had betrayed everything we believed in and promised each other. He had told me countless times since meeting me that he would never lie to me, and every time he ended up breaking that promise, he would tell me that it wouldn’t happen again. But there we were. Another promise broken. Even worse, though, was the fact that he made a vow to me on our wedding day that he would always be honest with me. Always. And yet he kept this a secret from me. Did he not trust that I could have kept it a secret? Morgan was right, I was owed an explanation. Hotch kept this from me, his wife for nine months. What else was he hiding? How was I ever supposed to trust him again?
I stood from my seat and turned for the door. No one said anything as I took a step but came to a sudden halt when I discovered that my path was blocked by someone in my way. I felt like I was going to pass out. There was she was: Emily Prentiss… standing right in front of me. Back from the dead.
She let out a sigh of regret, but also relief, when our eyes met. She looked sorry about how all of this happened. And she should have been. Morgan sat on our couch for months, crying because he missed her so much. I named my daughter after her because I missed her so much and I wanted her memory to be remembered. What was it all for, though, now that was standing there? What was the point now? Morgan was right to be vocally angry because he knew that I couldn’t. But as mad as I was at Hotch, I felt an overwhelming need to hug her. I crashed into her. I didn’t hesitate. Nine months without her. Nine fucking months.
Emily caught me as I embraced her as tightly as I could. She rubbed small circles on my back as she held me close. “I am so sorry, Y/N.”
I sniffled into her shoulder while I hugged her tighter. She was there. Like, really there. I wasn’t dreaming or hoping anymore. It was her. Our family was back. And I was just so relieved. I couldn’t even be mad at her. But I could tell as I pulled away from Emily and turned to look at the room with her that Morgan was pissed beyond compare. Everyone was crying because they were so grateful for the fact that she was still alive. However, Morgan wasn’t. He was standing there, pouting with disbelief, his hands curled into shaky fists that were holding in all of his anger.
She carefully let go of me and approached Morgan. “Derek, you have to believe me when I tell you that not a day went by where I didn’t think about you guys and how sorry I am that things turned out this way. You didn’t deserve that.”
“I didn’t deserve to know that you were alive?” he questioned through gritted teeth.
“You didn’t deserve to hold a dying friend in your arms.” Prentiss extended her arms for another hug, a silent offer for him, which he took slowly, as if he were afraid that he would fall right through her like she was some kind of ghost. When they embraced, however, I saw Morgan’s anger wash away. Now, he was just broken. “I’m so sorry.” He squeezed her waist as tight as he could for as long as he could before she tapped out and parted from him. She turned to face me again. “I want to meet baby Emily, if you’ll let me, once this is all over.”
I nodded, still too hesitant to say anything.
She smiled. “Okay… For now, I just want to focus on finding Declan and bringing him home safe. What do you guys know so far?”
Reid immediately jumped to his feet like this was his moment to shine, and he started asking her questions about Declan, Ian, and Louise, but I couldn’t hear anything. Everything seemed so washed out and distant. The only thing that was clear to me was the shame on my husband’s face as I stood there, staring at him, silently letting him know just how badly he had fucked up this time around. He could hardly look me in the eyes.
“Can I see you in my office?” he asked.
Everyone watched us silently, wondering if I would correct him on the fact that it was my office now. Or maybe they were waiting with anticipation to see if I was going to blow him off considering I had been glaring at him the entire time since I found out that Prentiss was alive this whole time—and he knew! But that was exactly why I indulged him. I didn’t correct him, I didn’t argue with him, I didn’t embarrass him. I simply followed him to his office.
I sat in my chair at my desk, which all used to be his before he left. He sat down across from me in one of the seats I would sit in whenever I was called to meet with him in his office. He seemed so out of his element in that chair, and, honestly, I was glad. I wanted him to be uncomfortable. I wanted him to be physically and emotionally uneasy because that was how I felt every day while he was away, and that was how I felt since finding out that our daughter’s namesake was actually alive and well.
“Were you ever going to tell me?” I inquired, reclining back. He shook his head. “You were going to just let me think that she was dead in order to, what, make me feel better about naming our daughter after her? Why didn’t you try to stop me—”
“I did try. I asked you if there were really no other names you could think of, but you were so adamant about it and I couldn’t tell you the truth; so, I just had to let it happen.”
“Is that why you left?”
Hotch froze for a moment while searching my eyes, and before he even began nodding, I knew the answer. He sighed. “It was hard to keep lying to everyone, but it felt impossible to keep lying to you specifically, especially since you took Emily’s death so hard… So, I just… I ran.”
“So, you would have stayed if I knew the truth?”
“Yes.”
I scoffed and let out an exasperated chuckle.
“Baby, I am so sorry, you have to believe me. I wish things hadn’t happened the way they did.”
“Yeah, well, you had a choice in that, didn’t you?”
“Y/N, that’s not fair. I was just trying to protect her—"
Suddenly, there was a knock on the door, proceeded by Garcia cautiously stepping into the office. “Sorry, ma’am, but I found something.”
I nodded her over, holding my hand out for the file of information she put together. Hotch eyed me. Usually, it was him sitting in my current seat, ignoring me in his seat as he accepted another file from Garcia who had addressed him in the first place. Everything felt so backwards. But, then again, maybe that was what he deserved now that I knew that he had been lying to me this entire time.
“I was narrowing down a list of Doyle’s top ten enemies from what we know and what Morgan’s getting out of him. The only one who’s been in the states recently is Mr. Richard Gerace. He’s been here for the past two weeks with a work visa.”
“Is there any way to connect him to this?” I asked while flipping through the file.
“The guy who cut off the camera feed at Declan’s house had a scar on his neck...” She turned the next page for me and pointed to a mugshot of Gerace. He had the same exact scar on his neck, meaning it was definitely him who took Declan.
“Get me everything you can on him, please, Penelope,” I said while closing the file and setting it down on my desk. She hesitated. “What is it?”
“That’s everything I have.”
I sighed. “Can you call Prentiss in, please?”
“Sure…”
“Thank you.”
When she left, I sighed and looked at Hotch again. Neither of us said anything as we stared at each other uncomfortably, a barrier of trust broken between us now. I didn’t like it when he lied to me. Every time I told him not to lie to me, he promised he wouldn’t, and then he would, and I would be mad for a few days before forgiving him and moving on. How much longer was I supposed to put up with it? I couldn’t even count how many times he lied to me on my fingers anymore. Meanwhile, I could count the number of times I lied to him on one hand. At some point, enough was going to be enough, and if he wasn’t careful, it was going to be soon. As much as I loved him, I couldn’t keep living with the lies. What was more important to him? Me or protecting his secrets? If the answer wasn’t immediately me, then there was an entirely different conversation we needed to have at some point that included what our future was going to look like.
“We’re going to talk later,” I finally told him.
As we were coming back to the office together, Hotch stopped to ask if I trusted him, and at the time, of course I did… But now? I wasn’t so sure. Then again, I’d said that a dozen times before when he pulled this shit. The worst part was, that I knew that I was going to end up crawling back to him, and it was going to be an excuse for him to lie to me all over again.
Hotch nodded. “Yeah…”
There was another knock at the door, this time from Prentiss, who seemed too shy to invite herself in, even though I had requested her presence. I waved her in.
“Gerace,” I said, holding the file out for her. She slowly walked over and accepted it. “Garcia thinks that he’s the one who took Declan.”
“That doesn’t make any sense,” she insisted before even looking at the file. I waited for her to explain why she thought that. “Gerace gave up on Doyle a long time ago, and he’s too much of a coward to pull something like this off. Not to mention how meticulous the abduction was. Gerace doesn’t have that level of patience and organization.”
I scoffed and wiped my palms over my face. “He was our only suspect and lead.”
“I mean…” She flipped through the file. “That’s definitely Gerace’s scar… So, I suppose he had something to do with it, but I wouldn’t say he’s solely responsible. If he’s working with a partner, he’s definitely the submissive.” She flipped another page. “Hold on—” I looked up at her. “Look at this.” She turned the file to show me one of the shots from the security footage of Gerace turning the cameras off. “Look in the background.” There was a woman standing there, watching Gerace while cocking a gun at her side. Prentiss’s jaw dropped as a realization dawned on her. “This might have something to do with Declan’s birth mother, considering the overkill towards Louise, who was Declan’s mother figure throughout his entire life.”
“Do you know who she is?”
She shook her head. “No. Doyle never told me.”
“Would you be willing to go in and get the information out of him now?”
She looked between me and Hotch, almost as if she were waiting for permission from him, but then she recalled that I was calling the shots right now until Hotch could be reinstated into the unit by Strauss and Cody. She finally gave in. “I don’t mind giving it a shot.”
Since Prentiss was still just a visitor, she couldn’t technically be allowed to wander the floor on her own. Even though she knew exactly where the interrogation room was and how to get there the fastest, I still had to usher her there. So, without saying anything to Hotch, I started walking out. Prentiss watched him silently for a moment before deciding to follow me.
“You know,” she said while catching up to me on the ramp outside of my office, “unit chief suits you.”
“It won’t last long. It’s just until Hotch can come back.”
We continued on towards the interrogation room in silence. When we stepped into the mirror room, I saw that Morgan and JJ were there, watching Doyle who was sitting on his own, staring at the wall while he silently thought and worried about his missing son. With others around to “officially” watch Prentiss, I took my leave almost immediately, heading back to my office so that I could sit down, catch my breath, and have a moment to think.
When I got there, Hotch was already gone. I glanced across the bullpen to see that he was in the boardroom with Rossi, Reid, and Garcia, all of them working on finding who Declan’s mother was in case that was the next fresh lead we would get from Doyle. They were fine on their own. For just a few minutes, I could disappear, and no one would notice. So, I closed the door, spun the dial of the blinds until they were shut tight, and I immediately broke down. I didn’t even make it to my chair or the couch. I just fell right then and there.
I started crying with my face hiding in my hands. Getting Hotch back should have been the best thing in the world, and yet it felt so shitty. I hated that he lied to me again. I hated that things went down like that. I hated that our daughter was mixed up in the drama of it all now just because of her name. I hated that I was mad at the love of my life after just getting him back, because instead of wanting to hold him close until our last breath, I wanted to kick his ankles until my anger was gone—If my anger would ever go away.
And then there was a knock at the door.
I tried catching my breath and calming down enough to wipe my tears away, but the door opened before I could collect myself entirely. I hid my face in the shadows. It didn’t matter who it was because I was just hoping that they would leave me alone now that they saw me collapsed on the floor and hiding my face while sniffling.
“Sunshine?”
I let out a choked sob when I realized that it was Morgan who had followed me from the mirror room to my office. “What?”
He knelt beside me and put a hand on my shoulder. “Come here…” He turned me slightly until I was facing him, and we immediately pulled each other in for a tight hug. I hid my face in his shirt, letting my tears soak the fabric without care. “It’s okay. I promise. Just breathe.” He rubbed circles on my back to help me calm down slowly.
I started to catch my breath by sucking in deep breaths and letting out short ones. “You shouldn’t have brought up Emily earlier,” I whispered.
He had been completely out of line when he brought up my daughter in the context of Prentiss still being alive. That was a conversation for me and my husband to have at a later time when we could talk privately. It didn’t give him permission to put our predicament on blast. I was sure that everyone had been thinking it, too. I mean, they were all probably wondering the same thing I was, which was what the hell were Hotch and I going to do now that Prentiss, our daughter’s namesake was back? But, honestly, it was none of their business. Yes, Morgan was her godfather, but he wasn’t her father, therefore, he had no say. He would be the first to know when a decision would be made, of course, but not before then.
I slid out of his arms and fell back against the wall carefully to just sit there and stare into the darkness. “It wasn’t fair,” I continued.
“I’m sorry, cupcake…” He slid down the wall to sit beside me, then took one of my cold hands in his. “My emotions have just been all over the place, and I snapped when I didn’t mean to. I know that’s not an excuse, but…” He sighed.
“I get it.” I rested my head on his shoulder. Of all people, I understood the most how Morgan felt.
“Did you talk to him?”
“Kind of…”
“Are you going to forgive him?”
“Unfortunately.”
Morgan swung his arm around me so that he could hug me close to his side. “Do you ever just miss the old days when things were so simple? Gideon and Elle were still on the team. Pretty boy was still dorky and quiet all the time. Hotch actually talked to us and trusted us.”
“You mean talked to you and trusted you.” I chuckled. “He didn’t talk to me when I first joined the team.”
“Yeah, because he had a schoolgirl crush on you. That doesn’t mean he didn’t trust you. But wasn’t it so much easier back then when you two weren’t together, and everything was just about putting the back guys away, and nothing was ever personal…” He sighed. “I think we just haven’t been the same ever since the Fisher King.”
He was right. That was the first case where we were all effected personally, and an Unsub got under our skin, and tore our family apart. I mean, I assumed that it was because of Randall Garner that my sister left the BAU. If we had never gotten mixed up in that case, maybe she and Gideon would still be around. But then again, if that were the case, then we wouldn’t’ve had Rossi or Emily on the team, and I was genuinely happy that I knew them and that they were apart of our family now. Rossi was Hotch’s best friend who liked to help me pick on him from time to time. Emily was one of my closest friends… At least before she left. I didn’t know where we stood now. That was what Morgan meant by things being easier back in the day, though, I supposed. Things weren’t so messy.
“I think you’re right,” I agreed quietly.
----
Not even an hour later, Prentiss had already convinced Doyle to tell us who the mother of his child was and how to find her. With Garcia’s help, we managed to track her down. Her name was Chloe Donaghy, and she was a notorious crime lord who ran a human trafficking and prostitution ring. She and Ian had met about eight years before Prentiss met him. When she found out that she was pregnant with Declan, she tried to kill herself to make sure that his son would never be born, but Doyle stopped her before she could even swallow a single pill, and from that point on during the pregnancy, she was his prisoner. He kept her locked away. She was chained to a bed in a locked room in his heavily guarded home for seven months straight. She was given a healthy diet for herself and the baby, and Doyle had a doctor go to check on her every other week or so to make sure that they were still okay. Once she had the baby, however, she left. She wanted nothing to do with Declan, and everyone knew it, so Doyle practically paid her to stay away and to never tell a soul about their son.
However, once she found out that Ian was a wanted man and was in hiding, she decided that she wanted a piece of him in the name of revenge. So, she got the same idea as us. Knowing that Ian would try to find Declan, she waited until someone found him first—which happened to be us—and once she knew where her son was, she jumped at the opportunity to take him when everyone was too distracted to notice. We managed to track her down to Baltimore. It looked like she and Gerace took Declan down there, and they were planning on selling him to another one of Doyle’s enemies, a man by the name of Lachlan. That wasn’t good. If Declan left U.S. soil, there was nothing we could do. It would be left up to the CIA or another agency similar to them, but they wouldn’t care. So, I made a call.
I decided that the best way to ensure that little boy would never die at the hands of a revenge driven monster like Lachlan was to have our team pursue this. I knew the risks involved with the choice. I knew that my career was on the line, and that I would inevitably be demoted as unit chief—if I were lucky. Worst case scenario, actually, was that I would be fired. But I didn’t care in the moment, because all I could think about was the fact that there was a little boy out there, wondering why his own mother was doing this, and why he was facing danger again. And then I thought about Jack. Declan was around Jack’s age. I couldn’t bear to think that it could have been Jack in Declan’s shoes, and that if that were the case, I would want someone to fight for him until he was safe at home.
So, I made the call to use Doyle as a pawn in our game. We loaded Doyle into an SUV with myself, Hotch, and Morgan, and we drove down to the airstrip where Garcia found out that Chloe and Lachlan were catching a flight at together. Emily was against the idea. She was concerned that Doyle was going to get away, or that Chloe was going to get Declan and Ian, and all of this was for nothing. I wasn’t going to let that happen. We were going to get Declan back, and we weren’t going to let Doyle get away. After everything that happened over the past few months, I was going to die before fucking this up.
“You wanna do it, or should I?” Hotch asked as we pulled up to the jet on the runway. I gestured that he could go for it, but I didn’t look at him or say anything in response. I still wasn’t ready to do that much. “Okay.” Morgan handed him the megaphone, and Hotch stepped out of the car to announce, “Lachlan McDermott and Cloe Donaghy, this is the FBI. We know that you have Declan Doyle. To ensure his safety, we would like to make a trade. Declan for Ian Doyle.”
That was our cue to show off Doyle to the world, so Morgan and I pulled him out of the car, making sure that our grip on him was strong enough that he couldn’t wiggle out. And then we heard a gunshot, and I saw a muzzle flash from within the jet. No. No, no, no. There was no way they just killed Declan. I refused to believe it, but just in case it was true, we kept Doyle close to the car so that we could stuff him back in and drive off if we had to.
Suddenly, the door of the jet started falling open, revealing the steps that Lachlan started storming down with Declan in front of him. I let out a quiet sigh of relief at the same time as Ian. It didn’t last long, however, since Lachlan immediately put a gun to Declan’s head.
“Bring him here!” Lachlan demanded, referencing Doyle.
Hotch looked over at me. “You’re still the unit chief. It’s your call.”
I sighed quietly while trying to quickly weigh the pros and cons. It certainly wasn’t ideal to put Doyle in Lachlan’s line of fire, but if it was the only way to get Declan…
“Now!” Lachlan yelled.
Within an instant, I was pushing forward, Morgan following lead by helping me move Doyle towards Lachlan. When we were close enough, we pushed him onto his knees and waited for something to happen. Lachlan smiled wickedly. He said something to Doyle, but I wasn’t listening while my attention was trained on Declan and trying to figure out how I was going to grab him on time while Morgan made sure he still had a grip on Doyle. I just had to trust the process, I supposed.
When I heard movement from the jet again, I looked up to see Chloe limping out, holding her stomach from the shot she took from Lachlan. They must have argued about taking our deal. Before I could tell her to stay back, though, I saw how she was raising her arm and aiming at me, Ian, and Morgan.
“Gun!” Morgan shouted, tackling me out of the way just as the sound of a gunshot rang through the air. I groaned as the wind was knocked out of me. Morgan rolled off me to check to see if I was okay. “Are you hit?” I shook my head. “Doyle—” He spun around on his knees to see Ian falling to the ground while holding his bleeding neck. “Shit.”
Another gunshot fired, this time from Chloe aiming directly at Lachlan, instantly killing him. In retaliation, Hotch, Reid, Prentiss, and JJ all shot down Chloe as fast as they could. Hotch ran over to me, putting his hands on my cheeks while scanning my body with my eyes to make sure that I was alright, the same way Morgan had. I inhaled sharply as I caught my breath. I got that Morgan was just trying to protect me, but did he have to go all high school footballer on me? Fucking hell. We were on concrete. The tackle was worse than getting shot, in my opinion.
“Next time, I make the call. And it won’t be this one,” Hotch whispered to me, brushing my hair out of my face.
“Ha. Ha,” I responded sarcastically. “Help me up.” We clasped our hands together, and he pulled me to my feet, letting me stumble into his chest somewhat before I caught my balance. “Thanks.” I patted his chest. “This is why you’re the unit chief.”
He kissed my forehead. “You thought you were doing the right thing, and that’s all we can ever do.”
I glanced over to Doyle who was reaching out for his son, knowing that it would be the last time. A part of me broke somewhat. As much as I despised Doyle, I had nothing against his son, and seeing a bond between a father and his son, even though they hardly knew each other at all, made me think about Hotch and Jack. It could have been Hotch instead of Haley who died in our house two years ago. We were beyond lucky that he was still with us. That I had a husband who loved me, that our daughter had a father who wanted nothing more than to raise her away from all of this chaos, that our son still had a father who would protect him no matter the cost. We were, by some definition, “lucky” that it wasn’t Hotch and Jack there instead of Ian and Declan.
----
By the time we cleaned up the mess at the airstrip, I had called Jessica and asked if she could bring Emily to the office to surprise Hotch, but also because Prentiss asked if she could meet her, and I felt like after the long day we had, we all deserved that. I mean, she was our little sunshine. If anyone could change the mood entirely, it was her.
So, when we all returned to Quantico, I told security to keep an out for Jessica and the baby—to which they all cheered about how excited they were to see her. The team went up to the boardroom to start tearing down all of the photos and evidence on the walls. Another chapter of our lives closed. Over months, we had been working on a case where we only had a fraction of the pieces. The only person who knew every part of it and could end this once and for all had left, but when she came back, it ended, and now we could just breathe again. We had our family back. That had to count for something.
When I saw Jessica arrive in the elevator, I met her there so that I could still surprise Hotch and Prentiss. “Thank you for bringing her,” I said while crouching down in front of the stroller to see my lil’ bug. She was half awake, but just lucid enough to play with my finger as I wiggled it in her face. “I owe you.”
“It’s no big deal, I swear. I was on my way to pick up Jack from a playdate, and this was on the way, anyhow.”
I glanced up at Jessica. “We can pick him up later. You don’t have to race around for us like this.”
Jessica smiled lightly. “What else am I going to do?”
I stopped to think about that for a moment, considering how we were her only family around now that Haley was gone, and Roy was… Well, Roy didn’t like having Jessica around because she was a very hands-on and task-oriented person when it came to family, and he didn’t like how overbearing and protective she was of him. So, we were the only ones that were around and in need of help. I just felt bad sometimes because it felt like she was our nanny or something. Then again, every time this conversation came up where I would tell her that all of this was unnecessary and that I felt bad, she would always argue that she loved doing it and that it was no big deal. If it really were a problem, she would have ditched us months ago.
“Okay,” I gave in. “We’ll see you at home, then.” I stood to hug her.
“See you at home.”
When we parted, she leaned over to say goodbye to baby Emily quickly, then waved to me as she walked towards the elevator. I waved back shortly before pushing the stroller up the ramp towards the boardroom. Inside, I could see and hear everyone chatting and catching up with one another. Rossi was hogging Aaron. He was so happy to have his buddy back after all this time, but he was still trying to convince him to shave the beard. Over my dead body. I loved that beard, and after everything that man put me through today, I was going to get the chance to scratch my fingers through it as he fucked me. Come hell or high water, I was going to make that happen.
I pushed the stroller into the room, and everyone fell silent. This was the first time Hotch had seen her in… months, and this was the first time Prentiss was going to meet her. Everyone wanted to be witness to it, and no one wanted to disrupt the moment. So, when I spun the stroller around, everyone held their breaths. I tried to ignore them as I pulled the top of stroller back to reveal Emily.
“May I hold her?” Prentiss asked, looking up at me with the brightest smile I had ever seen from her before. I nodded. Prentiss reached into the stroller, buried her hands under Emily, and lifted her up. “Hi, there, baby girl…” Prentiss cooed. Emily kicked and fussed in Prentiss’s arms.
As I watched Prentiss rock Emily side to side in her arms, I felt someone’s arms snake around my waist and pull me backwards into a hug. I smiled and rested my head on Hotch’s shoulder. His beard scratched my cheek slightly as he pressed a gentle kiss against it. I giggled and nuzzled into his touch as much as I could while thinking about how I missed this feeling so bad every single second he was away from me. All those rules we had for so long before he left didn't matter anymore. Not when he had been away and all we wanted was to hold each other and love each other.
“She has your eyes, Hotch,” Prentiss said while Emily giggled. “Yes, you do,” Prentiss teased with the same kind of baby-talk voice everyone used around kids. “You are just too darn cute.”
“I missed you so much,” he whispered in my ear before kissing me.
“I’m still mad at you, Aaron,” I whispered only for him to hear. He loosened his grip on me somewhat. “And I don’t know when I’ll stop being mad.” He hid his face in the crook of my neck to hide his disappointment from everyone else. “But I know that I love you, and that, with time, I’ll learn to forgive and forget about this. Until then, I just need you to bear with me while I navigate rebuilding what trust I had given you, and asking you to give me answers, or asking that you give me some space—whatever it is, I just need you to understand. Can you do that?”
He nodded against me.
“We’re going to be okay,” I sighed, almost like I was trying to convince myself, too. “We’re going to be okay…”
-----
criminal minds family: @peggy1999 @gorgeousdarkangel @alex--awesome--22 @oceaneblu @brithedemonspawn @absolutemarveltrash @bshelley322 @rousethemouse @sunshinepower17 @weexinling @pettttyyyc​ @Braty-angel
27 notes · View notes
maybankiara · 4 years
Text
I’M GETTING LOST ON YOU (YOU GOT ME DOING THINGS I NEVER THOUGHT I’D DO)
summary:  Neither of them says anything for a short while. JJ doesn’t know what’s going on, because she hasn’t really said anything, but he isn’t quite opposed to the situation.
  They’ve been close before. Closer, even. But it has never felt like this – so intimate.
  If either of them moved a few inches, their lips would be touching.
  When she finally says something, it’s so quiet he almost doesn’t hear. ‘I thought you didn’t like me anymore.’
— in which jj finally grows a pair and asks kiara out.
pairing: JJ x Kiara
words: 2.2k
read on archive of our own
No Pogue-on-Pogue macking; that was the rule number one. It was a rule they’d established a long time ago, when they were just kids, and when Kiara had just joined. Crushes or love hadn’t been on any of their minds, but rules had to be set, so they set them.
  Years went by, and somehow they all became entranced by Kiara. It wasn’t just that she was the only girl in their group of boys; it wasn’t just that she was forbidden; it wasn’t just that she didn’t want any of them.
  Or at least for JJ, it wasn’t that.
  She was the first girl he started flirting with. They were barely thirteen, and JJ liked it when she blushed and rolled her eyes at him, so he kept it going. He got better at it, to the point where he was better than most boys their age.
  She was the first girl he kissed, too. It was a dare, and it was a horrible, awkward, flimsy and clumsy first kiss, but it was the one JJ had never forgotten. He got a lot better at kissing since then, but that was the only time he’d ever kissed her.
  Kiara was always there. Kiara was someone he could always like at a safe distance, where lines between flirting and being overly friendly were blurred just enough.
  But now it’s the summer before university, they are almost eighteen, and Kiara is going away sooner rather than later.
  Her father opens the door. His eyes check JJ up and down, and he frowns. ‘Kiara! It’s for you!’
  ‘Thanks, Mr. C.’
  Mike Carrera closes the door and JJ is left standing outside.
  Panic washes over him. What if Kiara doesn’t like it? What if he’s about to screw everything up? Maybe he doesn’t look good enough, maybe she still has feelings for Pope, or John B, has she ever looked at him the way he looks at her?
  He walks in the spot, fingers in his mouth and teeth tugging at the nail bed.
  This was a big mistake.
  The flowers are too much. Kiara doesn’t like people who aren’t environmentally friendly. Is picking flowers environmentally unfriendly? Is she going to get mad at him over this?
  He doesn’t think he could stand her getting mad at him.
  ‘JJ! What’s happening?’
  JJ stops in his tracks, and looks at the porch.
  Kiara is standing there, wearing grey shorts and a plum tank top. Her hair is in a half-up do, and her skin is glistening with a sheer layer of sweat that’s a consequence of one of the hottest days Outer Banks has seen this summer. She’s half leaning out of the door, with one of her hands on the door frame, the other running through her hair.
  It was enough to make him remember why he chose to take this shot.
  ‘JJ,’ she says, softly. ‘What’s wrong?’
  ‘What?’
  ‘You’re pacing like crazy. Did something happen? Are you okay? What’s with the flowers?’
  It’s only now, standing five feet away from her, that JJ notices the worry in her face. It’s in the slight wrinkles on her forehead and the pout of her lips, and how she tilts her head and looks him up and down, in a way that he knows means she’s looking for any new bruises decorating his body.
  JJ smiles. His shoulders relax and there’s a spring in his step when he comes closer.
  ‘Everything’s good,’ he says. She squints at him, so he adds, ‘I promise.’
  ‘Okay… What’s with the flowers, then?’
  ‘Oh!’ He stretches his arms towards her, the bouquet almost hitting her in the stomach. ‘These are for you.’
  ‘For me?’
  ‘Yeah,’ says JJ. ‘I got them for you.’
  Kiara takes them and he sees the familiar redness across her cheeks, one that he hasn’t managed to achieve in longer than he’d like to admit. She smiles, sheepishly, and leans against the door frame.
  Somewhere in the distance, crickets are chirping.
  She thanks him quietly.
  JJ sighs and takes a step closer, until they’re merely a feet apart. Kiara doesn’t flinch, as she’s used to him being in everyone’s personal space, but she looks a little surprised by the action.
  A strand of her hair falls in front of her face and, without thinking, JJ’s finger tucks it behind her ear.
  When his hand lingers on her neck, right below her ear, he wishes he could do more than just tuck her hair.
  But he doesn’t. He isn’t being any more reckless tonight than he already is.  Instead, he lets his hand fall back into his pocket, and gives Kiara a warm smile.
  His eyes flicker to her lips, but that’s something he can’t control. He spent one too many night wondering if they would taste the same as they did all those years ago.
  ‘I’m sorry,’ he says, ‘if this is too much.’
  ‘JJ—’
  ‘No, listen. I know there’s the rule about no Pogue-macking, and I know both you and I have been proud of how independent we are, and I respect that.’ His smile falters, but he looks at Kiara’s face, and lets himself not worry for a hot second. ‘I just wanted to tell you that I see you as more than just a friend, Kie. And obviously, I totally respect your feelings, whatever they are, because you know, bros before hos or whatever. I just wanted you to know. Before you leave. That I’m in love with you.’
  JJ closes his eyes and takes a deep breath, already preparing for the inevitable rejection.
  He wasn’t supposed to do this. He wasn’t supposed to spill his guts to her, because that could only freak her out.
  When he starts shaking, he bits his lip. In his pockets, his fingers are fidgeting, and he starts finding the crickets a comic addition to the situation.
  Well. He knew he’d fuck this up.
  ‘Open your eyes, JJ.’
  He does.
  Kiara is looking at him with a face that he can’t really read, but she’s holding the bouquet close to her chest, so maybe that’s a good sign. Her eyes are gazing into his and he begins to wonder what she’s searching for.
  A soft breeze passes between them, and her hair sways on the wind a little. She looks so soft, so beautiful in that moment – all JJ wants her is to know that. To know how beautiful she is, both on the outside and the inside.
  So he says it.
  ‘It’s not – It’s not just about the way you look, Kie, I hope you know that.’ His fingers tuck the wind-swept strands away, and this time, he lets his hand linger on her neck with no plans of moving it away. ‘You’ve been there for me when I needed you. You banter me like no one else does, and you call me out on my shit when John B and Pope can’t be bothered. I love riling you up because I love when you get protective of the world, when you talk about your ideas to save it. I just… I just wanted you to know that.’
  She smiles, laughs a little, too, and JJ feels her breath hot against his own. He’s not sure if it’s the light, or if there’s a twinkle in her eyes.
  ‘I never took you for such a sap, JJ.’
  He leans his head back and groans, but when he looks back, she’s smiling even wider. His eyes fall to the strap of her top and he plays with it, twirling it around his fingers, trying to keep his mind away from thinking about how easy it would be just to take it off.
  Neither of them says anything for a short while. JJ doesn’t know what’s going on, because she hasn’t really said anything, but he isn’t quite opposed to the situation.
  They’ve been close before. Closer, even. But it has never felt like this – so intimate.
  If either of them moved a few inches, their lips would be touching.
  When she finally says something, it’s so quiet he almost doesn’t hear. ‘I thought you didn’t like me anymore.’
  JJ looks back at her, brows furrowed, head tilted. ‘What?’
  ‘When John B kissed me, and then everything with Pope happened.’ Kiara bites her lip and looks away, and JJ is now certain there’s a twinkle in her eyes. ‘I thought you moved on, and at first I was happy, but then I realised how much I miss you flirting with me and dancing with me and complimenting me and – and –’
   ‘I’m here, now,’ he says. ‘I never stopped liking you. I tried, but I couldn’t.’
  She looks back at him, hopeful, and JJ can barely hold himself from kissing her full on the mouth, like he’s been dreaming of all these years. The realisation of the meaning of her words hits him like a truck and he steps back, almost, gasping.
  Kiara’s hand catches his, the one that’s still on her neck. Her palm is warm and her fingers are tiny compared to his, and seeing them together like that is yet another blow.
  ‘JJ, are you okay?’
  He clenches his teeth, and nods. If he opens his mouth, he’ll kiss her, and he doesn’t know if she wants it.
  That’s when he realised that he didn’t actually believe this would go well, and it did, and he isn’t fucking prepared for this.
  What do you do when you confess your feelings to someone and they love you back?
  He laughs, a little, nervously. ‘I was actually supposed to just ask you out. Not confess my feelings.’
  ‘That’s okay. It’s not too late.’
  ‘For what?’
  ‘Asking me out,’ Kiara says, sheepishly, ‘if that’s what you want.’
  ‘Oh.’
  For a brief moment, JJ’s brain just… freezes. He stares at Kie and there’s a lot of things he’s thinking of doing, but formulating a question that he never even gave a thought about (because he never even thought he’d get the chance to) is not one of them.
  ‘Um.’
  Kiara laughs. ‘JJ, are you broken?’
  ‘Maybe,’ he says. ‘A little bit.’
  ‘Okay.’
  Taking his hand in hers, she takes it off her neck as she crouches and places the bouquet on the floor. She straightens her back and closes the distance between them and oh no, JJ is definitely thinking all the wrong thoughts and wraps her arms around his neck, the way she never has.
  The way he hasn’t even seen her do to Pope, for those brief three weeks they tried being a thing two years ago.
  No, this is a different, this is for JJ only, it’s intimate and possessive at the same time, and her lips are so close that—
  Kiara brushes her lips against his, so softly he barely feels it.
  ‘Kie,’ he whispers.
  Her voice is just as quiet. ‘Yes?’
  JJ’s hands wrap around her torso and press her against him. He leaves one hand at the small of her back, drawing circles with his thumb, and brings the other one to her ear, to get her hair out of the way for the third time in fifteen minutes. ‘You have hair all over your face.’
  She laughs at that and buries her head in his chest. It’s so soft and natural and earnest that he almost melts into the touch, seen and felt and understood and accepted for the first time.
  He places a finger under her chin and lifts it, gazing into her eyes with a little more adoration he think he should – but he can’t help it. Not anymore. Not when his world is in the palm of his hand.
  So he says, quietly, ‘I love you.’
  And then kisses her, softly. It’s a chaste kiss, and he sets rain to the flames within him, tells them to wait.
  When they part, JJ sees the world in a new light.
  Brighter, for starters, and maybe not out to get him.
  ‘Kiara Carrera,’ he begins, ‘do you want to go on a date with me?’
  ‘I would love to.’
  Kiara leans in for a kiss but he pulls back, letting go of her instead. When she looks at him, puzzled, he walks off the porch backwards, grinning at her wider than he thinks he’s ever grinned.
  ‘That’s all you’re getting for now,’ he tells her. ‘I’m a reserved man.’
  She flips him off and she laughs, and something makes him think that maybe this wasn’t a dumb idea. Maybe they were just the same, only better. Maybe he didn’t fuck everything up.
  Maybe for the first time in his life, he did the right thing.
134 notes · View notes
danetobelieve · 4 years
Text
Catchlight || Adam and Winston
TIMING: sometime before the S1 finale LOCATION: Non-descript forest in White Crest PARTIES: @walker-journal​ & @danetobelieve​ SUMMARY: Adam and Winston run into some ethereal creatures in the forest.  WARNINGS: violence cw, blood cw a little, no TWs
It took alot to call Adam Walker and his cadre hunting buddies back into town from hunting Alghouls, Redcaps, and other scourges out in the hinterlands, but a black ocean and people sprouting with eyeballs all over their bodies definitely qualified. One of the sacred duties of Hunters was to ensure that the supernatural world’s secrecy was maintained. The status quo between humanity and the paranormal was a fragile balance, one Adam often had to enforce using copious amounts of Nepenthe-filled syringes. 
But sometimes the balance just absolutely shat itself and imploded. Containment was a lost cause and it just came down to finding some way to make the madness stop. 
Adam had inquired with older Hunters and hit up various contacts in the grimier side in White Crest’s paranormal world. Bribes, promises, threats, and tacit agreements had thus far yielded very little to go on. Antedilvuian myths and mummerings about the Troubles come again were interesting, but not exactly helpful in the “how does it fucking stop” deparment. 
So it was that Adam had followed this trail of breadcrumbs out in the Outskirts where witnesses reported a beacon of unearthly light that was only visible when your eyes were closed. The shadows of late evening were deepening in the forest as Adam made his way through towards the beacon, M4 carbine slung over his shoulder and pausing at various points to close his eyes and reorient on the pillar of light. 
-
The eyeball was still in Winston’s hand and they were still very far from thrilled about that fact. It was the visions, they had gone from just monsters to everything else and Winston had found themselves drawn to the Outskirts of the town. The plan wasn’t a great one. They’d pulled on a jacket and shoes and grabbed their phone before heading out into the night. They just knew that they had to be there and as they drove they found themselves following a pillar of light that they could see with their third eye but nothing else. Winston parked a little way of and began their trek towards the light once there was no more road to follow. They were making their way there when they spotted a figure with what looked like an assault rifle on his shoulder. Doing their best to act natural, Winston held their hand behind their back so that Adam wouldn’t spot the eye and gave them a weak smile. “Nice night for a … walk … right?” 
-
Adam had been raised with a singular raison d'etre for his life, the protection of humanity. Thus when he crossed path with a human being dangerously close to the epicenter of some paranormal phenomena, the Hunter’s features creased in concern. His chestnut eyes scanned the horizon back from whence they’d both come, taking in the waning crepuscular rays of evening as they dimmed to frail strands peeking through the forest canopy. 
It’d be a setback to have to turn back so close to the objective, but Adam couldn’t in good conscience let a civilian draw so close to supernatural danger. If that meant this night was a bust and he drove this person home, so be it. “Afraid that's a no go my dude,” Adam said. “It's pretty dangerous out here at the moment, there’s been an accident up ahead,” said the young man whose kevlar tactical jacket and fatigues were an odd uniform for a ‘civil servant.’ “You should head back into town” 
-
Raising an eyebrow gently, Winston flashed Adam a quick smile and looked them over. “Listen, you’re clearly not WCPD because I know literally everyone that works there from the janitor to the captain,” working there helped with that, “I don’t see a badge or an identification being displayed so it’s a pretty good chance that you’re not any other sort of agency.” Winston took a deep breath, doing their best to pretend that they were authoritative in this situation and not completely terrified of the assault rifle that could riddle them with bloody holes. “I need to get over to the ‘accident’ up ahead, because there’s important information that I really need. I don’t want to argue or whatever and you’re more then welcome to come, but I’m … I’m going okay?” Winston tried to sound tough, they doubted they did. 
-
Adam stood on the rugged forest path, sizing his interlocutor up for a time. It was clear they were determined to go through with this and while the Hunter’s mutant strength made manhandling them back into town possible, it’d likely be a pointless effort that was more trouble than it was worth. Guess that meant accompanying this person into the anomaly and trying to make sure they made it back out intact. 
Definitely wasn’t Adam’s favorite arrangement, but with ancient demonic forces on the rise there wasn’t really the luxury of time. 
He nodded assent and offered a hand to the stranger. “Name’s Adam, and no, I’m not anyone with any cred,” he admitted with a chuckle. The young man started the trek towards the invisible luminosity, evidently having accepted that company just going to be a thing. “Who’re you?” 
-
“Winston, Dane.” Winston replied, “Cool to meet you I guess, Adam right?” Winston looked them up and down. Muscular. Armed too. Which was good if this guy was going to join Winston in their little trek which by the looks of things they seemed determined to do. They looked at their path and began down it, slowly and carefully. The strange light that you could only see when you closed your eyes might have been helpful, but it was still completely dark otherwise and Winston did not need to twist their ankle at a time like this. “I’m out here doing research,” Winston explained quickly, “obviously you’ve noticed all of the weird stuff that is going on around town. You know, you’d have to be living under a rock to miss it. But then again some people aren’t the most observant, I guess that because you’re here you’ve come for a similar reason?” 
-
Adam nodded acknowledgement, not seeing purpose for further obfuscation when it was clear Winston knew what was going on. He had no idea what the unseen light truly entailed, but keeping Winston in the dark on details seemed counterproductive if they were going to help each other. “I’m trying to find the connection between the increases in paranormal phenomena and a disease some of the locals have come down with that causes eye-looking tumors on the skin.” The Hunter noticed his non-mutant companion was struggling in the dimming twilight. Adam reached back into his pack and rummaging around before pulling out a rescue flashlight from amongst the more deadly equipment therein. Offering it to Winston, Adam forged ahead over roots and gullies that dipped into dense brush. 
“I’ve been tracking leads to see if there is some way to cure them and purify the water in the harbor. Since witnesses report that the light thing in these woods flares up whenever the sun goes dark, I want to check it out and compare notes with friends that are looking into other stuff around town.” 
-
Accepting the flashlight, Winston flashed Adam a grateful smile. “Oh don’t worry,” Winston replied with a laugh, “I am not convinced that they are just eye looking tumors and not just eyes.” Winston had an eye on their hand and although the way that they had received it had been very different, Winston however wasn’t about to admit to having  a weird-eye-hand. “Friends that are looking into other stuff around town? Have you started a supernatural-super-friends group?” Winston wanted to join honestly. If there were people trying to fix this so that Winston didn’t have to fly by the seat of their pants all of the time then that would truly be ideal. “I suspect that they’re all tied into the giant squid demon that is currently sat at the bottom of the lake, or the cult that appears to worship it … either way, investigate the strange light that flares up when the sun goes down?” Winston smirked, “I can help with that for sure.”
-
Adam turned back to give Wintson a long look, crooking a sandy eyebrow as his companion revealed that he knew more then the Hunter ever anticipated. “Born into it technically,” Adam admitted. “But it’s widened over these past few months as more and more people get sucked into the paranormal craziness. I’d prefer more secrecy about the demonic incursion...alot more secrecy,” the Hunter continued with a sigh that suggested perhaps the rather swashbuckling approach Blanche and others in the Scooby Gang took with supernatural secrecy did not sit well with Adam sometimes. “But supernatural awareness is always higher in an area with thin dimensional barriers like this. Doesn’t help that we may be having a repeat of Dead Sunday, in which case we’re totally fucked together regardless. So, y’know YOLO.”  
On that cheery note Adam led their party of two to a wooded hill covered in gnarled trees whose branches curled like grasping claws around large quartz boulders. Closing his eyes, Adam reoriented on the Unseen Light’s last vantage point. “Happy to have you along Winston...honestly any help is appreciated when the stakes are this high,” he said with a warm smile before turning back to blindly regard the small valley their vantage point overlooked. “Not to pry, but got any skills I should know about for the investigation? Also, are you comfortable with a weapon? I’ve got some sidearms if you are.” The hunter nodded to the incline that descended into a far densely wooded section. “I’m asking now because once we go down and close in on the anomaly. Weird shit will probably start to happen.” 
“It’s my duty to protect you,” the Hunter said without bothering to explain why. “But it’ll be easier if I know what you’re capable of and all that” 
-
“Born into it?” Winston replied with a raised eyebrow. They weren’t sure if Adam was implying what they thought he was, but that sounded like he definitely wasn’t entirely human. Which made Winston wonder what he could be. He didn’t seem like fae and he looked too tanned to be a vampire, but that didn’t really help Winston with anything. There was still so much more to learn. “I think that everyone in town would prefer more secrecy, mostly because those that are ignorant of the existence of the supernatural aren’t exactly the type to take it well if it were forcefully revealed to them.” Winston wasn’t worried about word about the supernatural getting out, everyone was clearly so oblivious to it that even if there was a giant demon swinging a flag saying that the supernatural was real people would still think that it was just cosplay. “Yeah, defo, yolo is really the response I want to something called … Dead sunday?” Winston wasn’t sure that they wanted to know. 
Winston stuck their hands in their pocket as they walked. They’d brought a few prototypes that they had been working on. Magical grenades that did different things, the spider bot and a few other very early prototypes. “I don’t want a gun,” Winston replied immediately, “I don’t … no guns. I’ve got …” Winston wasn’t sure that they wanted to really explain everything to Adam, although the longer this went on the more sure that Winston was that Adam was a Hunter. “I’ve got my own stuff that I can use to look out for us and I’ve got … other things that I can do if the worst comes to the worst.”
-
“My family have been paranormal investigators for a while,” Adam replied, technically telling the truth in a manner of speaking. “Dead Sunday was a colonial shitstorm alot like this,” was Adam’s attempt to summarize a rather complicated event whose aftermath had established both the Silver Bullet and the Hunter presence in White Crest. “Humanity lost...hard.” 
Winston’s assessment of his combat capacity as literally “stuff” and “things,” was hardly reassuring from a tactical point of view. 
“Alright man, watch your step,” Adam said as he began descending down into the wooded valley of Unseen Light, feet displacing rocks and soil as he eased his way along the trailess side of the incline. 
-
Raising an eyebrow thoughtfully in response to Adam’s reply, Winston wished that they could make the same claim. “Well, I’ll let you take the lead since you’re so much more experienced apparently,” Winston was being sincere. They were still really new to this and they were always happy to follow someone else’s lead if it meant that it kept them alive. Winston followed after them closely, their third eye blinking in the light that only it could see. In fact, right now Winston was wondering if the third eye was entirely accurate, because they were pretty sure that there was a squid like creature who’s tentacles were going to smack into them. 
Winston flew backwards with a thud, their back cracking with a painful crunch as their shoulder collided with a root, but when they looked around there was nothing there. “What the hell?” 
-
“Sorry...look, I didn’t mean it like that, you asked…,” Adam began, misinterpreting Winston’s statement as salty sarcasm. However that particular line of miscommunication born of tension was rendered moot as Winston was abruptly knocked back by some unseen force, hitting one of the many thick roots that formed vein-like patterns across the forest floor with an audible crack that made Adam instinctively wince. In the instantaneous reflex born of training, Adam drew a Sig Sauer service pistol from its holster and closed the distance to Winston, scanning for any movement in the darkness while he offered a free-hand to help Winston up. 
Yet the forest was absolutely quiet. The Hunter’s enhanced senses couldn’t even hear any ambient animals noises in the brush. The local fauna knew something was wrong. 
There came a feeling of frigid heat down Adam’s spine, and eerie hot-cold feeling as his Hunter clairvoyance picked up the approach of a paranormal creature. The sensation grew to painful intensity throughout Adam’s body as the inhuman presence grew nearer and nearer. 
But there was still nothing, not even the crack of a twig. 
“I can sense something here,” Adam mummurred to his companion, not really caring about revealing his abilities with a threat so close. “It might be glamored, just a sec.” The Hunter retrieved a grey metallic sphere emblazoned with a severed celtic knot from his pack. There came a mechanical click as Adam flicked a trigger on the Dispellate and wound up into a baseball pitch that hurled the grenade towards where the unnatural presence emanated strongest. 
A low hiss of released vacuum sealant was followed by a faintly luminous silver smoke that billowed through the air. Tendrils of the Dispellate’s alchemical cocktail of anti-illusion vapors spread amongst the branches like an argent fog. But though Adam’s aim had hit the area of that strange presence dead-on, nothing materialized. 
“Well uh..shit,” Adam huffed with a frustrated but rueful chuckle. “You ok? I’m not sure wha...gah motherfu.” The Hunter’s rumination was out in an outburst of pained cursing as deep gashes cut straight through his tactical kevlar from thin air, the impact knocking Adam to his knees as blood blossomed down his side. 
-
Winston was really getting sick and tired of fighting things that they could so clearly tell were well out of their league. It wasn’t that they weren’t capable. But this thing could either materialise and disappear from thin air or it was stopping them from seeing it somehow. Wracking their brain, Winston tried to come up with a solution to their current predicament but with less luck then they would like. It occurred to them that having a third eye should’ve been helpful in this situation but they were more preoccupied by the blood. They remonstrated with themselves for not having learned healing magic more thoroughly before standing to their feet. “I think - what if this thing is like the light and we can only see it with our … other eyes.” Winston pulled the bandage from their hand and closed their eyes, a blurry and distorted vision of the dark clearing came into view and Winston was distressed to see a truly eldritch horror in place. “Close your eyes,” Winston said, summoning magic, this was going to be a fight and they really weren’t sure that they would win. 
-
Doing as he was bidden, Adam closed his eyes. 
The Medusozoa creature was luminous and might’ve even been called angelic if the sunray-like spray of its tendrils were not bristling with stingers and barbs. It appeared as if several immense Lion’s Mane jellyfish had merged with one another, drifting on currents of light rather than water. The central mass of the numinous demon was a pulsating and translucent mesoglea within which shone orbs that seemed to be made of stella plasma or solidified light. They churned and writhed within the creature as if they were some kind of organs, an alien anatomy composed of celestial phenomena rather than flesh and blood. Its tendrils swayed and undulated like a thick field of white-golden vines that snaked around trees and stones. Each of tentacles seemed graceful and soft as gossamer, yet the surgical wounds bleeding down Adam’s side proved the thorn-like nematocysts bristling along those tendrils could slice through flesh and kevlar effortlessly. 
Adam opened his eyes again experimentally. Just as Winston had surmised, the grove in front of Adam was completely empty and dark, devoid of the immense predator and the surreal sourceless light.
“Good call…” But when Adam glanced to Winston, they seemed to be in the midst of concentration. A real combat situation only gave seconds to think, and it was clear that those tendrils were snaking back to take another lunge. Adam had no clue what Winston was doing, but it didn’t matter because they’d be impaled standing still like that. 
In the split second afforded to him, Adam decided he needed to draw the eldritch thing’s attention and either give Winston a chance to retreat or hopefully prove that the “stuff and things” they’d mentioned earlier wasn’t just empty bluster. 
With the alacrity of oft-repeated reflex, the Hunter closed his eyes and unshouldered the rifle. Sprinting away from Winston along the grove’s edge, Adam sent cracking bursts of gunfire towards the nearest shining tendrils. The whip-like recoil signaled that, even though it was unclear if any real damage had been done, Adam’d at least gotten the thing’s attention. Trying to fight back the growing pain in his side, Adam continued to engage in a guerrilla battle of firing short barrages before running like hell as lancing tendrils eviscerated the spot where he’d just been seconds ago. 
-
It was a giant jellyfish demon. Fuck. Winston had already been incredibly disappointed with the fact that their main enemy was a giant kraken thing or demon or whatever, calling it Squidward made it a little less terrifying to deal with but that didn’t mean that Winston liked this anymore. It’s luminous body seemed to radiate light when Winston looked at it through the eye in their hand and Winston kept their normal eyes closed. 
Adam seemed to be back up and moving and with the amount of blood that he had lost Winston was amazed that they were still going. But they weren’t about to let anything happen to their new friend if they could help it and as the creature shot tentacles in every direction, Winston summoned as much power as they possibly could and focussed on providing some kind of protective barrier. Between themselves and Adam and this thing. 
For a second, Winston held their breath, hoping and praying that this wasn’t one spell that they would inevitably mess up. 
The tentacles surged forward, invisible light flooding the clearing. Winston could see the luminous barbs and thorns pulsing towards them, getting closer with every second. Then they stopped, as if an invisible force was holding them in place. “Fucking shoot the thing,” Winston grunted as the tentacles drew back and slammed against Winston’s barriers. 
“Oh fuck,” Winston grunted with exertion as sweat began to pour down their face and a small trickle of blood pooled at the corner of their left nostril, “this things fucking strong.” 
-
In retrospect, as he looked at the myriad lanced tendrils poised above him many Swords of Damocles, Adam would note that Winston had just saved his life. Adam was superhuman yes, but not immortal. The bloodloss had been progressively slowed his reactions times down while the luminous demon’s unceasing barrage of piercing tendrils had been tearing apart the trees and stones just behind him. But sprinting and firing the rifle only made his wound bleed faster, There’d come a point where the numbness spreading up the young soldier’s side had caught up with him. Spines and shining stingers came down in a rain of blades from all directions. 
But just before Adam was torn to shreds by all these living plasma cutters, they all just...stopped. 
Adam’s sight darted from the horde of bright death stopped in midair, before opening his brown eyes to look over to Winston, who strained against some unseen burden as a rivulet of blood slid down their face. 
Was Winston a Spellcaster? Were they a Medium with crazy Jean Grey psychic powers? Adam really had no idea, and didn’t have time to spare the matter any thought. 
Adam used the precious seconds to retrieve one of the grenades he hadn’t had time to arm while running. He didn’t carry ordinances like these into town as the risk of civilian collateral was too high. But coming out here Adam hadn’t known what to expect, and like many Hunters, Adam reacted to unknown variables by scaling up the firepower. Unlike the harmless Dispellates he also carried, this particular explosive much resembled the standard issue M67 fragmentation grenade, though the United States Military would've looked askance at the bizarre materials and esoterica that comprised its payload. 
The Hunter wound up into a pitcher's throw that sent the grenade sailing towards the demon’s central bulbous mass. Cold iron shrapnel blossomed from an explosion silver fire, the gleaming metal glittering starkly against the creature’s bioluminescence. Adam moved with the alacrity of training and pretanatural reflexes, sending spheres of jagged death sailing towards the creature in a steady rhythm until he hadn’t any more left. For a few seconds the far side of the grove was a confusion of shredded gelatinous flesh, smoke, and a crisscrossing hail of sagittal cold iron. 
The following silence lasted until the tangled, hemorrhaging masses of luminous gelatin rose from where they’d lain lifeless on the ground like dandelion seeds taking to the air, each new being bulging as glimmering new celestial sphere-organs underwent cellular mitosis in their center. 
“Hey...uh, Winston, think we need to get the fuck outta Dodge.” 
-
Fuck. This thing was strong. Winston could feel the blood slowly trickling further and further down their face. It didn’t take long for it to reach the edge of their chin and drip slowly onto the leaves below their feet. The exertion that it required for Winston to hold this thing in place was staggering and they knew they wouldn’t be able to do it for long. 
Even now, as they struggled against its immense strength, Winston could feel their own failing. 
They were way more then out gunned here. It felt like they had brought a knife to a gun fight and Winston knew that at such a violent disadvantage there was very little they could do but eventually lose this fight. 
Fortunately Adam had brought grenades to the gun fight, and as the explosions rocked the ethereal creature that they’d encountered, Winston couldn’t help but recoil as they let their magic fade away and staggered slightly. 
Reaching up and wiping the blood from their chin, Winston blinked a few times somewhat obliviously. They could see Adam saying words, but it was taking them a little too long to translate those words from sounds to semantic meaning. 
“I agree!” they stammered as their feet finally remembered what they were meant to do and began backing away from this, they would have to find their answers somewhere else, “Let’s run. Running seems like a really fucking good idea.” 
-
Adam wasn’t capable of much magic beyond the alchemical processes for monster venom antidotes and warding sigils to hold off your average Ghostbusters bullshit, a combination of sketchy science and religious ritual that kind of went hand in hand with the Hunter gig. However he did know that sorcery on any scale always came with a price. 
Now, Adam wasn’t sure what the cosmic exchange rate for “Saving a dumbass from a giant invisible demon” currently was. But looking at the state Winston was in right now, Adam guessed he owed them way more than some beers and a “thanks bruh.”
He offered Winston a shoulder to throw their arm over to aid in walking if they cared to. Either way, Adam tried to forge ahead up the wooded hill they came down, trying to leave the valley of unseen light and its rejuvenating bloom of newborn Medusozoa demons behind. Unfortunately he had to keep his eyes open to actually see the precarious trail and its maze of stones and roots, during which time he couldn’t see what the invisible entities were up to. Between mounting injuries and having to close his eyes occasionally to keep track of the enemy, progress up the hilltop going at the agonizing pace of a dream, where you want to run, but are stuck in a slippery slow-motion. 
Part of surviving as a Hunter is an instinct for when the monsters have been a bit too quiet and heinous shenanigans were incoming. Sure enough, the distinct lack of being impaled during the last few minutes made Adam suspicious that something was up. Paranoia was rewarded with catching sight of one of the newborn shining creatures beginning to pulsate unsettlingly as its biolomunience built to a feverish incandescence. The Hunter’s mutant senses felt the energy discharge begin before he saw it. 
“Hit the deck!” 
Adam threw himself, and hopefully Winston too, flat against the dirt of the rabbit trail. A beam of pure concentrated photons lanced overhead, boring through nearby boulders like a knife through butter and scissoring through some trees ahead. The wood burst into eerie multicolored flames that burned bright but gave off no heat. 
As a test, Adam opened his eyes and had his suspicions confirmed. The prismatic flames were completely invisible to the naked eye. Branches and roots were blackening and charring without any visible source of conflagration. “Shit shit, c’mon let's go this way, can’t let the fire cut us off,” Adam said to Winston, huffed with exertion as he tried to find an alternate route through otherworldly flames and demonic photon beams. 
-
Hands still covered in their own blood from wiping it off of their face, Winston allowed Adam to lead them away from the demon that was fucking with them. They were going to start making signs for themselves. Things that said things like “Don’t go into that dark hallway on your own,” “don’t walk down that dark alleyway without your friends,” and “don’t fuck with giant floating jellyfish demons you have no chance of beating in a fair fight.” 
They were barely away from the original scene of battle when Winston spotted the beam incoming. Fortunately Adam had spotted it even before they had. “Fuck, ow.” Winston grunted as they hit the dirt and rolled away from the laser beam. 
“Thanks for that,” Winston said as they spat out blood and gravel, scrambling to their feet they followed Adam through the forest. 
Reaching into their bag, Winston pulled out a lightbulb that they had heavily edited and enchanted. Winding it back, they hurled it at the Medusozoa Demon and watched the light bulb sail through the air before smashing against the jellyfishes’ cap. 
A second later there was an electrical discharge as the micro-emp grenade that they had been testing exploded with a bright flash of electrical energy. 
“I don’t know if that helped,” Winston replied honestly as they turned on their heel and raced after Adam, blood still pouring -- though more slowly -- down their face. “But we should definitely run faster and not stay to find out.”
-
Adam didn’t need to be twice and tried to hussle it over the hill. There were tense moments as beams streaked after them, sending that strange multicolored flames cascading through the trees in brief conflagrations the naked eye couldn’t perceive. However whatever crazy discharge ordinance Winston had let loose seemed to have wounded or at the very least disoriented their closest pursuer, and their very painful trek amongst jagged stones, thick brush, and eldritch energies finally crested over the hill. 
Eventually Adam signalled for a stop after they’d put some distance between themselves and the valley where that strange light that only flared when the sun's eyes closed had first been sighted. “Just a sec Winston...wanna check something.” 
Trying to push the pain in his side away, the Hunter closed his eyes and leaned a nearby tree. Adam let his Hunter sense drift, clearing his mind of distracting thoughts as best he could. After a minute a dim impression surfaced.  Adam waited, but those presences in his Hunter senses didn’t draw any more intense or draw near the hill they’d just cleared. Eventually the clairvoyant feeling of the demons faded as they drifted beyond his range. 
Whether the valley was their ‘territory’ or those creatures maybe couldn’t even exist outside that unseen light, Adam couldn’t really say. But those were questions for later. 
The Hunter opened his eyes. “They aren’t following us….hey uh, thanks for having my back there Winston. You saved my ass,” was Adam’s awkward foray into gratitude and he continued along the last part of the trail towards their cars. “Owe you big time.” 
13 notes · View notes
morallygreyprompts · 5 years
Note
Ok so I'm a total sucker for villains getting whumped- can I get Villain getting captured by the 'good guys' and the last thing hero says to them essentially boils down to 'I hate you'. Except that said 'good guys' turn out to be terrible people, Hero and their friends find out they're hurting Villain and storm over there to rescue them but Villain's so messed up that they only recognize Hero?
Ask 94
SAME. I should do it more often, to be honest ^^’ This is a mega oof request and I cannot wait to get this written XD And I wrote A LOT. Like 2K (Mostly in lesson, might I add) so enjoy. It is quite the whump.
The heroes wrestled Villain down onto their knees and tied their hands behind their back whilst Hero loomed over them. “So you’ve won,” Villain grumbled, trying without success to snatch their hand back. “Now what? Prison, house arrest, cage in your cellar?”Hero shrugged. “I don’t know. The Government wants you, said they had a new secure facility for you and any other… villains, like you.”
“Sounds boring. But a villain? That’s what you think of me?” Villain smirked. Hero leaned in close to their face, Villain could only see some odd mix of anger and relief. “I think so much worse of you.”Villain turned their head to see a van pulling in, some grim and stony looking agents inside.
“Looks like my taxi is here,” Villain muttered.
Villain was lifted to their feet and dragged off in the direction of the van by the other heroes.“Hey, [Hero],” Villain called. “All this- this was fun. I hope we can do it again some time.”“Hold it,” Hero said. Their friends stopped, waiting for Hero to catch up. Villain tried to smile away the concern that rose in their chest. Hero was inches away from their face, but Villain wouldn’t break eye contact, wouldn’t wipe that smirk from their lips.
“Let’s make something very clear,” Hero resumed. “I hate you with every speck of my being. I never want to see you again and I don’t care what happens to you. You’re going to rot in there and it’s the least you deserve.”Villain’s face fell slightly. “You don’t mean that. We’ve hardly got ourselves a clear cut relationship here, but-”“Relationship? There is nothing between us, not even an acquaintanceship. You’re a thorn in my side I’m glad to get rid of. I mean every word- get them out of my sight.”
Villain fought against the heroes, shouting “You don’t mean that! I know you don’t! If you hated me you’d have killed me! Hey, where are these guys taking me? [Hero]!” Hero paid them no attention. They’d turned their back on Villain and waited until they were bundled into the van, restrained adequately and whisked away.
Hero stood in the still air, taking deep breaths, trying to block out Villain’s voice that still rang in their head. One of the other heroes came over and put a hand on their shoulder. “You good?”
Hero turned to face them and nodded. “Yeah, just- It. It’s finally done. They’re gone somewhere they won’t hurt anyone but… they weren’t that dangerous compared to some- so why is the Government taking them there of all people?”
Other Hero shrugged. “They’ll be looked after and the city is safe. I know you’re very good at it, but, don’t think too far into this.”
Hero nodded and watched the van fully disappear out of sight.
_____
“They don’t mean that,” Villain mumbled, their thin arms swaying in the shackles above their head. “They don’t mean that. They’ll be sorry, they’ll help. They’ll help, they care. They have to care…” They were so busy mumbling to themselves, trying to find a way to fix the hurt Hero had caused them that they hardly noticed the two guards coming into their cramped cell, constantly illuminated with white light which glowed on the white tiles they were sat on.
It was only when their hands were freed from the shackles that Villain said, “What’ll it be today? Hmmm? The chair? The table? The bath?”
Villain groaned as one of the guards grabbed them by the wrist and exposed a vein for the other to catch with a needle. “Hm, something new then?” they guessed before their eyes slipped shut.
New and underwhelming considering some of what they’d been through. They were hanging from the ceiling by their wrists, but the group of half a dozen guards around them was torture enough. It was quite clear what they had in mind, it had happened once before, but not with this many people. The leader didn’t hesitate in starting, they punched them until their knuckles bled. Villain could hardly stop their vision swimming by the time the next one came at them.
They felt bones break, skin tear and blood pour from their nose, lips, eyebrow and after that Villain was too out of it to know what was going on, but they didn’t stop. Not until all six of them were satisfied and Villain was left hanging, unable to keep their feet at all. Unconsciousness kept claiming them for a little while, like waves on a shore. The entire time they could only think of one person, Hero. That was the only person they could remember anymore.
“Bring them in, it’s time they got in on this,” one of the guards said. Villain was amazed to see one of their Henchmen walk in, one of the younger ones. Villain had considered them family. They didn’t let their surprise show, but that face filled them with hope.“Come on then. Show us what you’re made of,” one guard said.“If you hit them any more chances are they’ll die,” Henchman pointed out, acting unphased. Or were they unphased? Did they care? Hero didn’t, why should anyone else?
“Then aim for the gut, now come on!”
Villain didn’t look at them as they punched them in the gut, in the ribs, and the rare strike to the face. Villain was past the point of making noises, past the point of trying to raise their legs to protect themselves. All they could do was lie there limp and accept the pain at the hands of someone they once thought was a friend. It wasn’t like they had a choice, but the concept hurt them more than any blow, but not as much as Hero abandoning them.
A particularly nasty strike to the stomach winded them, and Henchman stopped, panting.“Burnt… out. A.already?” Villain tried to taunt. They couldn’t see their reaction for their blurring vision, but Villain was sure they saw surprise.“It’s a start,“ one of the guards grunted. "Drag them back to their cell. We’re done here.”
Villain fell into Henchman’s arms as they were released. They held them with as much care as they dared to show in front of the others.
Once they were back in their cell, Henchman waited at the door until they were alone. They took out their phone and Villain couldn’t stop them from taking pictures of them.
“Trophies?” Villain wheezed, curling up on themselves. Their ribs throbbed and their breathing was dangerously shallow. Their body shuddered with exertion. Henchman didn’t answer, but the pitiful look on their face was enough for them to guess that wasn’t what they were for.“Rest up. They’ll probably be back for more soon.”
And watched the door shut, just like they’d watched Hero turn their back on them. They were alone again, and that hurt more than anything.
“They care,” Villain started murmuring to themselves again. Their white lie, the false promise that kept them clinging onto hope. It was something they constantly chanted to themselves now. “[Hero] will come… They care… they, they have to care.”
________
Hero slammed Henchman into the wall of the store they’d ben trying to rob. “You picked the wrong store,” Hero grumbled. Henchman didn’t retaliate. “I need your help. That’s why I’m here, doing this.”“Yeah, right.” Hero grabbed them by the collar and marched them out of the store and into an alleyway.
Henchman used their arm like a windmill to free themselves and they stepped back. “Before you start anything. Look. Just look, I need to show you this. After that, you can do what you want.”
Henchman carefully took their phone out and shoved it toward their face. Hero took a moment to focus on what the image was.“[Villain]… What?”
“That ‘Government facility’? It’s a lie. It’s just a group of people who want to make [Villain] suffer for everything they’ve done. Everything they haven’t done, rather. You guys really don’t understand how [Villian] is. You’ve just pinned every crime on them that you can think of and they haven’t done half of it!”
“Is that so?” Hero raised an eyebrow.“[Villain] doesn’t kill,” Henchman snapped. “And they are getting tortured out there. You have to help them, please! You’re their only chance.”
Hero frowned. “Show me that picture again, I need to look again.”
Henchman did so, this time showing them more than one, showing Villain curling up, hugging themselves, the blood dripping onto the floor. Deep scars and sores showed on their wrists and arms. Their face was completely masked in bruises, and they were so thin. So weak…
“It’s been months since they went there… How did you get this?” Hero asked.“I infiltrated the place. I needed to know [Villain] was okay… They’re all I’ve got.” Hero could hear the upset in their voice, the dampness forming in their eyes.“Is this a recent photo?” Hero continued.
“I took this picture yesterday and it really doesn’t show how badly off they are. The first time I saw them I. I had to beat them. They need you, [Hero]. They murmur your name all the time, they’re clinging onto the hope that you care, that you’ll save them. That place, it’s driving them mad. I’ll do anything you say if it means you’ll get them out of there. Hell, I’ll beg on my knees! You’re their only chance.”
Hero’s face looked blank. So many emotions hit them at once that none of them seemed to register in their mind. Anger, confusion, guilt, upset. They couldn’t process it. “You’re coming with me. I’m getting the team and I’m getting them out of there.”
Henchman’s face lit up with hope. “Really?!” They bowed their head low in genuine gratitude. “Thank you, thank you so much! You’ve no idea how much this means to me- and to them.”
“Just get me in there, and if you’re lying then I’ll have you in a worse state than that picture.“
__________
Villain rocked back and forth in their corner, mumbling about Hero. Their ribs screamed at them but it was a pain they were accustomed to. They wanted to sleep, but that would only come to them when they were truly exhausted. They kept their swollen eyes closed to keep the painful light out. It was the only comfort they had now.
Their will was breaking. Villain had never wanted to go home so much in their life. They were never given time to heal and the injuries were stacking up so badly. They’d already suffered water torture this morning, getting dunked in the bath for what felt like hours. It probably was; whoever had them here enjoyed their suffering. No one was going to help them. No one cared. This was all they had to live for now until they were finally permitted to die.
“They don’t care…” Villain realised tears sneaked through their flimsy defences. The cold realisation that they were alone burned their chest, or was that just their over-exerted ribs and lungs? Villain didn’t know and didn’t care anymore. Seeing Henchman had been the final straw, the final betrayal. They were alone, with only the suffering to keep them company.
When the door opened that night, Villain braced for another attack. But when Hero showed up at the door, Villain couldn’t accept what they were seeing.
“[Hero] again… and who are these…” Villain dismissed their sloppy thought process. “Hallucinations… More lies… more trickery. [Hero] doesn’t care. No one cares…”Hero came closer to them, followed by their team, and Henchman in their uniform.
“No more… please… no more,” Villain whispered.
“Shh,” Hero said gently. “Sit up. You’re stressing your ribs, sit up for me and it’ll be better.”Villain didn’t move, didn’t acknowledge what was said. “You’re… not. Not real…”
“I am, [Villain]. I’m here, I’ve come to save you. [Henchman], they raised the alarm. We came as soon as we heard.”“[Hero] hates me… won’t come.”
Hero crouched down beside them and put a warm hand on their shoulder. “I am so sorry, [Villain]. You were right, I never meant those things, but I was upset. I want to fix this- I should never have let it happen in the first place but I want to help however I can now.”
Villain still didn’t answer.“Sit up, please,” Hero said gently. “It’ll help.”Villain strained their weak limbs to force themselves back against the wall, sitting up, face shiny with sweat, hair long and filthy, tatty and unkempt.
Villain reached forward and hugged Hero, not caring for the pain in their ribs. If they hugged it, it would disappear. It would go away and leave them to rot. But it didn’t. Hero stayed there, hugged them back, stroking the back of their head.
“I’m going to fix this,” Hero promised. They paused when they heard Villain sob, the way they pressed their head into the crook of Hero’s neck. Hero felt hot tears of their own.
“[Other Hero], get me that wheelchair. We need to get them back to the base, desperately.”Other Hero left and returned triumphant a few moments later. Hero struggled to help them in, but Villain would not let go of them. “Who are they?” they asked.
Hero looked at their team in confusion. “You… don’t remember? They’re my team. They want to help you too. It’s okay.” Hero managed to slip away from Villain so they could push the wheelchair, but Hero had to hang a hand over Villain’s shoulder for them to hold.
“They care,” Villain continued to murmur again with a weak smile. “I knew they cared…” They went limp in their chair.
1K notes · View notes
daisylincs · 4 years
Note
Ok sorry if you get multiple of the same ask just ignore the copies Anyway can you do a staticquake 53, 58, 62
53 = Mutual Pining, 58 = Accidental Eavesdropping and 62 = Love Confessor (Character A confessing their love for Character B to Character C). 
Hey thanks, anon! Now, for some reason, my brain at midnight decided it would be a really good idea to combine this prompt with, of all things, a Circus AU. I cannot for the life of me explain why I thought this, except maybe that I was scrolling through the prompt list and realised that I've never actually seen a Circus AU before. (If you've ever seen one, please send it to me!! I'd love to give it a read.) 
Anyway, however it happened, here's the bullet-point fic for you! I hope you like it 😁🎪
Shield Circus is known far and wide for its incredible acts, and for how its performers have formed their own tight-knit little family. From the ringmaster, Coulson, who's basically a dad to them all, to the fearless trapeze queen May and her protégés Daisy and Elena, to Fitz and Jemma with their ever-hilarious mad science show, everyone has a place where they fit in, and everyone knows, trusts and likes everyone. 
Naturally, there's a lot of teasing around, too - with a group as close as that, what else can you expect, really? 
And since Fitzsimmons have finally gotten their act together (ha, ha) the group's collective shipping interest has moved on to Daisy. 
Specifically, Daisy and Lincoln, who everybody can see like each other - except, obviously, Daisy and Lincoln. 
Lincoln's role in the circus is a musician, and while he's still a part of the Shield family, he's not as super-close as all the performers are. Him and Daisy are friends, sure, but they're not that close, and he's sure his massive crush is unrequited. 
After all, why would someone like her, who can literally fly through the air, fall in love with him, whose only sort-of skill is playing the circus theme?
 He's made it a habit to sit and watch her train with May and Elena in the evenings - she's absolutely radiant in the air, flushed and happy with success, and always laughing and delighted with herself when she comes down to the ground again. 
Oh, he's so far gone. 
"You two drive me up the wall," Hunter, the lion-tamer, informs Lincoln as he's setting up his act. (Lincoln doesn't think Hunter can really talk, since he literally describes his job with the lions as "they can't do anything worse to me than my hellbeast wife." But he doesn't say that - Hunter and Bobbi are in one of their on phases.) 
Instead he just grumbles, "yeah, whatever, man," as he trails the trapeze artists out of the ring so he can hopefully catch Daisy for a chat before dinner. 
He's hanging around outside the changing room - not creepily, thank you very much, he and Daisy actually have a standing arrangement to meet up after she's done with training to laugh together about the day's nonsense - when he hears May's voice. 
He's a little surprised, because May usually changes very quickly so she can go over the plans for the next show with Coulson before dinner. 
And he knows he shouldn't, but if his time with Shield has taught him anything, it's that privacy is a rather vague construct around here. Eavesdropping is a thing in the circus, especially if it can get you good bribing material. 
Nobody's ever got any good bribing material on May, so can you even blame him for listening in? 
What he hears, though, is decidedly not that. 
"... sure he doesn't feel the same way?" May is asking. 
It's Daisy who replies. "I don't know! He's never shown anything that couldn't just be interpreted as just, you know, being a good friend."
 May makes a kind of unconvinced "hmm" noise. 
"Oh, don't give me that," Daisy says, exasperated. "Fitz and Jemma used to meet up after training all the time - hell, they even slept in each other's trailers." 
"And Fitz and Jemma are now dating," May says drily. 
Lincoln can just see the scowl on Daisy's face. "Okay, fine, maybe not my best example. But -" her voice catches slightly "- I don't want to go into this unless I'm absolutely sure he feels the same way. I can't… the pain of thinking someone really cares about me, then having my heart torn out when I find out the truth - I can't go through that again." 
He has never heard her sound so vulnerable, and it makes his heart clench. He knows exactly who did this to her, and if Ward ever shows his face here again, Lincoln will punch him straight into the next county. (He suspects he'll have to stand in line, though.) 
There's a kind of soft rustle in the room, and he thinks it might mean May has pulled Daisy into a hug. His heart clenches again, for entirely different reasons this time - May might seem unapproachable and impassive, but God, she cares so much. 
"Daisy," May says softly, and Lincoln almost gasps at how gentle her voice is. "I know you've been hurt, but if my life has taught me one thing, it's that we're not meant to spend our lives alone. Love is something we should all have, and I know better than anyone how hard it can be to open your heart to it, because the first step is trust."
"Trust, Daisy," she continues. "I know how hard it is, but you have to trust that Lincoln is nothing like Ward. You have to trust that he's a good guy, and that he'd never hurt you that way." 
Lincoln almost misses the rest of her sentence, because his brain has frozen at the mention of his name. Wait, wait, wait. They were talking about him?
What???
Surely this doesn't mean… if she said she's not sure he feels the same way… if May is talking about love… 
His brain blanks out completely for a few seconds as it hits him that Daisy might actually return his feelings. 
Somewhat poetically, it's Daisy's voice that brings him back to reality. "He's a really, really good friend," she says, and her voice is so warm. "You really think he feels the same way?" 
Lincoln can practically hear May rolling her eyes. "Of course he does," she says. "Daisy, that boy is gaga for you." 
Daisy laughs softly, incredulously, as though she's only just letting herself believe it. (He totally gets the feeling.) 
"Thanks, May," she says, still in that slightly breathless tone of voice. 
And Lincoln has never heard May sound so warm as she replies, "Anytime." 
He's so caught up in wondering at it all that he almost doesn't realise when the door opens. He has to scramble like a mad thing to get behind one of the lockers so the women don't see him. 
"Lincoln should be somewhere nearby," Daisy says, and he winches at the absolute accuracy of that. "We always meet up for a chat around this time." 
"And you think you two don't have anything special," May snorts. 
Daisy shakes her head. "Yeah, yeah." But she's smiling as she walks off. 
Lincoln is smiling, too, and he's so relieved and happy that he totally forgets about the locker he ducked behind. As he stands up, he bangs his head against it, really, really loudly. 
Ow. 
And also, oh, shit. 
May whips around, her eyes narrowing. "Out," she orders. 
Lincoln sheepishly comes out from behind the locker, rubbing his head. That really hurt, and now he's busted. Just perfect. 
May's eyes narrow even further, if that's even possible. "How long were you there?" she asks in her soft, dangerous tone. 
Lincoln swallows. He wouldn't dare lie to May. So, shame-facedly, he admits, "since pretty much the beginning." 
"And why," May asks in a voice that bites like a whip, "did you think it was a good idea to listen in on a private conversation?" 
"I… was hoping… to get bribing material," he says weakly. "On, uh, you." 
He's sure he's just signed his own death warrant, but to his total surprise, May snorts and rolls her eyes. "This really is Coulson's circus, isn't it? He's really rubbed his no-personal-space-we're-a-family ethic off on you all, hasn't he?" 
"Uh, yeah," Lincoln says awkwardly. 
May shakes her head, but there's only fondness in her gaze. "He'd be so frustratingly happy to hear that." 
Her gaze sobers quickly, though. "I was right, wasn't I?" she asks, getting straight to the point. He's always respected that about her. "You do love Daisy?" 
"Of course I do," he says, and he can feel a grin tugging at his lips just to admit it. Wow, it feels good - but it's nothing compared to actually KNOWING she feels the same way. 
May's gaze is amused, but, he thinks, just a bit affectionate, too. "Good," she says. "Then I think you two will be very happy." 
"Thanks," he says, feeling his grin widen and his chest expand with warmth. "That means a lot, from you." 
Because it really does - May is basically Daisy's mom. The little exchange he just overheard pretty much proves that. 
May smiles at him - an actual smile. Whoa. "Off you go," she tells him. "Go find the girl you love." 
He can't stop grinning as he turns to walk away, feeling a literal bounce in his step. Fitz really wasn't exaggerating with the whole love-makes-you-so-happy-you-could-float-away thing, was he? 
"Just one more thing," May calls as he's about to round the corner. "I don't think I really need to tell you this, but if you hurt Daisy, I promise you I will put you in that cannon and fire you through the roof." 
Lincoln has to bite back a grin at the picture, but he's never been more serious as he replies, "I hope you'll never have to do that." 
May gives him a nod, and he really feels like he's passed a test. A nod and a smile from Melinda May? He never thought he'd see the day. 
He's so preoccupied in his proud thoughts that he completely forgets to look where he's going, and crashes right into someone on the other side of the corner. 
"What the… Daisy?!?!" he exclaims as he sees who it is. Then he processes her suspicious pose leaning against the corner. "What are you… were you eavesdropping on me?" 
Daisy’s cheeks are burning, but she folds her arms and stares him down. "Well, as it turns out, I was eavesdropping on you admitting that you had eavesdropped on me," she points out, confusingly but correctly. 
He has to concede that one - and he's pretty sure his cheeks are burning, too, as he realises what exactly she would have heard. 
"So, um, you… like me, too?" he asks. 
At the exact same time, she says, "I can't believe you like me, too!" 
"Well," he says, cheeks practically on fire from burning so much, "I thought I was actually pretty obvious, what with the whole watching you train and spending every free minute chatting with you -" 
"I thought you were just being a good friend!" she protests, and somehow she's drifting a step closer to him with every word. "I didn't know that I'd end up falling for you, even though your puns are the stupidest thing in this world!" 
"My puns are awesome," he argues, letting the last bit of distance between them draw closed. "And, really, didn't you see all the similarities between us and Fitzsimmons?"
"Well, you probably thought I didn't even notice you," she reasons, raising her hands to poke at his chest but ending up just resting them there. "As if I could not notice you!"
 "Should we just agree that we're a pair of oblivious idiots, then?" he asks, reaching out to brush a curl of hair behind her ear and letting his hand linger behind her ear. 
And, really, that line should not be as romantic as it is, but Daisy's gaze turns all soft and warm. 
"Yes," she says, and her voice is breathless in the best way possible. "We should."
And then they're kissing, her hands flattening and smoothing on his chest and his hands cradling the sides of her face. 
To think he ever thought she wouldn't notice him. To think she ever thought he was just being a good friend! 
If they weren't so busy kissing, he might even see what Hunter means. 
But as it is, he's fully occupied with the wonderful feeling of Daisy's lips on his. 
Well, at least until he gears someone start clapping slowly behind them. 
"Really, Hunter?" Daisy asks without even opening her eyes. 
Hunter doesn't even have the dignity to look ashamed of himself. "Come on, Dais, you know the drill." He holds out his hand to Fitz. "Come on, mate, pay up." 
Grumbling, Fitz forks over what looks like a hundred dollars. Lincoln just shakes his head. 
Then May walks calmly around the corner and takes the money from Hunter's hand. "Actually, I think you'll find that my bet predicted this to the day," she says, and holds out her hand for everyone to pay up. 
At Daisy and Lincoln's absolutely incredulous stares, she shrugs and pockets her money. "What do you expect? I married Phil Coulson." 
Then she's walking away like the badass she is, leaving everyone else to trail after her. 
Daisy buries her head in Lincoln's shoulder. "This is a crazy family we're part of, isn't it?" 
His eyes light up with mischief. "Yeah, it's a real circus." 
Daisy's eyes widen, and she groans and slaps at his shoulder. "Oh my Goooood." 
She can feel his laughter even more than she can hear it with her cheek pressed against his chest. 
And cliché as it might be, she's never been happier. 
She's found where she truly belongs. 
"They're crazy," she agrees, looking up at him with a soft smile, "but I wouldn't trade any of it for the world." 
The End. 
8 notes · View notes
quartusbellum-blog · 6 years
Photo
Tumblr media
ARCHIE for the role of SIRIUS BLACK, using the faceclaim BLAIR REDFORD.
Your application for Sirius is stunning, and incorporates lots of little details which flesh out the character wonderfully. You’ve portrayed the complexity of Sirius’ current situation so well, and there is some truly beautiful writing here. I am very excited to welcome you to Quartus Bellum!
ooc details
Name: Archie
Age: Twenty five
Pronouns: She/her
Activity Level: I’m a current PhD candidate, so my time is pretty strapped. I am also coming out of a writing hiatus, so I’m a little rusty, but this game was so alluring I just thought I’d be an idiot to pass up some world building and exploring. I can probably be online a few times a week, but I can promise lengthy replies in lieu of my absence. I hope that’s okay. I would definitely like to keep the mod team updated on things if I’m away for whatever reason, just so we’re on the same page and everything!
Other: No triggers! But thank you very much for asking. I’m just extremely motivated and intrigued by this plot, so I have to give major kudos to such an arresting idea. Please also note that I am applying from a mockblog I have created for the purpose of this application.
Acknowledgement: I acknowledge that the themes of this game may include triggering elements. I also acknowledge that my character may be harmed, coerced, or even killed (with player’s consent) during paras/events or may cause harm to or kill others during paras/events.
                                                 ჻    ჻    ჻   ჻
general ic details
Name: Sirius Orion Black
Age: Twenty one (November 3rd, 1959)
Ships: Chemistry. Full disclosure, the biography given for Sirius gave me a lot of Sirius/James feelings (like, a lot), but I also really ship Sirius/Remus. I write Sirius as gay, but overall I’m pretty relaxed about writing relationships provided they’re realistically depicted and well-paced.
Gender/Pronouns: Genderfluid (he/him or they/them)
Face Claim: Blair Redford (x), Luke Pasqualino, Sean Teale * * I’ve gone back and forth between these three for ages... Ordinarily I write the Black family as POC, so after a Great Struggle™ in which I seriously admired Luke Pasqualino in “Snatch”, I decided to do something different and go with Blair Redford. Now, I do have a possible headcanon around Sirius and Regulus being half-brothers, so that can give the Regulus player some freedom around choosing a faceclaim, as I know matching ethnicities can be tricky (especially as Blair is half unspecified Native American). I will say, however, that I am open to discussing Sirius’ faceclaim, so if you’re unsure I’m happy to talk about it with you.
                                                 ჻    ჻    ჻   ჻
biography:
Sirius wasn’t supposed to live past twenty one.
It’s a morbid, private thought, one best left for murmuring into the black velvet of late nights, supine with firewhiskey and muggle cigarettes. See, Sirius never expected to make it to sixteen, but then a certain shaggy-haired idiot named James F. Potter happened, and Sirius’ fears went out the window (literally – have you ever tried to pry open the latch of a very old semi-sentient house that doesn’t want its heir to escape? Harder than it looks). When the war started, Sirius accepted the likelihood of his imminent death with little fanfare. It was easier, anyway, to throw himself into missions and the gutsy bravado that gripped 1979 like a fever. The city was alive: subtropical clubs; the tongues of strangers; heady muggle music; the Order laughing, packed into some tiny apartment, drunk off their tits. And even before that haze dissipated, they all felt immortal. The war was real, of course, but so were they, and Sirius was young, and dumb, and he was one of the best duelers by far, so why shouldn’t he take to the streets, Doc Martens smacking the pavement, dodging after some Death Eater? The Black household was one shrouded in death, what with the dusty portraits of forgotten ancestors, their eyes following you in the gloom, and his own mother’s obsession with mortality, as if the Pox that claimed their father was a mere token of magic’s cruel whim to give and take away. The Marauders filled him with hope; the Order stoked those embers to flames. But there was always something within him, some stoic knowledge, that this was too good too last. He was a Black: his blood ran thick as oil.
If anyone asks (which they don’t, because despite his newfound control, Sirius can still be frightening), losing James was more than a sucker punch to the gut. The Order had lost so many brave witches and wizards at the height of the war, but those terrible deaths were nothing compared to James’ disappearance. No, not disappearance. Kidnapping. Theft. They stole Prongs from Sirius’ useless fingers, swept him away for good, and Sirius was powerless. Maybe that was what hurt most of all: knowing that no matter how deeply he felt for James, how fortifying and achingly tender their friendship was, it just wasn’t enough. Sirius thought he was incapable of love before he met James. But where did that get him? The yawning dark of an empty flat; shaking hands in the cold dawn light; the blood-pound of fear in his jugular, drumming hard enough to make his eyes spot black. Sirius didn’t give himself a chance to mourn, to wonder, to do anything other than drown himself in the rescue effort. The loss of Dumbledore was similarly shattering, but Dumbledore was more figurehead than individual: a manifestation of everything the Order wanted to be. James was real: he was blood and bone. He was laughter and the glossy gold of a snitch, he was private jokes and intense bravery. He was Sirius’ counterbalance. And then he was gone.
Sirius isn’t the same. None of them are. Everything they’d fought for was extinguished in twenty four hours. That might partly explain Sirius’ habitual visits to the muggle world. Disguised as Padfoot is as good as being invisible. He can slip through their ordered, ordinary world, and feel, at least for a few hours, that his pathetic excuse for an existence hasn’t been obliterated close beyond repair. Sirius tells himself they’ll claw it all back. Dumbledore, James, the Ministry. There is a terrible anger within him that is beyond anything he has ever felt. It is cavernous, infinite, far darker and bruised than any reservoir of loathing for his family. It is so intense that he cannot even speak about it. Sirius has always been a little frightened of how deeply he feels, but this redraws those boundaries. That feeling that his life is on a countdown has compounded. Sirius is willing to do anything to take back what is rightfully theirs. He spent his youth at war. It makes sense he’ll die at war too. He’s ready to throw open his arms and embrace the abyss, laughing in delirium, Is that all you’ve got? Well come on then!
                                                 ჻    ჻    ჻   ჻
my character is:
HOW IS YOUR CHARACTER LYING TO THEMSELVES (AND HOW IS THIS SHOWN EXTERNALLY)?
Everything is fine. Everything is, of course, not fine, and is in fact irreparably fucked. But the alternative to Sirius’ externally calm demeanour is Sirius totally losing his bottle, and no one can have that, mainly because it would be the least useful thing he could do for the cause. Sirius is used to being someone people admire – there was no shortage of that in school, and even in the early period of the war: someone catching his eye hopefully, waiting for his go ahead; the mere recollection warms him with a rare, near-forgotten sense of purpose – but it’s quite another to have the final say in something. Sirius doubts himself so much. He’s not exactly a rational thinker. His vengeance is cold and cruel, that is certain, but even that type of behaviour is inherently emotional. Sirius revels in disorder: he enjoys feeling unmoored, likes not knowing. He’s not like Moony or Peter, who needed some semblance of routine to feel comfortable. Sirius quite likes feeling out of his depth; discomfort demands action. But he’s not good at communicating that, and he struggles with giving solace to someone who very badly needs to know that things are under control. Sirius hasn’t quite stooped to going, “There, there,” and patting someone awkwardly on the shoulder, but it’s close to it. He’s the first to loudly suggest a drink at some muggle pub after a disastrous mission, and he’s the last to leave, still nursing his beer long after everyone else has straggled home. Sirius isn’t eloquent like James; he isn’t calm like Moony. Hell, he doesn’t even have Wormtail’s pragmatism (before he betrayed them all, the absolute fucking bastard). Sirius is waiting for someone to catch him out. He’s not built to be a leader. The only thing he’s good for is a shag and a fun time. Right? He’s not… he’s not what they think he is. He’s useless. He’s a joke. It’s a joke. But it’s a fine joke. Ergo: everything is fine. It has to be. Otherwise he’ll drag everyone else into the flames with him, and if there’s anything Sirius is truly frightened of, it’s someone else recognizing just how deep the streak of darkness within him runs.
YOUR CHARACTER’S JOB (WHAT DO THEY DO AND HOW DO THEY FEEL ABOUT IT?)
Sirius is dedicated to the Ashen Phoenix. Even when the Order of the Phoenix still existed, when it was little more than a ragged group of idealistic Hogwarts graduates and wayward aurors, when Dumbledore’s vague effluence alternately inspired or infuriated them, back when the war seemed – well, not winnable, but certainly surmountable – even then, Sirius was too much. Too brash, too rough, too much of a muchness that made people like old Mad Eye growl under his breath about upstart sprogs. There was something to be admired in Sirius’ explosive determination, even as his reckless behaviour and breathless duels with Death Eaters was more exasperating than useful. “What?” he’d retort defensively, to a room of tired Order members. “They were asking for it.”
Sirius had always been too much. When everything – when James – when it all went to utter shite, it’s probably no wonder that Sirius lost whatever loose grip on sanity he’d ever had, and tossed it all in to band up with Mary and Lily. Lily, whom he could barely stand on a good day, who suddenly became one of the most important people standing stalwart against the uncertain scaffolding containing his so-called life. Was it really that surprising? Sirius has always privately regarded his grip on reality to be tenuous at best. Combine that with a deep, unwavering streak of hatred for blood purists, and you’ve got a terrible combination. Successful, sure; but dangerous. He can’t afford to be the rambunctious “upstart” that once semi-terrorized the Order of the Phoenix, nor can he sit about on his laurels, skulking in espionage or plotting elaborate shadowy schemes. Sirius’ patience runs thin at the best of times. No, instead he’s squashed himself into a rather uncomfortable box between “probably could be classified as a war crime” and “slightly morally questionable but still alright enough to make Evans begrudgingly admit that was a good idea”. It’s not a comfortable fit, and Sirius still isn’t sure how he ended up growing up so bloody fast, but he’ll do anything to turn back the tide of the darkness that now laps menacingly against their throats.
Aside from that, he spends quite a lot of time inadvertently posing as a muggle homeless person. Or a big shaggy dog. In comparison to being a magical fugitive, it’s almost like going on holiday.
ADDRESS THE DIFFERENCES BETWEEN WHAT YOUR CHARACTER IS CURRENTLY DOING AND WHAT THEY WOULD PREFER TO DO.
Sirius does not like responsibility. It smacks of adulthood, and Sirius never thought he’d live to see that, let alone become a ruddy pillar of virtue. It’s not that he intensely dislikes fussing over details for the Phoenixes, but it does not come naturally to him – he’s no James, put it that way (James, who was forever buzzing around them all in a manner simultaneously carefree and watchful, who’d jokingly suggest you get a jumper otherwise you’d get a cold, you bellend, so just go grab one, oh, and would you get him a chocolate frog on the way, thanks). Sirius actually doesn’t like people looking up to him. What does he know? He’s just some irresponsible dog who’d much, much rather zip away on his motorbike to blast You-Know-Who’s bits off, and sod the consequences. If he didn’t have Mary and Lily keeping him in check, Merlin knew where he’d be. Probably sharing a cell with Dumbledore. Knitting scarves and gossiping. Some lark like that. Instead he’s relegated to asking mundane questions like, “Are you sure that’s a good idea?” and, “Wait, try this healing charm.” Sirius listens to his own blather and wants to be sick. He feels the words can no one see I don’t know what I’m doing! burning every inch of him, pounding against the underside of his skin, flaring across his pained expression. “I want James back,” he said (thoughtlessly) to Lily once; and she’d shot him a look and said, “We all do.” You don’t get it, Sirius thought. I want him back because I need him. I can’t do this on my own. I need him.
Sirius does not want to turn back the clock. Despite his irreverent mindset, Sirius isn’t a fantasist. He’s emotionally charged and often irrational, but he doesn’t indulge in make-believe. There is no way things are going back to normal. There’s not even a fragment of what they left behind: those few months after school ended and before the war began, when London was besieged by an oppressive summer humidity, and the Marauders tumbled in and out of parties, drinking and laughing, carefree and stupid; the sanctity of Hogwarts, and how innocent they’d been; even old Regulus, with his pinched, shrewd expression, but the way his eyes would loosen and warm whenever Sirius ruffled his hair and affectionately called him a tosspot. Sirius cherishes these slivers of the past, counting them out like his last coins, weighing their treasure in the palm of his hand. The memories he makes now are bleak. Undernourished effigies of a world devolved. Sirius might feel beset with fear about the future, but he is still… adaptable. He was at sixteen, when he left home on a wing and a prayer, and he is now, at twenty one. No more clever, and a great deal more out of control, but able to adapt, change, mold, mend. Sirius recognizes the strange surrounding landscape and has vowed, if silently, to learn its routes, to memorize its violent topography. Survival. That was what his parents had always taught him, right? That pure blood dominates. The Marauders taught him that too, albeit in compassionate terms of friendship and trust, things Sirius had to re-learn at eleven and still is, in a way. The Order drilled him in guerrilla warfare. Dumbledore’s capture stripped him of complacency. And James… Well, survival demands vigilance. Survival turned him into something else: someone sharper, more serious, blackened around the edges. Sirius doesn’t want to turn back the clock because that would mean leaving this new version of himself behind. And like it or not, this is the only version of Sirius sodding Black that could ever make it out the other end. So, tits up to you, Voldy. This bitch ain’t going nowhere.
OOC QUESTIONS
WRITING SAMPLE:
[ REDACTED ]
EXPLORATION:
A LONG-AWAITED REUNION — One of the major subplots for this group is, of course, the return or recapture of James. From Sirius’ perspective, I think James returning to his life would signal a number of critical things – it might even be a turning point in his characterization. I see Sirius at the moment as barely hanging on. Stress exacerbates his pre-existing feelings of insufficiency and vulnerability, so he is at a stage as a member of Ashen Phoenix where he’s strung out and exhausted, burning the midnight oil, hollow-eyed and discomfortingly stoic. The loss of James was an enormously heavy blow, and that’s no overstatement. If James was somehow returned, I think that would fill Sirius’ sails with the winds of renewal and hope. He and James were a double act; they were shadows of each other. Without him, Sirius feels like a fraud.
PLAYING NICE — Sirius has never been good at pretending. His emotions run close to the surface, flashing in his quicksilver eyes at the slightest provocation. It never used to take much for him to plunge from a euphoric high to a turbulent anger, his moodiness as tempestuous as a tide. But that’s not how you lead people. You can’t expect people to see past your thunderstorm behaviour to the reality of the situation: that Sirius has always felt a split second away from free-fall. Most people aren’t like James, or Moony, or even bloody Wormtail – they can’t see that it’s all an act; that Sirius’ vulnerabilities run swift and deep, and his bravado is just a reliable way to deflect unwanted probing. Since Mouldy Voldy started swinging his shriveled old cock around, Sirius has had an abrupt about-face. It’s never easy, and he often forgets that he’s supposed to be playing nice. In fact, one could make the argument that he hasn’t changed much at all: he’s still a moody bastard. But sometimes he takes a deep breath instead of bursting in rage; sometimes he clenches his fists instead of flying for his wand. I would like to explore Sirius working hard to keep a lid on his temper, particularly given the success of Ashen Phoenix relies, at least in part, on him keeping it together for a little while longer.
THE IMMORTALITY OF REGULUS ARCTURUS BLACK — Reuniting Sirius and Regulus is a massively important subplot for me, and I think it could have powerful consequences in this group. I’m not sure how Sirius will take Regulus’ vampirism – it’ll certainly be interesting to find out. In another context I could see him falling about in horrified laughter, because now Regulus will get to hang around with Walburga forever. But I wonder now if their prolonged absence from each other will stir within Sirius a long-buried sense of responsibility. He’ll probably start worrying about Regulus, terrified that the new Dark Ministry will hunt his brother down and exterminate him. He might even (gasp!) become horribly over-protective, hating himself all the while for needling Regulus about “feeding” and all that dosh (”Shut up!” Sirius snaps as Regulus raises a single eyebrow. “It’s not like there’s a manual about becoming a bleeding vampire, is there?” A pause. “No pun intended.”) No doubt there’s a degree of irony in an ex-Death Eater suddenly becoming the prey of his old buddies, but Sirius isn’t a masochist. As stupid as Regulus has behaved, they’re still brothers. Even before the war turned for the worst, Sirius still missed him. Yearned for him to be back. Regulus was the biggest idiot in Britain, but he was Sirius’ idiot, and if Sirius had heard more than a whisper about his brother he would have probably done something very stupid to rush to his side. Call Sirius a lot of things, but being disloyal could never be one of them.
EXTRAS:
I have created a mockblog for this group, which is the account I am submitting from. You can find it here. I have also written some general headcanons, which you can find below!
Sirius started getting muggle tattoos during the war. The first war, that is, back when they all thought it’d be over by Christmas. He’s got about seventeen, possibly a few more, all in black ink, most of them done in poky muggle tattoo parlours buried in the heart of London, but a couple of them are magical: the dragon across his left shoulder blade, for example, which sneezes fire when you tickle it just right. It’s an eclectic collection that illustrate Sirius’ natural whimsy: a series of ancient runes that Moony told him meant something cool (although Sirius has since suspected Moony was an absolute tosser, and the runes in fact spell “totally gullible gobshite”); an elaborate diagram of the planets in the middle of his back; a broom that zips around his arm (James’ fault, that one); an anatomically dubious pin-up girl (Sirius wanted a guy, but the tattoo artist looked frightening, and Sirius wasn’t in the mood to go toe-to-toe about his sexual preferences); and, for reasons best left alone, ancient constellations scattered most of his chest. There are some other tattoos squashed in here and there – a Gryffindor lion, a protection symbol that Moony literally laughed aloud at when Sirius showed it off – that are mainly impulse decisions. Sirius loves them all. The ink is so black against his brown skin, the magical designs flickering in the corner of his eye, and it all gives the illusion of him appearing alive, ever in motion, an intricate living illustration.
Sirius still owns his motorbike, although it’s too dangerous to ride it. Some arsehole (Bellatrix, probably) ratted him out, and now everyone and their mother is on the hunt for a sleek black motorbike. He isn’t stupid enough to ride it, no matter how burning the urge, although sometimes he does go out to Clapham, where he’d parked it in a muggle garage, just to linger over it for a few stolen moments. One day he’ll blaze it right over London, preferably in celebration of Balding Voldy’s bloody demise. One day. He will.
Sirius is gay. The revelation came unobtrusively. He’d always known there was… something awry. You’d have to be blind, deaf, and dumb not to be surrounded by all of those posh pureblood birds growing up and feel nothing more than resignation at their proximity (their brothers were far more intriguing). Sirius played along for a bit at school, going out with a few girls, making out with McKinnon at more than a few parties. It was all serviceable, except for the fact it was tremendously boring, and if there was anything Sirius resented, it was feeling confined. He came out (very loudly) in his sixth year (in the middle of the Great Hall; it was quite the gossip for a week or so), and has since been perfectly content with advertising his sexuality at the nearest opportunity. He’s no blushing violet, put it that way. While Wizarding society is more or less accepting of sexuality (his parents notwithstanding: Sirius was still expected to marry and produce an heir; and that thought, of dragging some brat into the world through duty alone, turned his stomach), the fragmentation in muggle society is something else. Sirius is still too enthusiastic about muggle life to ever really fit in – he’s been asked innumerable times if he’s a tourist, which absolutely delights him – but the gay and lesbian rights movements in recent years has captured his attention. He’s kept up with the news about protests, and once apparated into an alley adjacent to a march for queer liberation. We’re here, the muggles chanted, we’re queer! We won’t disappear! The feeling was incredible. Wixen didn’t have anything like this – it was all just taken for granted. But the fight of the muggles. Their determination; their spirit. Their strength in demanding what was theirs. It left him breathless, and for the first time in his life, proud.
Sirius spends a lot of time as Padfoot these days. It’s just easier. He’s a dab hand at disguise charms – had to be, when the war started to turn truly dark and a Black blood traitor head on a spit was a coveted prize – but outside of a handful of people, Sirius’ animagus form is a secret. Lily knows, of course, as does Macdonald, but they have to. Slinking around London as a dog makes for surreptitious travel, even if he’s taken on some bad habits as a human. Fleas genuinely are the worst, alright? He can’t help scratching himself fiercely at the slightest itch.
The way Sirius dresses now is a diluted version of the summer of 1979. Back then, London was a heaving cesspool of cramped, humid clubs, gigs outfitted in leather and gelled spikes, tight chokers, and a casual, careless androgyny that made Sirius’ heart beat fast. Back then he didn’t give a toss. Now, of course, he’s no longer a naive graduate, and the world has grown dim. He usually wears a leather jacket over some band t-shirt, a pair of black ripped jeans, Doc Martens. That’s toned down, for him. While a lot of the jewellery has gone, Sirius’ fingers are still bejewelled with rows of heavy silver rings, and a dragon tooth earring swings from his left ear. His eyes, once glittering with flirtatious humour, are ringed dark with wariness; and his cutting bone structure has sharpened with one too many missed meals. Sirius is probably physically bulkier than he was at school, simply because sleeping rough and hauling arse after a dozen Death Eaters tends to fill you out, but his body is still lean, with an echo of that languid grace that whispered of pureblood ballrooms and charity galas. Sirius’ hair has grown long, and he usually ties it sloppily away from his face, but he stays clean shaven… most of the time. Lily once said he looked like a Lennon on a bad trip after Sirius reappeared after a rendezvous in Dublin for four weeks. He’s still trying to figure out if that was an insult.
6 notes · View notes
samanthasmeyers · 4 years
Text
[Experiment] How AI is Changing the Way We Optimize at Unbounce
When I first encountered A/B testing, I immediately wanted to become the type of marketer who tested everything. The idea sounded fun to me. Like being a mad scientist running experiments to prove when my work was actually “working.”
Turns out though, there’s always a long list of other things to do first… blog posts to write, campaigns to launch, and don’t get me started on the meetings! I’m not alone in this, either. A lot of marketers are just too darned busy to follow up and optimize the stuff they’ve already shipped. According to HubSpot, only 17% of marketers use landing page A/B tests to improve conversion rates.
A small glimpse of my ever-growing to-do list. Ain’t nobody got time for A/B tests.
Sure, running a split test with one or two variants always sounds easy enough. But once you take a closer look at the process, you realize just how complex it can actually be. You need to make sure you have… 
The right duration and sample size.
Taken into account any external factors or validity threats.
Learned how to interpret the results correctly, too.
But—while there will always be a time and place for A/B testing—there’s also now an easier and faster way for marketers to optimize. Smart Traffic is a new Unbounce tool that uses the power of AI and machine learning to get you more conversions. Every day, more marketers are using Smart Traffic to “automagically” optimize their landing pages. But whenever we launch anything new, we like to test it out for ourselves to learn alongside you (and keep you up to speed on what to try next).
Here’s what I learned after taking Smart Traffic for a test drive myself…
Shifting Your Mindset to Optimize with AI
I know many marketers are (perhaps) skeptical when it comes to promises of machine learning, artificial intelligence, or magical “easy” buttons that get them better results. But AI is all around us and it’s already changing the way we do marketing. Landing page optimization is just one more area of the job where you no longer need to do everything yourself manually.
Smart Traffic augments your marketing skills and automatically sends visitors to the landing page variant where they’re most likely to convert (based on how similar page visitors have converted before). It makes routing decisions faster than any human ever could (thank you, AI magic), and “learns” which page variant is a perfect match for each different visitor. This ultimately means no more “champion” variants. Instead, you’re free to create multiple different pages to appeal to different groups of visitors and run ‘em all at once.
This is very different from A/B testing and honestly—it can feel kinda weird at first. You’ve got to trust in the machine learning to figure out what works best and what doesn’t. Data scientists call this the “black box” problem: data goes in, decisions come out, but you never really get the full understanding of what happened in between. 
Smart Traffic is fundamentally different from A/B Testing. You can learn more about how it works here.
For marketers using Smart Traffic, this means shifting your mindset and starting to think about optimization differently. Unlike A/B testing, you’re not looking for those “aha” moments to apply to your next campaign, or a one-size-fits-all “winning” variant. Instead, you’re looking to discover what works best for different subsets of your audience. This gives you unlimited creativity to try out new marketing ideas, makes it easier and less risky for you to optimize, and gives you an average conversion lift of 30% compared to splitting the traffic evenly across multiple variants. (Woah.)
My Experiment with Smart Traffic
I know all this because I recently experimented with variant creation myself to better understand this new AI optimization mindset. I created 15 variants across two separate landing pages using Smart Traffic to discover…
How easy is it to optimize with an AI-powered optimization tool?
Could I quickly set up the tests in Unbounce while still getting those other to-do’s done?
What kind of conversion lift would I see from just a few hours invested?
I took a little bit of my inspiration from Ms. Frizzle on the Magic School Bus. No, not her haircut, her catchphrase: “Take chances, make mistakes, get messy!”
Oh, so that’s where she got all her good ideas.
Creating 15 Variants in Under Two Hours
The beauty of Smart Traffic is there are no limits to how many variants you can create and it automatically starts optimizing in as few as 50 visits. Just hit the “optimize” button and you’re off to the races. Could it really be that simple?
My guinea pigs for this experiment would be two recent campaigns our marketing team had worked on: the ecommerce lookbook and the SaaS optimization guide. The team had created both of these ebook download pages in Unbounce, but we hadn’t been able to return to them and optimize very much in the months since we published.
The original landing pages would serve as my control variants. (Click to see the full pages.)
Before starting, I consulted with Anna Roginska, Growth Marketer at Unbounce, to get her input on how I should create my variants. She advised:
You can take the ‘spaghetti at the wall’ approach, where you create a bunch of variants and just leave them to Smart Traffic to see what happens. It’s that ‘set it and forget it’ mentality. That’s interesting, but when you look at a bowl of spaghetti… There’s a lot of noodles in there. You won’t necessarily get to explain why something is working or not working.
The other approach is to be more strategic and focused. I think there’s a huge benefit to going in with a plan. Create maybe only five variants and give them each a specific purpose. Then, you can see how they perform and create new iterations for different portions of the audience.
I had two landing pages to work with, so I thought I’d give both approaches a try. But with only a few hours scheduled in my calendar to complete all these variants, I needed to move fast.
The “Spaghetti at the Wall” Approach to Variant Creation
On the ecommerce lookbook page, I wanted to spend less time planning and more time creating. Whereas in A/B testing you need a proper test hypothesis and a careful plan for each variant, Smart Traffic lets you get creative and try out new ideas on the fly. Your variants don’t have to be perfect—they just need to be different enough to appeal to new audience segments.
This meant I didn’t have to make any hard or fast choices about which one element to “test” on the landing page. I could create 15 different variants that varied wildly from one another. Some used different colors, some had different headlines, some completely changed up the layout of the page.
This is something you just can’t do in a traditional A/B test where you’re looking to find a “winner” and understand why it “wins.” I had to remind myself I wasn’t looking for that one variant to rule them all (or for that one variant to bring them all and in the darkness bind them). I was looking to increase the chance of conversion for every single visitor. Certain pages were going to work better for certain audiences, and that was totally fine.
I wondered, though: how many variants would be too many? Would the machine learning recognize that some of these were not anything special and just stop sending traffic to them? And how long would it take to get results? With these questions in mind, I checked back on my first set of tests one month later…
Changing up the background color
Usually, color A/B tests are pretty much a waste of time. You need a lot of data to get accurate results, and most marketers don’t actually end up learning anything useful in the end. (Because color by itself means nothing, it always depends on the context of the page.)
That being said, we know there is some legitimate color theory and certain audience segments respond better to certain colors than others. So I thought it might be interesting to switch up the background on this landing page to see what would happen. And color me surprised—these variants are seeing some pretty dramatically different conversion rates:
Pink background – 12.82%
Green background – 21.43%
White background – 21.74%
Black background – 31.71%
One might start to speculate from these conversion rates that darker backgrounds perform better than the lighter backgrounds. But hold your horses, that’s thinking about this as an A/B test again. Here’s why Jordan Dawe, Senior Data Science Developer at Unbounce, says you should be cautious about drawing any conclusions from the conversion rates…
Smart Traffic is not sending visitors randomly—it’s trying to get the best traffic to the best variant. So in this case, it doesn’t mean that a black background will always convert higher than a pink background. There are likely portions of the audience going to each color that would be doing worse on others. Here’s what you can conclude: the color black is preferred by a portion of the traffic that converts highly.
It’s hard to shake that mindset of looking for a “winner” and trying to figure out “why” something is working. But I was starting to accept that different portions of the audience would always respond better to different variants—this was just the first time I’d been able to use AI to automatically serve up the best version.
Making big (and small) changes to the headline
For the next group of variants, I switched up the H1 in both small and big ways to see what effect that would have on the conversion rate. In some cases, this meant just swapping a single adjective (e.g., “jaw-dropping” for “drool-worthy”). In other cases, I went with a completely new line of copy altogether.
Here’s how the variants stacked up against each other:
See 27 Sales-Ready Ecommerce Landing Pages in Our Ultimate Lookbook – 25.81%
See 27 Stunning Ecommerce Landing Pages in Our Ultimate Lookbook – 25.93%
Get Ready to See 27 Jaw-Dropping Ecommerce Landing Page Examples – 28.13%
Get Serious Inspo for Supercharging Your Ecomm Sales – 35%
See 27 Drool-Worthy Ecommerce Landing Pages in Our Ultimate Lookbook – 40%
Again, each variant yielded a different conversion rate. I wondered if I kept testing different variations of the headlines and found one that performed best, could I deactivate all the other headline variants and just go with the “best” one? 
Here’s how Floss Taylor, Data Analyst at Unbounce, responded…
Smart Traffic doesn’t have champion variants. You don’t pick one at the end like you would in an A/B test. Although one variant may appear to be performing poorly, there could be a subset of traffic that it’s ideal for. You’re better off leaving it on long-term so it can work its magic.
Trying out different page layouts and hierarchies
The last set of variants I created messed with the actual structure and hierarchy of the page. I wanted to see if moving things around (or removing sections entirely) would influence the conversion rate. Here’s a sample of some of the experiments…
Removing the Headline – 16.67%
Adding a Double CTA – 21.95%
Moving the Testimonial Up the Page – 27.27%
Nothing too surprising here. And because I had created so many variants, Smart Traffic was taking longer than usual in “Learning Mode” to start giving me a conversion lift. Here’s how Floss Taylor explains it…
Smart Traffic needs approximately 50 visitors to understand which traffic would perform well for each new variant. If you have 15 variants and ~100 visitors per month, you’re going to have a long learning period where Smart Traffic cannot make accurate recommendations. I’d suggest starting off with a lower number of variants, and only adding more once once you have sufficient traffic.
The “Strategic Marketer” Approach
So throwing spaghetti at the wall turned out to be… messy. (New parents beware.) For the SaaS optimization guide page, I wanted to be a bit more strategic. And I actually had a leg up for this one, because Anna Roginska, Growth Marketer at Unbounce, had already started with a Smart Traffic experiment on this page four months ago.
Anna had set up a test between two different variants. One had an image of the ecommerce lookbook as the hero graphic on the page, while the other used the image of conversion expert and author Talia Wolf. Anna says she decided on this second variant because of research she had seen on how photographs of people tend to convert better than products.
I put Talia up front because I knew from other tests I’ve run and research I’ve done. [Photographs of] people tend to convert better. I didn’t know if it would work better in this particular case, but I was able to set up a variant and use Smart Traffic to find out. And it just so happens that the algorithm started sending way more traffic to this variant.
Anna seemed to be onto something, too: her variant was converting at nearly double the rate for a large traffic subset. And while I now know we can’t consider this a “champion” variant like in an A/B test and learn from the results, we could iterate based on her design to target new audience segments.
I created a simple spreadsheet to develop my gameplan. The goal was to create five new versions of the page that would appeal to different visitors based on their attributes:
Reducing the word count to target mobile and “ready to download” visitors
For inspiration on my first variant, I consulted the 2020 Conversion Benchmark Report. The machine learning insights here suggested that SaaS landing pages with lower word counts and easier-to-read copy tend to perform better than their long-winded counterparts.
And while the original version of our download page was easy enough to read, it did have a long, wordy intro with a lot of extra detail. Could I increase our conversion rate for a portion of our audience if just focused on the bare essentials? I was ready to kill some darlings to find out…
Original Long-Form Version – 10%
Low Word Count Version – 21.43%
It seems there’s a segment of our traffic coming to this page who didn’t need to see all that extra info before they decided to fill out the form. I speculated that this variant might also perform better on mobile devices since it would be faster-loading and easier to scroll through. Interesting!
Switching the headline to target different audience segments
Next, I created an additional four page variants to speak to the different pain points and reasons our audience might want to download the guide. (Actually, this is something Talia herself recommends you do in the SaaS optimization guide.) I switched up the headline copy here, as well as some of the supporting text underneath to match. After a month, here’s what the conversion rates look like:
Get Talia’s Guide to Optimize – 19.05%
You Can’t Just Build – 23.08%
Optimization is a Lot of Work – 24%
Not Sure How to Optimize? – 33.33%
Each variant is serving a different segment of the audience, by speaking to the particular reason they want to download the guide most (e.g., maybe they don’t have the time to optimize, or maybe they don’t know how to get started). As Smart Traffic learns more about which variants perform best for which audience segments, we become that much more likely to score a conversion.
What I Learned Running These Smart Traffic Experiments
Smart Traffic absolutely makes optimization easier and faster for marketers who previously never had the time (or experience) to run A/B tests. It took me under two hours to set up and launch these experiments, and we’re already seeing some pretty impressive results just over a month later.
While the ecommerce lookbook page is still optimizing, the SaaS ebook page is showing a 12% lift in conversions compared to evenly splitting traffic among all these variants. And this is after only a month—the algorithm will keep improving to get us even better results over time. (Like a fine wine, or that suspiciously old cheese in my fridge.)
At the same time, I did walk away with a few important lessons learned. If you’re planning to use Smart Traffic to optimize your landing pages, here are some things to keep in mind before you get started:
There are no champion variants – Unlike traditional A/B testing, you won’t be able to point to one landing page variant at the end of your test and call it a winner. The machine learning algorithm automatically routes audiences differently based on their individual attributes, which means you have to be cautious when you’re analyzing the results.
The more variants you create, the longer you’ll wait – While it can be tempting to throw spaghetti at the wall and create dozens of variants for your landing page, this means you’ll also have to wait longer to see what sticks. Try starting out with three to five variations and take a more strategic approach based on research in your industry. (The 2020 Conversion Benchmark Report is a great place to start for some ideas.)
It’s (usually) better to leave low-converting variants active – Because Smart Traffic learns over time and continually improves, you’re typically better off leaving your variants active—even if their conversion rates aren’t all that impressive. The AI takes the risk out of optimization by automatically sending visitors to the page that suits them best. If you turn off variants, you may lose out on some of those conversions altogether.
It can be a lot of fun to get creative with the different page elements and try out new ideas. You just might want to come up with a bit of a plan first and be strategic with your approach. Still, it’s better to experiment and optimize with Smart Traffic (even if you make some mistakes along the way) than to never optimize at all.
(And in case you were worried, yep—I managed to get my to-do list done, too. )
from Marketing https://unbounce.com/marketing-ai/smart-traffic-experiments/ via http://www.rssmix.com/
0 notes
roypstickney · 4 years
Text
[Experiment] How AI is Changing the Way We Optimize at Unbounce
When I first encountered A/B testing, I immediately wanted to become the type of marketer who tested everything. The idea sounded fun to me. Like being a mad scientist running experiments to prove when my work was actually “working.”
Turns out though, there’s always a long list of other things to do first… blog posts to write, campaigns to launch, and don’t get me started on the meetings! I’m not alone in this, either. A lot of marketers are just too darned busy to follow up and optimize the stuff they’ve already shipped. According to HubSpot, only 17% of marketers use landing page A/B tests to improve conversion rates.
A small glimpse of my ever-growing to-do list. Ain’t nobody got time for A/B tests.
Sure, running a split test with one or two variants always sounds easy enough. But once you take a closer look at the process, you realize just how complex it can actually be. You need to make sure you have… 
The right duration and sample size.
Taken into account any external factors or validity threats.
Learned how to interpret the results correctly, too.
But—while there will always be a time and place for A/B testing—there’s also now an easier and faster way for marketers to optimize. Smart Traffic is a new Unbounce tool that uses the power of AI and machine learning to get you more conversions. Every day, more marketers are using Smart Traffic to “automagically” optimize their landing pages. But whenever we launch anything new, we like to test it out for ourselves to learn alongside you (and keep you up to speed on what to try next).
Here’s what I learned after taking Smart Traffic for a test drive myself…
Shifting Your Mindset to Optimize with AI
I know many marketers are (perhaps) skeptical when it comes to promises of machine learning, artificial intelligence, or magical “easy” buttons that get them better results. But AI is all around us and it’s already changing the way we do marketing. Landing page optimization is just one more area of the job where you no longer need to do everything yourself manually.
Smart Traffic augments your marketing skills and automatically sends visitors to the landing page variant where they’re most likely to convert (based on how similar page visitors have converted before). It makes routing decisions faster than any human ever could (thank you, AI magic), and “learns” which page variant is a perfect match for each different visitor. This ultimately means no more “champion” variants. Instead, you’re free to create multiple different pages to appeal to different groups of visitors and run ‘em all at once.
This is very different from A/B testing and honestly—it can feel kinda weird at first. You’ve got to trust in the machine learning to figure out what works best and what doesn’t. Data scientists call this the “black box” problem: data goes in, decisions come out, but you never really get the full understanding of what happened in between. 
Smart Traffic is fundamentally different from A/B Testing. You can learn more about how it works here.
For marketers using Smart Traffic, this means shifting your mindset and starting to think about optimization differently. Unlike A/B testing, you’re not looking for those “aha” moments to apply to your next campaign, or a one-size-fits-all “winning” variant. Instead, you’re looking to discover what works best for different subsets of your audience. This gives you unlimited creativity to try out new marketing ideas, makes it easier and less risky for you to optimize, and gives you an average conversion lift of 30% compared to splitting the traffic evenly across multiple variants. (Woah.)
My Experiment with Smart Traffic
I know all this because I recently experimented with variant creation myself to better understand this new AI optimization mindset. I created 15 variants across two separate landing pages using Smart Traffic to discover…
How easy is it to optimize with an AI-powered optimization tool?
Could I quickly set up the tests in Unbounce while still getting those other to-do’s done?
What kind of conversion lift would I see from just a few hours invested?
I took a little bit of my inspiration from Ms. Frizzle on the Magic School Bus. No, not her haircut, her catchphrase: “Take chances, make mistakes, get messy!”
Oh, so that’s where she got all her good ideas.
Creating 15 Variants in Under Two Hours
The beauty of Smart Traffic is there are no limits to how many variants you can create and it automatically starts optimizing in as few as 50 visits. Just hit the “optimize” button and you’re off to the races. Could it really be that simple?
My guinea pigs for this experiment would be two recent campaigns our marketing team had worked on: the ecommerce lookbook and the SaaS optimization guide. The team had created both of these ebook download pages in Unbounce, but we hadn’t been able to return to them and optimize very much in the months since we published.
The original landing pages would serve as my control variants. (Click to see the full pages.)
Before starting, I consulted with Anna Roginska, Growth Marketer at Unbounce, to get her input on how I should create my variants. She advised:
You can take the ‘spaghetti at the wall’ approach, where you create a bunch of variants and just leave them to Smart Traffic to see what happens. It’s that ‘set it and forget it’ mentality. That’s interesting, but when you look at a bowl of spaghetti… There’s a lot of noodles in there. You won’t necessarily get to explain why something is working or not working.
The other approach is to be more strategic and focused. I think there’s a huge benefit to going in with a plan. Create maybe only five variants and give them each a specific purpose. Then, you can see how they perform and create new iterations for different portions of the audience.
I had two landing pages to work with, so I thought I’d give both approaches a try. But with only a few hours scheduled in my calendar to complete all these variants, I needed to move fast.
The “Spaghetti at the Wall” Approach to Variant Creation
On the ecommerce lookbook page, I wanted to spend less time planning and more time creating. Whereas in A/B testing you need a proper test hypothesis and a careful plan for each variant, Smart Traffic lets you get creative and try out new ideas on the fly. Your variants don’t have to be perfect—they just need to be different enough to appeal to new audience segments.
This meant I didn’t have to make any hard or fast choices about which one element to “test” on the landing page. I could create 15 different variants that varied wildly from one another. Some used different colors, some had different headlines, some completely changed up the layout of the page.
This is something you just can’t do in a traditional A/B test where you’re looking to find a “winner” and understand why it “wins.” I had to remind myself I wasn’t looking for that one variant to rule them all (or for that one variant to bring them all and in the darkness bind them). I was looking to increase the chance of conversion for every single visitor. Certain pages were going to work better for certain audiences, and that was totally fine.
I wondered, though: how many variants would be too many? Would the machine learning recognize that some of these were not anything special and just stop sending traffic to them? And how long would it take to get results? With these questions in mind, I checked back on my first set of tests one month later…
Changing up the background color
Usually, color A/B tests are pretty much a waste of time. You need a lot of data to get accurate results, and most marketers don’t actually end up learning anything useful in the end. (Because color by itself means nothing, it always depends on the context of the page.)
That being said, we know there is some legitimate color theory and certain audience segments respond better to certain colors than others. So I thought it might be interesting to switch up the background on this landing page to see what would happen. And color me surprised—these variants are seeing some pretty dramatically different conversion rates:
Pink background – 12.82%
Green background – 21.43%
White background – 21.74%
Black background – 31.71%
One might start to speculate from these conversion rates that darker backgrounds perform better than the lighter backgrounds. But hold your horses, that’s thinking about this as an A/B test again. Here’s why Jordan Dawe, Senior Data Science Developer at Unbounce, says you should be cautious about drawing any conclusions from the conversion rates…
Smart Traffic is not sending visitors randomly—it’s trying to get the best traffic to the best variant. So in this case, it doesn’t mean that a black background will always convert higher than a pink background. There are likely portions of the audience going to each color that would be doing worse on others. Here’s what you can conclude: the color black is preferred by a portion of the traffic that converts highly.
It’s hard to shake that mindset of looking for a “winner” and trying to figure out “why” something is working. But I was starting to accept that different portions of the audience would always respond better to different variants—this was just the first time I’d been able to use AI to automatically serve up the best version.
Making big (and small) changes to the headline
For the next group of variants, I switched up the H1 in both small and big ways to see what effect that would have on the conversion rate. In some cases, this meant just swapping a single adjective (e.g., “jaw-dropping” for “drool-worthy”). In other cases, I went with a completely new line of copy altogether.
Here’s how the variants stacked up against each other:
See 27 Sales-Ready Ecommerce Landing Pages in Our Ultimate Lookbook – 25.81%
See 27 Stunning Ecommerce Landing Pages in Our Ultimate Lookbook – 25.93%
Get Ready to See 27 Jaw-Dropping Ecommerce Landing Page Examples – 28.13%
Get Serious Inspo for Supercharging Your Ecomm Sales – 35%
See 27 Drool-Worthy Ecommerce Landing Pages in Our Ultimate Lookbook – 40%
Again, each variant yielded a different conversion rate. I wondered if I kept testing different variations of the headlines and found one that performed best, could I deactivate all the other headline variants and just go with the “best” one? 
Here’s how Floss Taylor, Data Analyst at Unbounce, responded…
Smart Traffic doesn’t have champion variants. You don’t pick one at the end like you would in an A/B test. Although one variant may appear to be performing poorly, there could be a subset of traffic that it’s ideal for. You’re better off leaving it on long-term so it can work its magic.
Trying out different page layouts and hierarchies
The last set of variants I created messed with the actual structure and hierarchy of the page. I wanted to see if moving things around (or removing sections entirely) would influence the conversion rate. Here’s a sample of some of the experiments…
Removing the Headline – 16.67%
Adding a Double CTA – 21.95%
Moving the Testimonial Up the Page – 27.27%
Nothing too surprising here. And because I had created so many variants, Smart Traffic was taking longer than usual in “Learning Mode” to start giving me a conversion lift. Here’s how Floss Taylor explains it…
Smart Traffic needs approximately 50 visitors to understand which traffic would perform well for each new variant. If you have 15 variants and ~100 visitors per month, you’re going to have a long learning period where Smart Traffic cannot make accurate recommendations. I’d suggest starting off with a lower number of variants, and only adding more once once you have sufficient traffic.
The “Strategic Marketer” Approach
So throwing spaghetti at the wall turned out to be… messy. (New parents beware.) For the SaaS optimization guide page, I wanted to be a bit more strategic. And I actually had a leg up for this one, because Anna Roginska, Growth Marketer at Unbounce, had already started with a Smart Traffic experiment on this page four months ago.
Anna had set up a test between two different variants. One had an image of the ecommerce lookbook as the hero graphic on the page, while the other used the image of conversion expert and author Talia Wolf. Anna says she decided on this second variant because of research she had seen on how photographs of people tend to convert better than products.
I put Talia up front because I knew from other tests I’ve run and research I’ve done. [Photographs of] people tend to convert better. I didn’t know if it would work better in this particular case, but I was able to set up a variant and use Smart Traffic to find out. And it just so happens that the algorithm started sending way more traffic to this variant.
Anna seemed to be onto something, too: her variant was converting at nearly double the rate for a large traffic subset. And while I now know we can’t consider this a “champion” variant like in an A/B test and learn from the results, we could iterate based on her design to target new audience segments.
I created a simple spreadsheet to develop my gameplan. The goal was to create five new versions of the page that would appeal to different visitors based on their attributes:
Reducing the word count to target mobile and “ready to download” visitors
For inspiration on my first variant, I consulted the 2020 Conversion Benchmark Report. The machine learning insights here suggested that SaaS landing pages with lower word counts and easier-to-read copy tend to perform better than their long-winded counterparts.
And while the original version of our download page was easy enough to read, it did have a long, wordy intro with a lot of extra detail. Could I increase our conversion rate for a portion of our audience if just focused on the bare essentials? I was ready to kill some darlings to find out…
Original Long-Form Version – 10%
Low Word Count Version – 21.43%
It seems there’s a segment of our traffic coming to this page who didn’t need to see all that extra info before they decided to fill out the form. I speculated that this variant might also perform better on mobile devices since it would be faster-loading and easier to scroll through. Interesting!
Switching the headline to target different audience segments
Next, I created an additional four page variants to speak to the different pain points and reasons our audience might want to download the guide. (Actually, this is something Talia herself recommends you do in the SaaS optimization guide.) I switched up the headline copy here, as well as some of the supporting text underneath to match. After a month, here’s what the conversion rates look like:
Get Talia’s Guide to Optimize – 19.05%
You Can’t Just Build – 23.08%
Optimization is a Lot of Work – 24%
Not Sure How to Optimize? – 33.33%
Each variant is serving a different segment of the audience, by speaking to the particular reason they want to download the guide most (e.g., maybe they don’t have the time to optimize, or maybe they don’t know how to get started). As Smart Traffic learns more about which variants perform best for which audience segments, we become that much more likely to score a conversion.
What I Learned Running These Smart Traffic Experiments
Smart Traffic absolutely makes optimization easier and faster for marketers who previously never had the time (or experience) to run A/B tests. It took me under two hours to set up and launch these experiments, and we’re already seeing some pretty impressive results just over a month later.
While the ecommerce lookbook page is still optimizing, the SaaS ebook page is showing a 12% lift in conversions compared to evenly splitting traffic among all these variants. And this is after only a month—the algorithm will keep improving to get us even better results over time. (Like a fine wine, or that suspiciously old cheese in my fridge.)
At the same time, I did walk away with a few important lessons learned. If you’re planning to use Smart Traffic to optimize your landing pages, here are some things to keep in mind before you get started:
There are no champion variants – Unlike traditional A/B testing, you won’t be able to point to one landing page variant at the end of your test and call it a winner. The machine learning algorithm automatically routes audiences differently based on their individual attributes, which means you have to be cautious when you’re analyzing the results.
The more variants you create, the longer you’ll wait – While it can be tempting to throw spaghetti at the wall and create dozens of variants for your landing page, this means you’ll also have to wait longer to see what sticks. Try starting out with three to five variations and take a more strategic approach based on research in your industry. (The 2020 Conversion Benchmark Report is a great place to start for some ideas.)
It’s (usually) better to leave low-converting variants active – Because Smart Traffic learns over time and continually improves, you’re typically better off leaving your variants active—even if their conversion rates aren’t all that impressive. The AI takes the risk out of optimization by automatically sending visitors to the page that suits them best. If you turn off variants, you may lose out on some of those conversions altogether.
It can be a lot of fun to get creative with the different page elements and try out new ideas. You just might want to come up with a bit of a plan first and be strategic with your approach. Still, it’s better to experiment and optimize with Smart Traffic (even if you make some mistakes along the way) than to never optimize at all.
(And in case you were worried, yep—I managed to get my to-do list done, too. )
0 notes
jjonassevilla · 4 years
Text
[Experiment] How AI is Changing the Way We Optimize at Unbounce
When I first encountered A/B testing, I immediately wanted to become the type of marketer who tested everything. The idea sounded fun to me. Like being a mad scientist running experiments to prove when my work was actually “working.”
Turns out though, there’s always a long list of other things to do first… blog posts to write, campaigns to launch, and don’t get me started on the meetings! I’m not alone in this, either. A lot of marketers are just too darned busy to follow up and optimize the stuff they’ve already shipped. According to HubSpot, only 17% of marketers use landing page A/B tests to improve conversion rates.
A small glimpse of my ever-growing to-do list. Ain’t nobody got time for A/B tests.
Sure, running a split test with one or two variants always sounds easy enough. But once you take a closer look at the process, you realize just how complex it can actually be. You need to make sure you have… 
The right duration and sample size.
Taken into account any external factors or validity threats.
Learned how to interpret the results correctly, too.
But—while there will always be a time and place for A/B testing—there’s also now an easier and faster way for marketers to optimize. Smart Traffic is a new Unbounce tool that uses the power of AI and machine learning to get you more conversions. Every day, more marketers are using Smart Traffic to “automagically” optimize their landing pages. But whenever we launch anything new, we like to test it out for ourselves to learn alongside you (and keep you up to speed on what to try next).
Here’s what I learned after taking Smart Traffic for a test drive myself…
Shifting Your Mindset to Optimize with AI
I know many marketers are (perhaps) skeptical when it comes to promises of machine learning, artificial intelligence, or magical “easy” buttons that get them better results. But AI is all around us and it’s already changing the way we do marketing. Landing page optimization is just one more area of the job where you no longer need to do everything yourself manually.
Smart Traffic augments your marketing skills and automatically sends visitors to the landing page variant where they’re most likely to convert (based on how similar page visitors have converted before). It makes routing decisions faster than any human ever could (thank you, AI magic), and “learns” which page variant is a perfect match for each different visitor. This ultimately means no more “champion” variants. Instead, you’re free to create multiple different pages to appeal to different groups of visitors and run ‘em all at once.
This is very different from A/B testing and honestly—it can feel kinda weird at first. You’ve got to trust in the machine learning to figure out what works best and what doesn’t. Data scientists call this the “black box” problem: data goes in, decisions come out, but you never really get the full understanding of what happened in between. 
Smart Traffic is fundamentally different from A/B Testing. You can learn more about how it works here.
For marketers using Smart Traffic, this means shifting your mindset and starting to think about optimization differently. Unlike A/B testing, you’re not looking for those “aha” moments to apply to your next campaign, or a one-size-fits-all “winning” variant. Instead, you’re looking to discover what works best for different subsets of your audience. This gives you unlimited creativity to try out new marketing ideas, makes it easier and less risky for you to optimize, and gives you an average conversion lift of 30% compared to splitting the traffic evenly across multiple variants. (Woah.)
My Experiment with Smart Traffic
I know all this because I recently experimented with variant creation myself to better understand this new AI optimization mindset. I created 15 variants across two separate landing pages using Smart Traffic to discover…
How easy is it to optimize with an AI-powered optimization tool?
Could I quickly set up the tests in Unbounce while still getting those other to-do’s done?
What kind of conversion lift would I see from just a few hours invested?
I took a little bit of my inspiration from Ms. Frizzle on the Magic School Bus. No, not her haircut, her catchphrase: “Take chances, make mistakes, get messy!”
Oh, so that’s where she got all her good ideas.
Creating 15 Variants in Under Two Hours
The beauty of Smart Traffic is there are no limits to how many variants you can create and it automatically starts optimizing in as few as 50 visits. Just hit the “optimize” button and you’re off to the races. Could it really be that simple?
My guinea pigs for this experiment would be two recent campaigns our marketing team had worked on: the ecommerce lookbook and the SaaS optimization guide. The team had created both of these ebook download pages in Unbounce, but we hadn’t been able to return to them and optimize very much in the months since we published.
The original landing pages would serve as my control variants. (Click to see the full pages.)
Before starting, I consulted with Anna Roginska, Growth Marketer at Unbounce, to get her input on how I should create my variants. She advised:
You can take the ‘spaghetti at the wall’ approach, where you create a bunch of variants and just leave them to Smart Traffic to see what happens. It’s that ‘set it and forget it’ mentality. That’s interesting, but when you look at a bowl of spaghetti… There’s a lot of noodles in there. You won’t necessarily get to explain why something is working or not working.
The other approach is to be more strategic and focused. I think there’s a huge benefit to going in with a plan. Create maybe only five variants and give them each a specific purpose. Then, you can see how they perform and create new iterations for different portions of the audience.
I had two landing pages to work with, so I thought I’d give both approaches a try. But with only a few hours scheduled in my calendar to complete all these variants, I needed to move fast.
The “Spaghetti at the Wall” Approach to Variant Creation
On the ecommerce lookbook page, I wanted to spend less time planning and more time creating. Whereas in A/B testing you need a proper test hypothesis and a careful plan for each variant, Smart Traffic lets you get creative and try out new ideas on the fly. Your variants don’t have to be perfect—they just need to be different enough to appeal to new audience segments.
This meant I didn’t have to make any hard or fast choices about which one element to “test” on the landing page. I could create 15 different variants that varied wildly from one another. Some used different colors, some had different headlines, some completely changed up the layout of the page.
This is something you just can’t do in a traditional A/B test where you’re looking to find a “winner” and understand why it “wins.” I had to remind myself I wasn’t looking for that one variant to rule them all (or for that one variant to bring them all and in the darkness bind them). I was looking to increase the chance of conversion for every single visitor. Certain pages were going to work better for certain audiences, and that was totally fine.
I wondered, though: how many variants would be too many? Would the machine learning recognize that some of these were not anything special and just stop sending traffic to them? And how long would it take to get results? With these questions in mind, I checked back on my first set of tests one month later…
Changing up the background color
Usually, color A/B tests are pretty much a waste of time. You need a lot of data to get accurate results, and most marketers don’t actually end up learning anything useful in the end. (Because color by itself means nothing, it always depends on the context of the page.)
That being said, we know there is some legitimate color theory and certain audience segments respond better to certain colors than others. So I thought it might be interesting to switch up the background on this landing page to see what would happen. And color me surprised—these variants are seeing some pretty dramatically different conversion rates:
Pink background – 12.82%
Green background – 21.43%
White background – 21.74%
Black background – 31.71%
One might start to speculate from these conversion rates that darker backgrounds perform better than the lighter backgrounds. But hold your horses, that’s thinking about this as an A/B test again. Here’s why Jordan Dawe, Senior Data Science Developer at Unbounce, says you should be cautious about drawing any conclusions from the conversion rates…
Smart Traffic is not sending visitors randomly—it’s trying to get the best traffic to the best variant. So in this case, it doesn’t mean that a black background will always convert higher than a pink background. There are likely portions of the audience going to each color that would be doing worse on others. Here’s what you can conclude: the color black is preferred by a portion of the traffic that converts highly.
It’s hard to shake that mindset of looking for a “winner” and trying to figure out “why” something is working. But I was starting to accept that different portions of the audience would always respond better to different variants—this was just the first time I’d been able to use AI to automatically serve up the best version.
Making big (and small) changes to the headline
For the next group of variants, I switched up the H1 in both small and big ways to see what effect that would have on the conversion rate. In some cases, this meant just swapping a single adjective (e.g., “jaw-dropping” for “drool-worthy”). In other cases, I went with a completely new line of copy altogether.
Here’s how the variants stacked up against each other:
See 27 Sales-Ready Ecommerce Landing Pages in Our Ultimate Lookbook – 25.81%
See 27 Stunning Ecommerce Landing Pages in Our Ultimate Lookbook – 25.93%
Get Ready to See 27 Jaw-Dropping Ecommerce Landing Page Examples – 28.13%
Get Serious Inspo for Supercharging Your Ecomm Sales – 35%
See 27 Drool-Worthy Ecommerce Landing Pages in Our Ultimate Lookbook – 40%
Again, each variant yielded a different conversion rate. I wondered if I kept testing different variations of the headlines and found one that performed best, could I deactivate all the other headline variants and just go with the “best” one? 
Here’s how Floss Taylor, Data Analyst at Unbounce, responded…
Smart Traffic doesn’t have champion variants. You don’t pick one at the end like you would in an A/B test. Although one variant may appear to be performing poorly, there could be a subset of traffic that it’s ideal for. You’re better off leaving it on long-term so it can work its magic.
Trying out different page layouts and hierarchies
The last set of variants I created messed with the actual structure and hierarchy of the page. I wanted to see if moving things around (or removing sections entirely) would influence the conversion rate. Here’s a sample of some of the experiments…
Removing the Headline – 16.67%
Adding a Double CTA – 21.95%
Moving the Testimonial Up the Page – 27.27%
Nothing too surprising here. And because I had created so many variants, Smart Traffic was taking longer than usual in “Learning Mode” to start giving me a conversion lift. Here’s how Floss Taylor explains it…
Smart Traffic needs approximately 50 visitors to understand which traffic would perform well for each new variant. If you have 15 variants and ~100 visitors per month, you’re going to have a long learning period where Smart Traffic cannot make accurate recommendations. I’d suggest starting off with a lower number of variants, and only adding more once once you have sufficient traffic.
The “Strategic Marketer” Approach
So throwing spaghetti at the wall turned out to be… messy. (New parents beware.) For the SaaS optimization guide page, I wanted to be a bit more strategic. And I actually had a leg up for this one, because Anna Roginska, Growth Marketer at Unbounce, had already started with a Smart Traffic experiment on this page four months ago.
Anna had set up a test between two different variants. One had an image of the ecommerce lookbook as the hero graphic on the page, while the other used the image of conversion expert and author Talia Wolf. Anna says she decided on this second variant because of research she had seen on how photographs of people tend to convert better than products.
I put Talia up front because I knew from other tests I’ve run and research I’ve done. [Photographs of] people tend to convert better. I didn’t know if it would work better in this particular case, but I was able to set up a variant and use Smart Traffic to find out. And it just so happens that the algorithm started sending way more traffic to this variant.
Anna seemed to be onto something, too: her variant was converting at nearly double the rate for a large traffic subset. And while I now know we can’t consider this a “champion” variant like in an A/B test and learn from the results, we could iterate based on her design to target new audience segments.
I created a simple spreadsheet to develop my gameplan. The goal was to create five new versions of the page that would appeal to different visitors based on their attributes:
Reducing the word count to target mobile and “ready to download” visitors
For inspiration on my first variant, I consulted the 2020 Conversion Benchmark Report. The machine learning insights here suggested that SaaS landing pages with lower word counts and easier-to-read copy tend to perform better than their long-winded counterparts.
And while the original version of our download page was easy enough to read, it did have a long, wordy intro with a lot of extra detail. Could I increase our conversion rate for a portion of our audience if just focused on the bare essentials? I was ready to kill some darlings to find out…
Original Long-Form Version – 10%
Low Word Count Version – 21.43%
It seems there’s a segment of our traffic coming to this page who didn’t need to see all that extra info before they decided to fill out the form. I speculated that this variant might also perform better on mobile devices since it would be faster-loading and easier to scroll through. Interesting!
Switching the headline to target different audience segments
Next, I created an additional four page variants to speak to the different pain points and reasons our audience might want to download the guide. (Actually, this is something Talia herself recommends you do in the SaaS optimization guide.) I switched up the headline copy here, as well as some of the supporting text underneath to match. After a month, here’s what the conversion rates look like:
Get Talia’s Guide to Optimize – 19.05%
You Can’t Just Build – 23.08%
Optimization is a Lot of Work – 24%
Not Sure How to Optimize? – 33.33%
Each variant is serving a different segment of the audience, by speaking to the particular reason they want to download the guide most (e.g., maybe they don’t have the time to optimize, or maybe they don’t know how to get started). As Smart Traffic learns more about which variants perform best for which audience segments, we become that much more likely to score a conversion.
What I Learned Running These Smart Traffic Experiments
Smart Traffic absolutely makes optimization easier and faster for marketers who previously never had the time (or experience) to run A/B tests. It took me under two hours to set up and launch these experiments, and we’re already seeing some pretty impressive results just over a month later.
While the ecommerce lookbook page is still optimizing, the SaaS ebook page is showing a 12% lift in conversions compared to evenly splitting traffic among all these variants. And this is after only a month—the algorithm will keep improving to get us even better results over time. (Like a fine wine, or that suspiciously old cheese in my fridge.)
At the same time, I did walk away with a few important lessons learned. If you’re planning to use Smart Traffic to optimize your landing pages, here are some things to keep in mind before you get started:
There are no champion variants – Unlike traditional A/B testing, you won’t be able to point to one landing page variant at the end of your test and call it a winner. The machine learning algorithm automatically routes audiences differently based on their individual attributes, which means you have to be cautious when you’re analyzing the results.
The more variants you create, the longer you’ll wait – While it can be tempting to throw spaghetti at the wall and create dozens of variants for your landing page, this means you’ll also have to wait longer to see what sticks. Try starting out with three to five variations and take a more strategic approach based on research in your industry. (The 2020 Conversion Benchmark Report is a great place to start for some ideas.)
It’s (usually) better to leave low-converting variants active – Because Smart Traffic learns over time and continually improves, you’re typically better off leaving your variants active—even if their conversion rates aren’t all that impressive. The AI takes the risk out of optimization by automatically sending visitors to the page that suits them best. If you turn off variants, you may lose out on some of those conversions altogether.
It can be a lot of fun to get creative with the different page elements and try out new ideas. You just might want to come up with a bit of a plan first and be strategic with your approach. Still, it’s better to experiment and optimize with Smart Traffic (even if you make some mistakes along the way) than to never optimize at all.
(And in case you were worried, yep—I managed to get my to-do list done, too. )
from Marketing https://unbounce.com/marketing-ai/smart-traffic-experiments/ via http://www.rssmix.com/
0 notes
kennethmontiveros · 4 years
Text
[Experiment] How AI is Changing the Way We Optimize at Unbounce
When I first encountered A/B testing, I immediately wanted to become the type of marketer who tested everything. The idea sounded fun to me. Like being a mad scientist running experiments to prove when my work was actually “working.”
Turns out though, there’s always a long list of other things to do first… blog posts to write, campaigns to launch, and don’t get me started on the meetings! I’m not alone in this, either. A lot of marketers are just too darned busy to follow up and optimize the stuff they’ve already shipped. According to HubSpot, only 17% of marketers use landing page A/B tests to improve conversion rates.
A small glimpse of my ever-growing to-do list. Ain’t nobody got time for A/B tests.
Sure, running a split test with one or two variants always sounds easy enough. But once you take a closer look at the process, you realize just how complex it can actually be. You need to make sure you have… 
The right duration and sample size.
Taken into account any external factors or validity threats.
Learned how to interpret the results correctly, too.
But—while there will always be a time and place for A/B testing—there’s also now an easier and faster way for marketers to optimize. Smart Traffic is a new Unbounce tool that uses the power of AI and machine learning to get you more conversions. Every day, more marketers are using Smart Traffic to “automagically” optimize their landing pages. But whenever we launch anything new, we like to test it out for ourselves to learn alongside you (and keep you up to speed on what to try next).
Here’s what I learned after taking Smart Traffic for a test drive myself…
Shifting Your Mindset to Optimize with AI
I know many marketers are (perhaps) skeptical when it comes to promises of machine learning, artificial intelligence, or magical “easy” buttons that get them better results. But AI is all around us and it’s already changing the way we do marketing. Landing page optimization is just one more area of the job where you no longer need to do everything yourself manually.
Smart Traffic augments your marketing skills and automatically sends visitors to the landing page variant where they’re most likely to convert (based on how similar page visitors have converted before). It makes routing decisions faster than any human ever could (thank you, AI magic), and “learns” which page variant is a perfect match for each different visitor. This ultimately means no more “champion” variants. Instead, you’re free to create multiple different pages to appeal to different groups of visitors and run ‘em all at once.
This is very different from A/B testing and honestly—it can feel kinda weird at first. You’ve got to trust in the machine learning to figure out what works best and what doesn’t. Data scientists call this the “black box” problem: data goes in, decisions come out, but you never really get the full understanding of what happened in between. 
Smart Traffic is fundamentally different from A/B Testing. You can learn more about how it works here.
For marketers using Smart Traffic, this means shifting your mindset and starting to think about optimization differently. Unlike A/B testing, you’re not looking for those “aha” moments to apply to your next campaign, or a one-size-fits-all “winning” variant. Instead, you’re looking to discover what works best for different subsets of your audience. This gives you unlimited creativity to try out new marketing ideas, makes it easier and less risky for you to optimize, and gives you an average conversion lift of 30% compared to splitting the traffic evenly across multiple variants. (Woah.)
My Experiment with Smart Traffic
I know all this because I recently experimented with variant creation myself to better understand this new AI optimization mindset. I created 15 variants across two separate landing pages using Smart Traffic to discover…
How easy is it to optimize with an AI-powered optimization tool?
Could I quickly set up the tests in Unbounce while still getting those other to-do’s done?
What kind of conversion lift would I see from just a few hours invested?
I took a little bit of my inspiration from Ms. Frizzle on the Magic School Bus. No, not her haircut, her catchphrase: “Take chances, make mistakes, get messy!”
Oh, so that’s where she got all her good ideas.
Creating 15 Variants in Under Two Hours
The beauty of Smart Traffic is there are no limits to how many variants you can create and it automatically starts optimizing in as few as 50 visits. Just hit the “optimize” button and you’re off to the races. Could it really be that simple?
My guinea pigs for this experiment would be two recent campaigns our marketing team had worked on: the ecommerce lookbook and the SaaS optimization guide. The team had created both of these ebook download pages in Unbounce, but we hadn’t been able to return to them and optimize very much in the months since we published.
The original landing pages would serve as my control variants. (Click to see the full pages.)
Before starting, I consulted with Anna Roginska, Growth Marketer at Unbounce, to get her input on how I should create my variants. She advised:
You can take the ‘spaghetti at the wall’ approach, where you create a bunch of variants and just leave them to Smart Traffic to see what happens. It’s that ‘set it and forget it’ mentality. That’s interesting, but when you look at a bowl of spaghetti… There’s a lot of noodles in there. You won’t necessarily get to explain why something is working or not working.
The other approach is to be more strategic and focused. I think there’s a huge benefit to going in with a plan. Create maybe only five variants and give them each a specific purpose. Then, you can see how they perform and create new iterations for different portions of the audience.
I had two landing pages to work with, so I thought I’d give both approaches a try. But with only a few hours scheduled in my calendar to complete all these variants, I needed to move fast.
The “Spaghetti at the Wall” Approach to Variant Creation
On the ecommerce lookbook page, I wanted to spend less time planning and more time creating. Whereas in A/B testing you need a proper test hypothesis and a careful plan for each variant, Smart Traffic lets you get creative and try out new ideas on the fly. Your variants don’t have to be perfect—they just need to be different enough to appeal to new audience segments.
This meant I didn’t have to make any hard or fast choices about which one element to “test” on the landing page. I could create 15 different variants that varied wildly from one another. Some used different colors, some had different headlines, some completely changed up the layout of the page.
This is something you just can’t do in a traditional A/B test where you’re looking to find a “winner” and understand why it “wins.” I had to remind myself I wasn’t looking for that one variant to rule them all (or for that one variant to bring them all and in the darkness bind them). I was looking to increase the chance of conversion for every single visitor. Certain pages were going to work better for certain audiences, and that was totally fine.
I wondered, though: how many variants would be too many? Would the machine learning recognize that some of these were not anything special and just stop sending traffic to them? And how long would it take to get results? With these questions in mind, I checked back on my first set of tests one month later…
Changing up the background color
Usually, color A/B tests are pretty much a waste of time. You need a lot of data to get accurate results, and most marketers don’t actually end up learning anything useful in the end. (Because color by itself means nothing, it always depends on the context of the page.)
That being said, we know there is some legitimate color theory and certain audience segments respond better to certain colors than others. So I thought it might be interesting to switch up the background on this landing page to see what would happen. And color me surprised—these variants are seeing some pretty dramatically different conversion rates:
Pink background – 12.82%
Green background – 21.43%
White background – 21.74%
Black background – 31.71%
One might start to speculate from these conversion rates that darker backgrounds perform better than the lighter backgrounds. But hold your horses, that’s thinking about this as an A/B test again. Here’s why Jordan Dawe, Senior Data Science Developer at Unbounce, says you should be cautious about drawing any conclusions from the conversion rates…
Smart Traffic is not sending visitors randomly—it’s trying to get the best traffic to the best variant. So in this case, it doesn’t mean that a black background will always convert higher than a pink background. There are likely portions of the audience going to each color that would be doing worse on others. Here’s what you can conclude: the color black is preferred by a portion of the traffic that converts highly.
It’s hard to shake that mindset of looking for a “winner” and trying to figure out “why” something is working. But I was starting to accept that different portions of the audience would always respond better to different variants—this was just the first time I’d been able to use AI to automatically serve up the best version.
Making big (and small) changes to the headline
For the next group of variants, I switched up the H1 in both small and big ways to see what effect that would have on the conversion rate. In some cases, this meant just swapping a single adjective (e.g., “jaw-dropping” for “drool-worthy”). In other cases, I went with a completely new line of copy altogether.
Here’s how the variants stacked up against each other:
See 27 Sales-Ready Ecommerce Landing Pages in Our Ultimate Lookbook – 25.81%
See 27 Stunning Ecommerce Landing Pages in Our Ultimate Lookbook – 25.93%
Get Ready to See 27 Jaw-Dropping Ecommerce Landing Page Examples – 28.13%
Get Serious Inspo for Supercharging Your Ecomm Sales – 35%
See 27 Drool-Worthy Ecommerce Landing Pages in Our Ultimate Lookbook – 40%
Again, each variant yielded a different conversion rate. I wondered if I kept testing different variations of the headlines and found one that performed best, could I deactivate all the other headline variants and just go with the “best” one? 
Here’s how Floss Taylor, Data Analyst at Unbounce, responded…
Smart Traffic doesn’t have champion variants. You don’t pick one at the end like you would in an A/B test. Although one variant may appear to be performing poorly, there could be a subset of traffic that it’s ideal for. You’re better off leaving it on long-term so it can work its magic.
Trying out different page layouts and hierarchies
The last set of variants I created messed with the actual structure and hierarchy of the page. I wanted to see if moving things around (or removing sections entirely) would influence the conversion rate. Here’s a sample of some of the experiments…
Removing the Headline – 16.67%
Adding a Double CTA – 21.95%
Moving the Testimonial Up the Page – 27.27%
Nothing too surprising here. And because I had created so many variants, Smart Traffic was taking longer than usual in “Learning Mode” to start giving me a conversion lift. Here’s how Floss Taylor explains it…
Smart Traffic needs approximately 50 visitors to understand which traffic would perform well for each new variant. If you have 15 variants and ~100 visitors per month, you’re going to have a long learning period where Smart Traffic cannot make accurate recommendations. I’d suggest starting off with a lower number of variants, and only adding more once once you have sufficient traffic.
The “Strategic Marketer” Approach
So throwing spaghetti at the wall turned out to be… messy. (New parents beware.) For the SaaS optimization guide page, I wanted to be a bit more strategic. And I actually had a leg up for this one, because Anna Roginska, Growth Marketer at Unbounce, had already started with a Smart Traffic experiment on this page four months ago.
Anna had set up a test between two different variants. One had an image of the ecommerce lookbook as the hero graphic on the page, while the other used the image of conversion expert and author Talia Wolf. Anna says she decided on this second variant because of research she had seen on how photographs of people tend to convert better than products.
I put Talia up front because I knew from other tests I’ve run and research I’ve done. [Photographs of] people tend to convert better. I didn’t know if it would work better in this particular case, but I was able to set up a variant and use Smart Traffic to find out. And it just so happens that the algorithm started sending way more traffic to this variant.
Anna seemed to be onto something, too: her variant was converting at nearly double the rate for a large traffic subset. And while I now know we can’t consider this a “champion” variant like in an A/B test and learn from the results, we could iterate based on her design to target new audience segments.
I created a simple spreadsheet to develop my gameplan. The goal was to create five new versions of the page that would appeal to different visitors based on their attributes:
Reducing the word count to target mobile and “ready to download” visitors
For inspiration on my first variant, I consulted the 2020 Conversion Benchmark Report. The machine learning insights here suggested that SaaS landing pages with lower word counts and easier-to-read copy tend to perform better than their long-winded counterparts.
And while the original version of our download page was easy enough to read, it did have a long, wordy intro with a lot of extra detail. Could I increase our conversion rate for a portion of our audience if just focused on the bare essentials? I was ready to kill some darlings to find out…
Original Long-Form Version – 10%
Low Word Count Version – 21.43%
It seems there’s a segment of our traffic coming to this page who didn’t need to see all that extra info before they decided to fill out the form. I speculated that this variant might also perform better on mobile devices since it would be faster-loading and easier to scroll through. Interesting!
Switching the headline to target different audience segments
Next, I created an additional four page variants to speak to the different pain points and reasons our audience might want to download the guide. (Actually, this is something Talia herself recommends you do in the SaaS optimization guide.) I switched up the headline copy here, as well as some of the supporting text underneath to match. After a month, here’s what the conversion rates look like:
Get Talia’s Guide to Optimize – 19.05%
You Can’t Just Build – 23.08%
Optimization is a Lot of Work – 24%
Not Sure How to Optimize? – 33.33%
Each variant is serving a different segment of the audience, by speaking to the particular reason they want to download the guide most (e.g., maybe they don’t have the time to optimize, or maybe they don’t know how to get started). As Smart Traffic learns more about which variants perform best for which audience segments, we become that much more likely to score a conversion.
What I Learned Running These Smart Traffic Experiments
Smart Traffic absolutely makes optimization easier and faster for marketers who previously never had the time (or experience) to run A/B tests. It took me under two hours to set up and launch these experiments, and we’re already seeing some pretty impressive results just over a month later.
While the ecommerce lookbook page is still optimizing, the SaaS ebook page is showing a 12% lift in conversions compared to evenly splitting traffic among all these variants. And this is after only a month—the algorithm will keep improving to get us even better results over time. (Like a fine wine, or that suspiciously old cheese in my fridge.)
At the same time, I did walk away with a few important lessons learned. If you’re planning to use Smart Traffic to optimize your landing pages, here are some things to keep in mind before you get started:
There are no champion variants – Unlike traditional A/B testing, you won’t be able to point to one landing page variant at the end of your test and call it a winner. The machine learning algorithm automatically routes audiences differently based on their individual attributes, which means you have to be cautious when you’re analyzing the results.
The more variants you create, the longer you’ll wait – While it can be tempting to throw spaghetti at the wall and create dozens of variants for your landing page, this means you’ll also have to wait longer to see what sticks. Try starting out with three to five variations and take a more strategic approach based on research in your industry. (The 2020 Conversion Benchmark Report is a great place to start for some ideas.)
It’s (usually) better to leave low-converting variants active – Because Smart Traffic learns over time and continually improves, you’re typically better off leaving your variants active—even if their conversion rates aren’t all that impressive. The AI takes the risk out of optimization by automatically sending visitors to the page that suits them best. If you turn off variants, you may lose out on some of those conversions altogether.
It can be a lot of fun to get creative with the different page elements and try out new ideas. You just might want to come up with a bit of a plan first and be strategic with your approach. Still, it’s better to experiment and optimize with Smart Traffic (even if you make some mistakes along the way) than to never optimize at all.
(And in case you were worried, yep—I managed to get my to-do list done, too. )
[Experiment] How AI is Changing the Way We Optimize at Unbounce published first on http://nickpontemktg.blogspot.com/
0 notes
josephkchoi · 4 years
Text
[Experiment] How AI is Changing the Way We Optimize at Unbounce
When I first encountered A/B testing, I immediately wanted to become the type of marketer who tested everything. The idea sounded fun to me. Like being a mad scientist running experiments to prove when my work was actually “working.”
Turns out though, there’s always a long list of other things to do first… blog posts to write, campaigns to launch, and don’t get me started on the meetings! I’m not alone in this, either. A lot of marketers are just too darned busy to follow up and optimize the stuff they’ve already shipped. According to HubSpot, only 17% of marketers use landing page A/B tests to improve conversion rates.
A small glimpse of my ever-growing to-do list. Ain’t nobody got time for A/B tests.
Sure, running a split test with one or two variants always sounds easy enough. But once you take a closer look at the process, you realize just how complex it can actually be. You need to make sure you have… 
The right duration and sample size.
Taken into account any external factors or validity threats.
Learned how to interpret the results correctly, too.
But—while there will always be a time and place for A/B testing—there’s also now an easier and faster way for marketers to optimize. Smart Traffic is a new Unbounce tool that uses the power of AI and machine learning to get you more conversions. Every day, more marketers are using Smart Traffic to “automagically” optimize their landing pages. But whenever we launch anything new, we like to test it out for ourselves to learn alongside you (and keep you up to speed on what to try next).
Here’s what I learned after taking Smart Traffic for a test drive myself…
Shifting Your Mindset to Optimize with AI
I know many marketers are (perhaps) skeptical when it comes to promises of machine learning, artificial intelligence, or magical “easy” buttons that get them better results. But AI is all around us and it’s already changing the way we do marketing. Landing page optimization is just one more area of the job where you no longer need to do everything yourself manually.
Smart Traffic augments your marketing skills and automatically sends visitors to the landing page variant where they’re most likely to convert (based on how similar page visitors have converted before). It makes routing decisions faster than any human ever could (thank you, AI magic), and “learns” which page variant is a perfect match for each different visitor. This ultimately means no more “champion” variants. Instead, you’re free to create multiple different pages to appeal to different groups of visitors and run ‘em all at once.
This is very different from A/B testing and honestly—it can feel kinda weird at first. You’ve got to trust in the machine learning to figure out what works best and what doesn’t. Data scientists call this the “black box” problem: data goes in, decisions come out, but you never really get the full understanding of what happened in between. 
Smart Traffic is fundamentally different from A/B Testing. You can learn more about how it works here.
For marketers using Smart Traffic, this means shifting your mindset and starting to think about optimization differently. Unlike A/B testing, you’re not looking for those “aha” moments to apply to your next campaign, or a one-size-fits-all “winning” variant. Instead, you’re looking to discover what works best for different subsets of your audience. This gives you unlimited creativity to try out new marketing ideas, makes it easier and less risky for you to optimize, and gives you an average conversion lift of 30% compared to splitting the traffic evenly across multiple variants. (Woah.)
My Experiment with Smart Traffic
I know all this because I recently experimented with variant creation myself to better understand this new AI optimization mindset. I created 15 variants across two separate landing pages using Smart Traffic to discover…
How easy is it to optimize with an AI-powered optimization tool?
Could I quickly set up the tests in Unbounce while still getting those other to-do’s done?
What kind of conversion lift would I see from just a few hours invested?
I took a little bit of my inspiration from Ms. Frizzle on the Magic School Bus. No, not her haircut, her catchphrase: “Take chances, make mistakes, get messy!”
Oh, so that’s where she got all her good ideas.
Creating 15 Variants in Under Two Hours
The beauty of Smart Traffic is there are no limits to how many variants you can create and it automatically starts optimizing in as few as 50 visits. Just hit the “optimize” button and you’re off to the races. Could it really be that simple?
My guinea pigs for this experiment would be two recent campaigns our marketing team had worked on: the ecommerce lookbook and the SaaS optimization guide. The team had created both of these ebook download pages in Unbounce, but we hadn’t been able to return to them and optimize very much in the months since we published.
The original landing pages would serve as my control variants. (Click to see the full pages.)
Before starting, I consulted with Anna Roginska, Growth Marketer at Unbounce, to get her input on how I should create my variants. She advised:
You can take the ‘spaghetti at the wall’ approach, where you create a bunch of variants and just leave them to Smart Traffic to see what happens. It’s that ‘set it and forget it’ mentality. That’s interesting, but when you look at a bowl of spaghetti… There’s a lot of noodles in there. You won’t necessarily get to explain why something is working or not working.
The other approach is to be more strategic and focused. I think there’s a huge benefit to going in with a plan. Create maybe only five variants and give them each a specific purpose. Then, you can see how they perform and create new iterations for different portions of the audience.
I had two landing pages to work with, so I thought I’d give both approaches a try. But with only a few hours scheduled in my calendar to complete all these variants, I needed to move fast.
The “Spaghetti at the Wall” Approach to Variant Creation
On the ecommerce lookbook page, I wanted to spend less time planning and more time creating. Whereas in A/B testing you need a proper test hypothesis and a careful plan for each variant, Smart Traffic lets you get creative and try out new ideas on the fly. Your variants don’t have to be perfect—they just need to be different enough to appeal to new audience segments.
This meant I didn’t have to make any hard or fast choices about which one element to “test” on the landing page. I could create 15 different variants that varied wildly from one another. Some used different colors, some had different headlines, some completely changed up the layout of the page.
This is something you just can’t do in a traditional A/B test where you’re looking to find a “winner” and understand why it “wins.” I had to remind myself I wasn’t looking for that one variant to rule them all (or for that one variant to bring them all and in the darkness bind them). I was looking to increase the chance of conversion for every single visitor. Certain pages were going to work better for certain audiences, and that was totally fine.
I wondered, though: how many variants would be too many? Would the machine learning recognize that some of these were not anything special and just stop sending traffic to them? And how long would it take to get results? With these questions in mind, I checked back on my first set of tests one month later…
Changing up the background color
Usually, color A/B tests are pretty much a waste of time. You need a lot of data to get accurate results, and most marketers don’t actually end up learning anything useful in the end. (Because color by itself means nothing, it always depends on the context of the page.)
That being said, we know there is some legitimate color theory and certain audience segments respond better to certain colors than others. So I thought it might be interesting to switch up the background on this landing page to see what would happen. And color me surprised—these variants are seeing some pretty dramatically different conversion rates:
Pink background – 12.82%
Green background – 21.43%
White background – 21.74%
Black background – 31.71%
One might start to speculate from these conversion rates that darker backgrounds perform better than the lighter backgrounds. But hold your horses, that’s thinking about this as an A/B test again. Here’s why Jordan Dawe, Senior Data Science Developer at Unbounce, says you should be cautious about drawing any conclusions from the conversion rates…
Smart Traffic is not sending visitors randomly—it’s trying to get the best traffic to the best variant. So in this case, it doesn’t mean that a black background will always convert higher than a pink background. There are likely portions of the audience going to each color that would be doing worse on others. Here’s what you can conclude: the color black is preferred by a portion of the traffic that converts highly.
It’s hard to shake that mindset of looking for a “winner” and trying to figure out “why” something is working. But I was starting to accept that different portions of the audience would always respond better to different variants—this was just the first time I’d been able to use AI to automatically serve up the best version.
Making big (and small) changes to the headline
For the next group of variants, I switched up the H1 in both small and big ways to see what effect that would have on the conversion rate. In some cases, this meant just swapping a single adjective (e.g., “jaw-dropping” for “drool-worthy”). In other cases, I went with a completely new line of copy altogether.
Here’s how the variants stacked up against each other:
See 27 Sales-Ready Ecommerce Landing Pages in Our Ultimate Lookbook – 25.81%
See 27 Stunning Ecommerce Landing Pages in Our Ultimate Lookbook – 25.93%
Get Ready to See 27 Jaw-Dropping Ecommerce Landing Page Examples – 28.13%
Get Serious Inspo for Supercharging Your Ecomm Sales – 35%
See 27 Drool-Worthy Ecommerce Landing Pages in Our Ultimate Lookbook – 40%
Again, each variant yielded a different conversion rate. I wondered if I kept testing different variations of the headlines and found one that performed best, could I deactivate all the other headline variants and just go with the “best” one? 
Here’s how Floss Taylor, Data Analyst at Unbounce, responded…
Smart Traffic doesn’t have champion variants. You don’t pick one at the end like you would in an A/B test. Although one variant may appear to be performing poorly, there could be a subset of traffic that it’s ideal for. You’re better off leaving it on long-term so it can work its magic.
Trying out different page layouts and hierarchies
The last set of variants I created messed with the actual structure and hierarchy of the page. I wanted to see if moving things around (or removing sections entirely) would influence the conversion rate. Here’s a sample of some of the experiments…
Removing the Headline – 16.67%
Adding a Double CTA – 21.95%
Moving the Testimonial Up the Page – 27.27%
Nothing too surprising here. And because I had created so many variants, Smart Traffic was taking longer than usual in “Learning Mode” to start giving me a conversion lift. Here’s how Floss Taylor explains it…
Smart Traffic needs approximately 50 visitors to understand which traffic would perform well for each new variant. If you have 15 variants and ~100 visitors per month, you’re going to have a long learning period where Smart Traffic cannot make accurate recommendations. I’d suggest starting off with a lower number of variants, and only adding more once once you have sufficient traffic.
The “Strategic Marketer” Approach
So throwing spaghetti at the wall turned out to be… messy. (New parents beware.) For the SaaS optimization guide page, I wanted to be a bit more strategic. And I actually had a leg up for this one, because Anna Roginska, Growth Marketer at Unbounce, had already started with a Smart Traffic experiment on this page four months ago.
Anna had set up a test between two different variants. One had an image of the ecommerce lookbook as the hero graphic on the page, while the other used the image of conversion expert and author Talia Wolf. Anna says she decided on this second variant because of research she had seen on how photographs of people tend to convert better than products.
I put Talia up front because I knew from other tests I’ve run and research I’ve done. [Photographs of] people tend to convert better. I didn’t know if it would work better in this particular case, but I was able to set up a variant and use Smart Traffic to find out. And it just so happens that the algorithm started sending way more traffic to this variant.
Anna seemed to be onto something, too: her variant was converting at nearly double the rate for a large traffic subset. And while I now know we can’t consider this a “champion” variant like in an A/B test and learn from the results, we could iterate based on her design to target new audience segments.
I created a simple spreadsheet to develop my gameplan. The goal was to create five new versions of the page that would appeal to different visitors based on their attributes:
Reducing the word count to target mobile and “ready to download” visitors
For inspiration on my first variant, I consulted the 2020 Conversion Benchmark Report. The machine learning insights here suggested that SaaS landing pages with lower word counts and easier-to-read copy tend to perform better than their long-winded counterparts.
And while the original version of our download page was easy enough to read, it did have a long, wordy intro with a lot of extra detail. Could I increase our conversion rate for a portion of our audience if just focused on the bare essentials? I was ready to kill some darlings to find out…
Original Long-Form Version – 10%
Low Word Count Version – 21.43%
It seems there’s a segment of our traffic coming to this page who didn’t need to see all that extra info before they decided to fill out the form. I speculated that this variant might also perform better on mobile devices since it would be faster-loading and easier to scroll through. Interesting!
Switching the headline to target different audience segments
Next, I created an additional four page variants to speak to the different pain points and reasons our audience might want to download the guide. (Actually, this is something Talia herself recommends you do in the SaaS optimization guide.) I switched up the headline copy here, as well as some of the supporting text underneath to match. After a month, here’s what the conversion rates look like:
Get Talia’s Guide to Optimize – 19.05%
You Can’t Just Build – 23.08%
Optimization is a Lot of Work – 24%
Not Sure How to Optimize? – 33.33%
Each variant is serving a different segment of the audience, by speaking to the particular reason they want to download the guide most (e.g., maybe they don’t have the time to optimize, or maybe they don’t know how to get started). As Smart Traffic learns more about which variants perform best for which audience segments, we become that much more likely to score a conversion.
What I Learned Running These Smart Traffic Experiments
Smart Traffic absolutely makes optimization easier and faster for marketers who previously never had the time (or experience) to run A/B tests. It took me under two hours to set up and launch these experiments, and we’re already seeing some pretty impressive results just over a month later.
While the ecommerce lookbook page is still optimizing, the SaaS ebook page is showing a 12% lift in conversions compared to evenly splitting traffic among all these variants. And this is after only a month—the algorithm will keep improving to get us even better results over time. (Like a fine wine, or that suspiciously old cheese in my fridge.)
At the same time, I did walk away with a few important lessons learned. If you’re planning to use Smart Traffic to optimize your landing pages, here are some things to keep in mind before you get started:
There are no champion variants – Unlike traditional A/B testing, you won’t be able to point to one landing page variant at the end of your test and call it a winner. The machine learning algorithm automatically routes audiences differently based on their individual attributes, which means you have to be cautious when you’re analyzing the results.
The more variants you create, the longer you’ll wait – While it can be tempting to throw spaghetti at the wall and create dozens of variants for your landing page, this means you’ll also have to wait longer to see what sticks. Try starting out with three to five variations and take a more strategic approach based on research in your industry. (The 2020 Conversion Benchmark Report is a great place to start for some ideas.)
It’s (usually) better to leave low-converting variants active – Because Smart Traffic learns over time and continually improves, you’re typically better off leaving your variants active—even if their conversion rates aren’t all that impressive. The AI takes the risk out of optimization by automatically sending visitors to the page that suits them best. If you turn off variants, you may lose out on some of those conversions altogether.
It can be a lot of fun to get creative with the different page elements and try out new ideas. You just might want to come up with a bit of a plan first and be strategic with your approach. Still, it’s better to experiment and optimize with Smart Traffic (even if you make some mistakes along the way) than to never optimize at all.
(And in case you were worried, yep—I managed to get my to-do list done, too. )
[Experiment] How AI is Changing the Way We Optimize at Unbounce published first on https://nickpontemrktg.wordpress.com/
0 notes
reviewandbonuss · 4 years
Text
[Experiment] How AI is Changing the Way We Optimize at Unbounce
When I first encountered A/B testing, I immediately wanted to become the type of marketer who tested everything. The idea sounded fun to me. Like being a mad scientist running experiments to prove when my work was actually “working.”
Turns out though, there’s always a long list of other things to do first… blog posts to write, campaigns to launch, and don’t get me started on the meetings! I’m not alone in this, either. A lot of marketers are just too darned busy to follow up and optimize the stuff they’ve already shipped. According to HubSpot, only 17% of marketers use landing page A/B tests to improve conversion rates.
A small glimpse of my ever-growing to-do list. Ain’t nobody got time for A/B tests.
Sure, running a split test with one or two variants always sounds easy enough. But once you take a closer look at the process, you realize just how complex it can actually be. You need to make sure you have… 
The right duration and sample size.
Taken into account any external factors or validity threats.
Learned how to interpret the results correctly, too.
But—while there will always be a time and place for A/B testing—there’s also now an easier and faster way for marketers to optimize. Smart Traffic is a new Unbounce tool that uses the power of AI and machine learning to get you more conversions. Every day, more marketers are using Smart Traffic to “automagically” optimize their landing pages. But whenever we launch anything new, we like to test it out for ourselves to learn alongside you (and keep you up to speed on what to try next).
Here’s what I learned after taking Smart Traffic for a test drive myself…
Shifting Your Mindset to Optimize with AI
I know many marketers are (perhaps) skeptical when it comes to promises of machine learning, artificial intelligence, or magical “easy” buttons that get them better results. But AI is all around us and it’s already changing the way we do marketing. Landing page optimization is just one more area of the job where you no longer need to do everything yourself manually.
Smart Traffic augments your marketing skills and automatically sends visitors to the landing page variant where they’re most likely to convert (based on how similar page visitors have converted before). It makes routing decisions faster than any human ever could (thank you, AI magic), and “learns” which page variant is a perfect match for each different visitor. This ultimately means no more “champion” variants. Instead, you’re free to create multiple different pages to appeal to different groups of visitors and run ‘em all at once.
This is very different from A/B testing and honestly—it can feel kinda weird at first. You’ve got to trust in the machine learning to figure out what works best and what doesn’t. Data scientists call this the “black box” problem: data goes in, decisions come out, but you never really get the full understanding of what happened in between. 
Smart Traffic is fundamentally different from A/B Testing. You can learn more about how it works here.
For marketers using Smart Traffic, this means shifting your mindset and starting to think about optimization differently. Unlike A/B testing, you’re not looking for those “aha” moments to apply to your next campaign, or a one-size-fits-all “winning” variant. Instead, you’re looking to discover what works best for different subsets of your audience. This gives you unlimited creativity to try out new marketing ideas, makes it easier and less risky for you to optimize, and gives you an average conversion lift of 30% compared to splitting the traffic evenly across multiple variants. (Woah.)
My Experiment with Smart Traffic
I know all this because I recently experimented with variant creation myself to better understand this new AI optimization mindset. I created 15 variants across two separate landing pages using Smart Traffic to discover…
How easy is it to optimize with an AI-powered optimization tool?
Could I quickly set up the tests in Unbounce while still getting those other to-do’s done?
What kind of conversion lift would I see from just a few hours invested?
I took a little bit of my inspiration from Ms. Frizzle on the Magic School Bus. No, not her haircut, her catchphrase: “Take chances, make mistakes, get messy!”
Oh, so that’s where she got all her good ideas.
Creating 15 Variants in Under Two Hours
The beauty of Smart Traffic is there are no limits to how many variants you can create and it automatically starts optimizing in as few as 50 visits. Just hit the “optimize” button and you’re off to the races. Could it really be that simple?
My guinea pigs for this experiment would be two recent campaigns our marketing team had worked on: the ecommerce lookbook and the SaaS optimization guide. The team had created both of these ebook download pages in Unbounce, but we hadn’t been able to return to them and optimize very much in the months since we published.
The original landing pages would serve as my control variants. (Click to see the full pages.)
Before starting, I consulted with Anna Roginska, Growth Marketer at Unbounce, to get her input on how I should create my variants. She advised:
You can take the ‘spaghetti at the wall’ approach, where you create a bunch of variants and just leave them to Smart Traffic to see what happens. It’s that ‘set it and forget it’ mentality. That’s interesting, but when you look at a bowl of spaghetti… There’s a lot of noodles in there. You won’t necessarily get to explain why something is working or not working.
The other approach is to be more strategic and focused. I think there’s a huge benefit to going in with a plan. Create maybe only five variants and give them each a specific purpose. Then, you can see how they perform and create new iterations for different portions of the audience.
I had two landing pages to work with, so I thought I’d give both approaches a try. But with only a few hours scheduled in my calendar to complete all these variants, I needed to move fast.
The “Spaghetti at the Wall” Approach to Variant Creation
On the ecommerce lookbook page, I wanted to spend less time planning and more time creating. Whereas in A/B testing you need a proper test hypothesis and a careful plan for each variant, Smart Traffic lets you get creative and try out new ideas on the fly. Your variants don’t have to be perfect—they just need to be different enough to appeal to new audience segments.
This meant I didn’t have to make any hard or fast choices about which one element to “test” on the landing page. I could create 15 different variants that varied wildly from one another. Some used different colors, some had different headlines, some completely changed up the layout of the page.
This is something you just can’t do in a traditional A/B test where you’re looking to find a “winner” and understand why it “wins.” I had to remind myself I wasn’t looking for that one variant to rule them all (or for that one variant to bring them all and in the darkness bind them). I was looking to increase the chance of conversion for every single visitor. Certain pages were going to work better for certain audiences, and that was totally fine.
I wondered, though: how many variants would be too many? Would the machine learning recognize that some of these were not anything special and just stop sending traffic to them? And how long would it take to get results? With these questions in mind, I checked back on my first set of tests one month later…
Changing up the background color
Usually, color A/B tests are pretty much a waste of time. You need a lot of data to get accurate results, and most marketers don’t actually end up learning anything useful in the end. (Because color by itself means nothing, it always depends on the context of the page.)
That being said, we know there is some legitimate color theory and certain audience segments respond better to certain colors than others. So I thought it might be interesting to switch up the background on this landing page to see what would happen. And color me surprised—these variants are seeing some pretty dramatically different conversion rates:
Pink background – 12.82%
Green background – 21.43%
White background – 21.74%
Black background – 31.71%
One might start to speculate from these conversion rates that darker backgrounds perform better than the lighter backgrounds. But hold your horses, that’s thinking about this as an A/B test again. Here’s why Jordan Dawe, Senior Data Science Developer at Unbounce, says you should be cautious about drawing any conclusions from the conversion rates…
Smart Traffic is not sending visitors randomly—it’s trying to get the best traffic to the best variant. So in this case, it doesn’t mean that a black background will always convert higher than a pink background. There are likely portions of the audience going to each color that would be doing worse on others. Here’s what you can conclude: the color black is preferred by a portion of the traffic that converts highly.
It’s hard to shake that mindset of looking for a “winner” and trying to figure out “why” something is working. But I was starting to accept that different portions of the audience would always respond better to different variants—this was just the first time I’d been able to use AI to automatically serve up the best version.
Making big (and small) changes to the headline
For the next group of variants, I switched up the H1 in both small and big ways to see what effect that would have on the conversion rate. In some cases, this meant just swapping a single adjective (e.g., “jaw-dropping” for “drool-worthy”). In other cases, I went with a completely new line of copy altogether.
Here’s how the variants stacked up against each other:
See 27 Sales-Ready Ecommerce Landing Pages in Our Ultimate Lookbook – 25.81%
See 27 Stunning Ecommerce Landing Pages in Our Ultimate Lookbook – 25.93%
Get Ready to See 27 Jaw-Dropping Ecommerce Landing Page Examples – 28.13%
Get Serious Inspo for Supercharging Your Ecomm Sales – 35%
See 27 Drool-Worthy Ecommerce Landing Pages in Our Ultimate Lookbook – 40%
Again, each variant yielded a different conversion rate. I wondered if I kept testing different variations of the headlines and found one that performed best, could I deactivate all the other headline variants and just go with the “best” one? 
Here’s how Floss Taylor, Data Analyst at Unbounce, responded…
Smart Traffic doesn’t have champion variants. You don’t pick one at the end like you would in an A/B test. Although one variant may appear to be performing poorly, there could be a subset of traffic that it’s ideal for. You’re better off leaving it on long-term so it can work its magic.
Trying out different page layouts and hierarchies
The last set of variants I created messed with the actual structure and hierarchy of the page. I wanted to see if moving things around (or removing sections entirely) would influence the conversion rate. Here’s a sample of some of the experiments…
Removing the Headline – 16.67%
Adding a Double CTA – 21.95%
Moving the Testimonial Up the Page – 27.27%
Nothing too surprising here. And because I had created so many variants, Smart Traffic was taking longer than usual in “Learning Mode” to start giving me a conversion lift. Here’s how Floss Taylor explains it…
Smart Traffic needs approximately 50 visitors to understand which traffic would perform well for each new variant. If you have 15 variants and ~100 visitors per month, you’re going to have a long learning period where Smart Traffic cannot make accurate recommendations. I’d suggest starting off with a lower number of variants, and only adding more once once you have sufficient traffic.
The “Strategic Marketer” Approach
So throwing spaghetti at the wall turned out to be… messy. (New parents beware.) For the SaaS optimization guide page, I wanted to be a bit more strategic. And I actually had a leg up for this one, because Anna Roginska, Growth Marketer at Unbounce, had already started with a Smart Traffic experiment on this page four months ago.
Anna had set up a test between two different variants. One had an image of the ecommerce lookbook as the hero graphic on the page, while the other used the image of conversion expert and author Talia Wolf. Anna says she decided on this second variant because of research she had seen on how photographs of people tend to convert better than products.
I put Talia up front because I knew from other tests I’ve run and research I’ve done. [Photographs of] people tend to convert better. I didn’t know if it would work better in this particular case, but I was able to set up a variant and use Smart Traffic to find out. And it just so happens that the algorithm started sending way more traffic to this variant.
Anna seemed to be onto something, too: her variant was converting at nearly double the rate for a large traffic subset. And while I now know we can’t consider this a “champion” variant like in an A/B test and learn from the results, we could iterate based on her design to target new audience segments.
I created a simple spreadsheet to develop my gameplan. The goal was to create five new versions of the page that would appeal to different visitors based on their attributes:
Reducing the word count to target mobile and “ready to download” visitors
For inspiration on my first variant, I consulted the 2020 Conversion Benchmark Report. The machine learning insights here suggested that SaaS landing pages with lower word counts and easier-to-read copy tend to perform better than their long-winded counterparts.
And while the original version of our download page was easy enough to read, it did have a long, wordy intro with a lot of extra detail. Could I increase our conversion rate for a portion of our audience if just focused on the bare essentials? I was ready to kill some darlings to find out…
Original Long-Form Version – 10%
Low Word Count Version – 21.43%
It seems there’s a segment of our traffic coming to this page who didn’t need to see all that extra info before they decided to fill out the form. I speculated that this variant might also perform better on mobile devices since it would be faster-loading and easier to scroll through. Interesting!
Switching the headline to target different audience segments
Next, I created an additional four page variants to speak to the different pain points and reasons our audience might want to download the guide. (Actually, this is something Talia herself recommends you do in the SaaS optimization guide.) I switched up the headline copy here, as well as some of the supporting text underneath to match. After a month, here’s what the conversion rates look like:
Get Talia’s Guide to Optimize – 19.05%
You Can’t Just Build – 23.08%
Optimization is a Lot of Work – 24%
Not Sure How to Optimize? – 33.33%
Each variant is serving a different segment of the audience, by speaking to the particular reason they want to download the guide most (e.g., maybe they don’t have the time to optimize, or maybe they don’t know how to get started). As Smart Traffic learns more about which variants perform best for which audience segments, we become that much more likely to score a conversion.
What I Learned Running These Smart Traffic Experiments
Smart Traffic absolutely makes optimization easier and faster for marketers who previously never had the time (or experience) to run A/B tests. It took me under two hours to set up and launch these experiments, and we’re already seeing some pretty impressive results just over a month later.
While the ecommerce lookbook page is still optimizing, the SaaS ebook page is showing a 12% lift in conversions compared to evenly splitting traffic among all these variants. And this is after only a month—the algorithm will keep improving to get us even better results over time. (Like a fine wine, or that suspiciously old cheese in my fridge.)
At the same time, I did walk away with a few important lessons learned. If you’re planning to use Smart Traffic to optimize your landing pages, here are some things to keep in mind before you get started:
There are no champion variants – Unlike traditional A/B testing, you won’t be able to point to one landing page variant at the end of your test and call it a winner. The machine learning algorithm automatically routes audiences differently based on their individual attributes, which means you have to be cautious when you’re analyzing the results.
The more variants you create, the longer you’ll wait – While it can be tempting to throw spaghetti at the wall and create dozens of variants for your landing page, this means you’ll also have to wait longer to see what sticks. Try starting out with three to five variations and take a more strategic approach based on research in your industry. (The 2020 Conversion Benchmark Report is a great place to start for some ideas.)
It’s (usually) better to leave low-converting variants active – Because Smart Traffic learns over time and continually improves, you’re typically better off leaving your variants active—even if their conversion rates aren’t all that impressive. The AI takes the risk out of optimization by automatically sending visitors to the page that suits them best. If you turn off variants, you may lose out on some of those conversions altogether.
It can be a lot of fun to get creative with the different page elements and try out new ideas. You just might want to come up with a bit of a plan first and be strategic with your approach. Still, it’s better to experiment and optimize with Smart Traffic (even if you make some mistakes along the way) than to never optimize at all.
(And in case you were worried, yep—I managed to get my to-do list done, too. )
https://unbounce.com/marketing-ai/smart-traffic-experiments/
0 notes
itsjessicaisreal · 4 years
Text
[Experiment] How AI is Changing the Way We Optimize at Unbounce
When I first encountered A/B testing, I immediately wanted to become the type of marketer who tested everything. The idea sounded fun to me. Like being a mad scientist running experiments to prove when my work was actually “working.”
Turns out though, there’s always a long list of other things to do first… blog posts to write, campaigns to launch, and don’t get me started on the meetings! I’m not alone in this, either. A lot of marketers are just too darned busy to follow up and optimize the stuff they’ve already shipped. According to HubSpot, only 17% of marketers use landing page A/B tests to improve conversion rates.
A small glimpse of my ever-growing to-do list. Ain’t nobody got time for A/B tests.
Sure, running a split test with one or two variants always sounds easy enough. But once you take a closer look at the process, you realize just how complex it can actually be. You need to make sure you have… 
The right duration and sample size.
Taken into account any external factors or validity threats.
Learned how to interpret the results correctly, too.
But—while there will always be a time and place for A/B testing—there’s also now an easier and faster way for marketers to optimize. Smart Traffic is a new Unbounce tool that uses the power of AI and machine learning to get you more conversions. Every day, more marketers are using Smart Traffic to “automagically” optimize their landing pages. But whenever we launch anything new, we like to test it out for ourselves to learn alongside you (and keep you up to speed on what to try next).
Here’s what I learned after taking Smart Traffic for a test drive myself…
Shifting Your Mindset to Optimize with AI
I know many marketers are (perhaps) skeptical when it comes to promises of machine learning, artificial intelligence, or magical “easy” buttons that get them better results. But AI is all around us and it’s already changing the way we do marketing. Landing page optimization is just one more area of the job where you no longer need to do everything yourself manually.
Smart Traffic augments your marketing skills and automatically sends visitors to the landing page variant where they’re most likely to convert (based on how similar page visitors have converted before). It makes routing decisions faster than any human ever could (thank you, AI magic), and “learns” which page variant is a perfect match for each different visitor. This ultimately means no more “champion” variants. Instead, you’re free to create multiple different pages to appeal to different groups of visitors and run ‘em all at once.
This is very different from A/B testing and honestly—it can feel kinda weird at first. You’ve got to trust in the machine learning to figure out what works best and what doesn’t. Data scientists call this the “black box” problem: data goes in, decisions come out, but you never really get the full understanding of what happened in between. 
Smart Traffic is fundamentally different from A/B Testing. You can learn more about how it works here.
For marketers using Smart Traffic, this means shifting your mindset and starting to think about optimization differently. Unlike A/B testing, you’re not looking for those “aha” moments to apply to your next campaign, or a one-size-fits-all “winning” variant. Instead, you’re looking to discover what works best for different subsets of your audience. This gives you unlimited creativity to try out new marketing ideas, makes it easier and less risky for you to optimize, and gives you an average conversion lift of 30% compared to splitting the traffic evenly across multiple variants. (Woah.)
My Experiment with Smart Traffic
I know all this because I recently experimented with variant creation myself to better understand this new AI optimization mindset. I created 15 variants across two separate landing pages using Smart Traffic to discover…
How easy is it to optimize with an AI-powered optimization tool?
Could I quickly set up the tests in Unbounce while still getting those other to-do’s done?
What kind of conversion lift would I see from just a few hours invested?
I took a little bit of my inspiration from Ms. Frizzle on the Magic School Bus. No, not her haircut, her catchphrase: “Take chances, make mistakes, get messy!”
Oh, so that’s where she got all her good ideas.
Creating 15 Variants in Under Two Hours
The beauty of Smart Traffic is there are no limits to how many variants you can create and it automatically starts optimizing in as few as 50 visits. Just hit the “optimize” button and you’re off to the races. Could it really be that simple?
My guinea pigs for this experiment would be two recent campaigns our marketing team had worked on: the ecommerce lookbook and the SaaS optimization guide. The team had created both of these ebook download pages in Unbounce, but we hadn’t been able to return to them and optimize very much in the months since we published.
The original landing pages would serve as my control variants. (Click to see the full pages.)
Before starting, I consulted with Anna Roginska, Growth Marketer at Unbounce, to get her input on how I should create my variants. She advised:
You can take the ‘spaghetti at the wall’ approach, where you create a bunch of variants and just leave them to Smart Traffic to see what happens. It’s that ‘set it and forget it’ mentality. That’s interesting, but when you look at a bowl of spaghetti… There’s a lot of noodles in there. You won’t necessarily get to explain why something is working or not working.
The other approach is to be more strategic and focused. I think there’s a huge benefit to going in with a plan. Create maybe only five variants and give them each a specific purpose. Then, you can see how they perform and create new iterations for different portions of the audience.
I had two landing pages to work with, so I thought I’d give both approaches a try. But with only a few hours scheduled in my calendar to complete all these variants, I needed to move fast.
The “Spaghetti at the Wall” Approach to Variant Creation
On the ecommerce lookbook page, I wanted to spend less time planning and more time creating. Whereas in A/B testing you need a proper test hypothesis and a careful plan for each variant, Smart Traffic lets you get creative and try out new ideas on the fly. Your variants don’t have to be perfect—they just need to be different enough to appeal to new audience segments.
This meant I didn’t have to make any hard or fast choices about which one element to “test” on the landing page. I could create 15 different variants that varied wildly from one another. Some used different colors, some had different headlines, some completely changed up the layout of the page.
This is something you just can’t do in a traditional A/B test where you’re looking to find a “winner” and understand why it “wins.” I had to remind myself I wasn’t looking for that one variant to rule them all (or for that one variant to bring them all and in the darkness bind them). I was looking to increase the chance of conversion for every single visitor. Certain pages were going to work better for certain audiences, and that was totally fine.
I wondered, though: how many variants would be too many? Would the machine learning recognize that some of these were not anything special and just stop sending traffic to them? And how long would it take to get results? With these questions in mind, I checked back on my first set of tests one month later…
Changing up the background color
Usually, color A/B tests are pretty much a waste of time. You need a lot of data to get accurate results, and most marketers don’t actually end up learning anything useful in the end. (Because color by itself means nothing, it always depends on the context of the page.)
That being said, we know there is some legitimate color theory and certain audience segments respond better to certain colors than others. So I thought it might be interesting to switch up the background on this landing page to see what would happen. And color me surprised—these variants are seeing some pretty dramatically different conversion rates:
Pink background – 12.82%
Green background – 21.43%
White background – 21.74%
Black background – 31.71%
One might start to speculate from these conversion rates that darker backgrounds perform better than the lighter backgrounds. But hold your horses, that’s thinking about this as an A/B test again. Here’s why Jordan Dawe, Senior Data Science Developer at Unbounce, says you should be cautious about drawing any conclusions from the conversion rates…
Smart Traffic is not sending visitors randomly—it’s trying to get the best traffic to the best variant. So in this case, it doesn’t mean that a black background will always convert higher than a pink background. There are likely portions of the audience going to each color that would be doing worse on others. Here’s what you can conclude: the color black is preferred by a portion of the traffic that converts highly.
It’s hard to shake that mindset of looking for a “winner” and trying to figure out “why” something is working. But I was starting to accept that different portions of the audience would always respond better to different variants—this was just the first time I’d been able to use AI to automatically serve up the best version.
Making big (and small) changes to the headline
For the next group of variants, I switched up the H1 in both small and big ways to see what effect that would have on the conversion rate. In some cases, this meant just swapping a single adjective (e.g., “jaw-dropping” for “drool-worthy”). In other cases, I went with a completely new line of copy altogether.
Here’s how the variants stacked up against each other:
See 27 Sales-Ready Ecommerce Landing Pages in Our Ultimate Lookbook – 25.81%
See 27 Stunning Ecommerce Landing Pages in Our Ultimate Lookbook – 25.93%
Get Ready to See 27 Jaw-Dropping Ecommerce Landing Page Examples – 28.13%
Get Serious Inspo for Supercharging Your Ecomm Sales – 35%
See 27 Drool-Worthy Ecommerce Landing Pages in Our Ultimate Lookbook – 40%
Again, each variant yielded a different conversion rate. I wondered if I kept testing different variations of the headlines and found one that performed best, could I deactivate all the other headline variants and just go with the “best” one? 
Here’s how Floss Taylor, Data Analyst at Unbounce, responded…
Smart Traffic doesn’t have champion variants. You don’t pick one at the end like you would in an A/B test. Although one variant may appear to be performing poorly, there could be a subset of traffic that it’s ideal for. You’re better off leaving it on long-term so it can work its magic.
Trying out different page layouts and hierarchies
The last set of variants I created messed with the actual structure and hierarchy of the page. I wanted to see if moving things around (or removing sections entirely) would influence the conversion rate. Here’s a sample of some of the experiments…
Removing the Headline – 16.67%
Adding a Double CTA – 21.95%
Moving the Testimonial Up the Page – 27.27%
Nothing too surprising here. And because I had created so many variants, Smart Traffic was taking longer than usual in “Learning Mode” to start giving me a conversion lift. Here’s how Floss Taylor explains it…
Smart Traffic needs approximately 50 visitors to understand which traffic would perform well for each new variant. If you have 15 variants and ~100 visitors per month, you’re going to have a long learning period where Smart Traffic cannot make accurate recommendations. I’d suggest starting off with a lower number of variants, and only adding more once you have sufficient traffic.
The “Strategic Marketer” Approach
So throwing spaghetti at the wall turned out to be… messy. (New parents beware.) For the SaaS optimization guide page, I wanted to be a bit more strategic. And I actually had a leg up for this one, because Anna Roginska, Growth Marketer at Unbounce, had already started with a Smart Traffic experiment on this page four months ago.
Anna had set up a test between two different variants. One had an image of the ecommerce lookbook as the hero graphic on the page, while the other used the image of conversion expert and author Talia Wolf. Anna says she decided on this second variant because of research she had seen on how photographs of people tend to convert better than products.
I put Talia up front because I knew from other tests I’ve run and research I’ve done. [Photographs of] people tend to convert better. I didn’t know if it would work better in this particular case, but I was able to set up a variant and use Smart Traffic to find out. And it just so happens that the algorithm started sending way more traffic to this variant.
Anna seemed to be onto something, too: her variant was converting at nearly double the rate for a large traffic subset. And while I now know we can’t consider this a “champion” variant like in an A/B test and learn from the results, we could iterate based on her design to target new audience segments.
I created a simple spreadsheet to develop my gameplan. The goal was to create five new versions of the page that would appeal to different visitors based on their attributes:
Reducing the word count to target mobile and “ready to download” visitors
For inspiration on my first variant, I consulted the 2020 Conversion Benchmark Report. The machine learning insights here suggested that SaaS landing pages with lower word counts and easier-to-read copy tend to perform better than their long-winded counterparts.
And while the original version of our download page was easy enough to read, it did have a long, wordy intro with a lot of extra detail. Could I increase our conversion rate for a portion of our audience if just focused on the bare essentials? I was ready to kill some darlings to find out…
Original Long-Form Version – 10%
Low Word Count Version – 21.43%
It seems there’s a segment of our traffic coming to this page who didn’t need to see all that extra info before they decided to fill out the form. I speculated that this variant might also perform better on mobile devices since it would be faster-loading and easier to scroll through. Interesting!
Switching the headline to target different audience segments
Next, I created an additional four page variants to speak to the different pain points and reasons our audience might want to download the guide. (Actually, this is something Talia herself recommends you do in the SaaS optimization guide.) I switched up the headline copy here, as well as some of the supporting text underneath to match. After a month, here’s what the conversion rates look like:
Get Talia’s Guide to Optimize – 19.05%
You Can’t Just Build – 23.08%
Optimization is a Lot of Work – 24%
Not Sure How to Optimize? – 33.33%
Each variant is serving a different segment of the audience, by speaking to the particular reason they want to download the guide most (e.g., maybe they don’t have the time to optimize, or maybe they don’t know how to get started). As Smart Traffic learns more about which variants perform best for which audience segments, we become that much more likely to score a conversion.
What I Learned Running These Smart Traffic Experiments
Smart Traffic absolutely makes optimization easier and faster for marketers who previously never had the time (or experience) to run A/B tests. It took me under two hours to set up and launch these experiments, and we’re already seeing some pretty impressive results just over a month later.
While the ecommerce lookbook page is still optimizing, the SaaS ebook page is showing a 12% lift in conversions compared to evenly splitting traffic among all these variants. And this is after only a month—the algorithm will keep improving to get us even better results over time. (Like a fine wine, or that suspiciously old cheese in my fridge.)
At the same time, I did walk away with a few important lessons learned. If you’re planning to use Smart Traffic to optimize your landing pages, here are some things to keep in mind before you get started:
There are no champion variants – Unlike traditional A/B testing, you won’t be able to point to one landing page variant at the end of your test and call it a winner. The machine learning algorithm automatically routes audiences differently based on their individual attributes, which means you have to be cautious when you’re analyzing the results.
The more variants you create, the longer you’ll wait – While it can be tempting to throw spaghetti at the wall and create dozens of variants for your landing page, this means you’ll also have to wait longer to see what sticks. Try starting out with three to five variations and take a more strategic approach based on research in your industry. (The 2020 Conversion Benchmark Report is a great place to start for some ideas.)
It’s (usually) better to leave low-converting variants active – Because Smart Traffic learns over time and continually improves, you’re typically better off leaving your variants active—even if their conversion rates aren’t all that impressive. The AI takes the risk out of optimization by automatically sending visitors to the page that suits them best. If you turn off variants, you may lose out on some of those conversions altogether.
It can be a lot of fun to get creative with the different page elements and try out new ideas. You just might want to come up with a bit of a plan first and be strategic with your approach. Still, it’s better to experiment and optimize with Smart Traffic (even if you make some mistakes along the way) than to never optimize at all.
(And in case you were worried, yep—I managed to get my to-do list done, too. )
from Marketing https://unbounce.com/marketing-ai/smart-traffic-experiments/ via http://www.rssmix.com/
0 notes
itsjessicaisreal · 4 years
Text
[Experiment] How AI is Changing the Way We Optimize at Unbounce
When I first encountered A/B testing, I immediately wanted to become the type of marketer who tested everything. The idea sounded fun to me. Like being a mad scientist running experiments to prove when my work was actually “working.”
Turns out though, there’s always a long list of other things to do first… blog posts to write, campaigns to launch, and don’t get me started on the meetings! I’m not alone in this, either. A lot of marketers are just too darned busy to follow up and optimize the stuff they’ve already shipped. According to HubSpot, only 17% of marketers use landing page A/B tests to improve conversion rates.
A small glimpse of my ever-growing to-do list. Ain’t nobody got time for A/B tests.
Sure, running a split test with one or two variants always sounds easy enough. But once you take a closer look at the process, you realize just how complex it can actually be. You need to make sure you have… 
The right duration and sample size.
Taken into account any external factors or validity threats.
Learned how to interpret the results correctly, too.
But—while there will always be a time and place for A/B testing—there’s also now an easier and faster way for marketers to optimize. Smart Traffic is a new Unbounce tool that uses the power of AI and machine learning to get you more conversions. Every day, more marketers are using Smart Traffic to “automagically” optimize their landing pages. But whenever we launch anything new, we like to test it out for ourselves to learn alongside you (and keep you up to speed on what to try next).
Here’s what I learned after taking Smart Traffic for a test drive myself…
Shifting Your Mindset to Optimize with AI
I know many marketers are (perhaps) skeptical when it comes to promises of machine learning, artificial intelligence, or magical “easy” buttons that get them better results. But AI is all around us and it’s already changing the way we do marketing. Landing page optimization is just one more area of the job where you no longer need to do everything yourself manually.
Smart Traffic augments your marketing skills and automatically sends visitors to the landing page variant where they’re most likely to convert (based on how similar page visitors have converted before). It makes routing decisions faster than any human ever could (thank you, AI magic), and “learns” which page variant is a perfect match for each different visitor. This ultimately means no more “champion” variants. Instead, you’re free to create multiple different pages to appeal to different groups of visitors and run ‘em all at once.
This is very different from A/B testing and honestly—it can feel kinda weird at first. You’ve got to trust in the machine learning to figure out what works best and what doesn’t. Data scientists call this the “black box” problem: data goes in, decisions come out, but you never really get the full understanding of what happened in between. 
Smart Traffic is fundamentally different from A/B Testing. You can learn more about how it works here.
For marketers using Smart Traffic, this means shifting your mindset and starting to think about optimization differently. Unlike A/B testing, you’re not looking for those “aha” moments to apply to your next campaign, or a one-size-fits-all “winning” variant. Instead, you’re looking to discover what works best for different subsets of your audience. This gives you unlimited creativity to try out new marketing ideas, makes it easier and less risky for you to optimize, and gives you an average conversion lift of 30% compared to splitting the traffic evenly across multiple variants. (Woah.)
My Experiment with Smart Traffic
I know all this because I recently experimented with variant creation myself to better understand this new AI optimization mindset. I created 15 variants across two separate landing pages using Smart Traffic to discover…
How easy is it to optimize with an AI-powered optimization tool?
Could I quickly set up the tests in Unbounce while still getting those other to-do’s done?
What kind of conversion lift would I see from just a few hours invested?
I took a little bit of my inspiration from Ms. Frizzle on the Magic School Bus. No, not her haircut, her catchphrase: “Take chances, make mistakes, get messy!”
Oh, so that’s where she got all her good ideas.
Creating 15 Variants in Under Two Hours
The beauty of Smart Traffic is there are no limits to how many variants you can create and it automatically starts optimizing in as few as 50 visits. Just hit the “optimize” button and you’re off to the races. Could it really be that simple?
My guinea pigs for this experiment would be two recent campaigns our marketing team had worked on: the ecommerce lookbook and the SaaS optimization guide. The team had created both of these ebook download pages in Unbounce, but we hadn’t been able to return to them and optimize very much in the months since we published.
The original landing pages would serve as my control variants. (Click to see the full pages.)
Before starting, I consulted with Anna Roginska, Growth Marketer at Unbounce, to get her input on how I should create my variants. She advised:
You can take the ‘spaghetti at the wall’ approach, where you create a bunch of variants and just leave them to Smart Traffic to see what happens. It’s that ‘set it and forget it’ mentality. That’s interesting, but when you look at a bowl of spaghetti… There’s a lot of noodles in there. You won’t necessarily get to explain why something is working or not working.
The other approach is to be more strategic and focused. I think there’s a huge benefit to going in with a plan. Create maybe only five variants and give them each a specific purpose. Then, you can see how they perform and create new iterations for different portions of the audience.
I had two landing pages to work with, so I thought I’d give both approaches a try. But with only a few hours scheduled in my calendar to complete all these variants, I needed to move fast.
The “Spaghetti at the Wall” Approach to Variant Creation
On the ecommerce lookbook page, I wanted to spend less time planning and more time creating. Whereas in A/B testing you need a proper test hypothesis and a careful plan for each variant, Smart Traffic lets you get creative and try out new ideas on the fly. Your variants don’t have to be perfect—they just need to be different enough to appeal to new audience segments.
This meant I didn’t have to make any hard or fast choices about which one element to “test” on the landing page. I could create 15 different variants that varied wildly from one another. Some used different colors, some had different headlines, some completely changed up the layout of the page.
This is something you just can’t do in a traditional A/B test where you’re looking to find a “winner” and understand why it “wins.” I had to remind myself I wasn’t looking for that one variant to rule them all (or for that one variant to bring them all and in the darkness bind them). I was looking to increase the chance of conversion for every single visitor. Certain pages were going to work better for certain audiences, and that was totally fine.
I wondered, though: how many variants would be too many? Would the machine learning recognize that some of these were not anything special and just stop sending traffic to them? And how long would it take to get results? With these questions in mind, I checked back on my first set of tests one month later…
Changing up the background color
Usually, color A/B tests are pretty much a waste of time. You need a lot of data to get accurate results, and most marketers don’t actually end up learning anything useful in the end. (Because color by itself means nothing, it always depends on the context of the page.)
That being said, we know there is some legitimate color theory and certain audience segments respond better to certain colors than others. So I thought it might be interesting to switch up the background on this landing page to see what would happen. And color me surprised—these variants are seeing some pretty dramatically different conversion rates:
Pink background – 12.82%
Green background – 21.43%
White background – 21.74%
Black background – 31.71%
One might start to speculate from these conversion rates that darker backgrounds perform better than the lighter backgrounds. But hold your horses, that’s thinking about this as an A/B test again. Here’s why Jordan Dawe, Senior Data Science Developer at Unbounce, says you should be cautious about drawing any conclusions from the conversion rates…
Smart Traffic is not sending visitors randomly—it’s trying to get the best traffic to the best variant. So in this case, it doesn’t mean that a black background will always convert higher than a pink background. There are likely portions of the audience going to each color that would be doing worse on others. Here’s what you can conclude: the color black is preferred by a portion of the traffic that converts highly.
It’s hard to shake that mindset of looking for a “winner” and trying to figure out “why” something is working. But I was starting to accept that different portions of the audience would always respond better to different variants—this was just the first time I’d been able to use AI to automatically serve up the best version.
Making big (and small) changes to the headline
For the next group of variants, I switched up the H1 in both small and big ways to see what effect that would have on the conversion rate. In some cases, this meant just swapping a single adjective (e.g., “jaw-dropping” for “drool-worthy”). In other cases, I went with a completely new line of copy altogether.
Here’s how the variants stacked up against each other:
See 27 Sales-Ready Ecommerce Landing Pages in Our Ultimate Lookbook – 25.81%
See 27 Stunning Ecommerce Landing Pages in Our Ultimate Lookbook – 25.93%
Get Ready to See 27 Jaw-Dropping Ecommerce Landing Page Examples – 28.13%
Get Serious Inspo for Supercharging Your Ecomm Sales – 35%
See 27 Drool-Worthy Ecommerce Landing Pages in Our Ultimate Lookbook – 40%
Again, each variant yielded a different conversion rate. I wondered if I kept testing different variations of the headlines and found one that performed best, could I deactivate all the other headline variants and just go with the “best” one? 
Here’s how Floss Taylor, Data Analyst at Unbounce, responded…
Smart Traffic doesn’t have champion variants. You don’t pick one at the end like you would in an A/B test. Although one variant may appear to be performing poorly, there could be a subset of traffic that it’s ideal for. You’re better off leaving it on long-term so it can work its magic.
Trying out different page layouts and hierarchies
The last set of variants I created messed with the actual structure and hierarchy of the page. I wanted to see if moving things around (or removing sections entirely) would influence the conversion rate. Here’s a sample of some of the experiments…
Removing the Headline – 16.67%
Adding a Double CTA – 21.95%
Moving the Testimonial Up the Page – 27.27%
Nothing too surprising here. And because I had created so many variants, Smart Traffic was taking longer than usual in “Learning Mode” to start giving me a conversion lift. Here’s how Floss Taylor explains it…
Smart Traffic needs approximately 50 visitors to understand which traffic would perform well for each new variant. If you have 15 variants and ~100 visitors per month, you’re going to have a long learning period where Smart Traffic cannot make accurate recommendations. I’d suggest starting off with a lower number of variants, and only adding more once once you have sufficient traffic.
The “Strategic Marketer” Approach
So throwing spaghetti at the wall turned out to be… messy. (New parents beware.) For the SaaS optimization guide page, I wanted to be a bit more strategic. And I actually had a leg up for this one, because Anna Roginska, Growth Marketer at Unbounce, had already started with a Smart Traffic experiment on this page four months ago.
Anna had set up a test between two different variants. One had an image of the ecommerce lookbook as the hero graphic on the page, while the other used the image of conversion expert and author Talia Wolf. Anna says she decided on this second variant because of research she had seen on how photographs of people tend to convert better than products.
I put Talia up front because I knew from other tests I’ve run and research I’ve done. [Photographs of] people tend to convert better. I didn’t know if it would work better in this particular case, but I was able to set up a variant and use Smart Traffic to find out. And it just so happens that the algorithm started sending way more traffic to this variant.
Anna seemed to be onto something, too: her variant was converting at nearly double the rate for a large traffic subset. And while I now know we can’t consider this a “champion” variant like in an A/B test and learn from the results, we could iterate based on her design to target new audience segments.
I created a simple spreadsheet to develop my gameplan. The goal was to create five new versions of the page that would appeal to different visitors based on their attributes:
Reducing the word count to target mobile and “ready to download” visitors
For inspiration on my first variant, I consulted the 2020 Conversion Benchmark Report. The machine learning insights here suggested that SaaS landing pages with lower word counts and easier-to-read copy tend to perform better than their long-winded counterparts.
And while the original version of our download page was easy enough to read, it did have a long, wordy intro with a lot of extra detail. Could I increase our conversion rate for a portion of our audience if just focused on the bare essentials? I was ready to kill some darlings to find out…
Original Long-Form Version – 10%
Low Word Count Version – 21.43%
It seems there’s a segment of our traffic coming to this page who didn’t need to see all that extra info before they decided to fill out the form. I speculated that this variant might also perform better on mobile devices since it would be faster-loading and easier to scroll through. Interesting!
Switching the headline to target different audience segments
Next, I created an additional four page variants to speak to the different pain points and reasons our audience might want to download the guide. (Actually, this is something Talia herself recommends you do in the SaaS optimization guide.) I switched up the headline copy here, as well as some of the supporting text underneath to match. After a month, here’s what the conversion rates look like:
Get Talia’s Guide to Optimize – 19.05%
You Can’t Just Build – 23.08%
Optimization is a Lot of Work – 24%
Not Sure How to Optimize? – 33.33%
Each variant is serving a different segment of the audience, by speaking to the particular reason they want to download the guide most (e.g., maybe they don’t have the time to optimize, or maybe they don’t know how to get started). As Smart Traffic learns more about which variants perform best for which audience segments, we become that much more likely to score a conversion.
What I Learned Running These Smart Traffic Experiments
Smart Traffic absolutely makes optimization easier and faster for marketers who previously never had the time (or experience) to run A/B tests. It took me under two hours to set up and launch these experiments, and we’re already seeing some pretty impressive results just over a month later.
While the ecommerce lookbook page is still optimizing, the SaaS ebook page is showing a 12% lift in conversions compared to evenly splitting traffic among all these variants. And this is after only a month—the algorithm will keep improving to get us even better results over time. (Like a fine wine, or that suspiciously old cheese in my fridge.)
At the same time, I did walk away with a few important lessons learned. If you’re planning to use Smart Traffic to optimize your landing pages, here are some things to keep in mind before you get started:
There are no champion variants – Unlike traditional A/B testing, you won’t be able to point to one landing page variant at the end of your test and call it a winner. The machine learning algorithm automatically routes audiences differently based on their individual attributes, which means you have to be cautious when you’re analyzing the results.
The more variants you create, the longer you’ll wait – While it can be tempting to throw spaghetti at the wall and create dozens of variants for your landing page, this means you’ll also have to wait longer to see what sticks. Try starting out with three to five variations and take a more strategic approach based on research in your industry. (The 2020 Conversion Benchmark Report is a great place to start for some ideas.)
It’s (usually) better to leave low-converting variants active – Because Smart Traffic learns over time and continually improves, you’re typically better off leaving your variants active—even if their conversion rates aren’t all that impressive. The AI takes the risk out of optimization by automatically sending visitors to the page that suits them best. If you turn off variants, you may lose out on some of those conversions altogether.
It can be a lot of fun to get creative with the different page elements and try out new ideas. You just might want to come up with a bit of a plan first and be strategic with your approach. Still, it’s better to experiment and optimize with Smart Traffic (even if you make some mistakes along the way) than to never optimize at all.
(And in case you were worried, yep—I managed to get my to-do list done, too. )
from Marketing https://unbounce.com/marketing-ai/smart-traffic-experiments/ via http://www.rssmix.com/
0 notes
annaxkeating · 4 years
Text
[Experiment] How AI is Changing the Way We Optimize at Unbounce
When I first encountered A/B testing, I immediately wanted to become the type of marketer who tested everything. The idea sounded fun to me. Like being a mad scientist running experiments to prove when my work was actually “working.”
Turns out though, there’s always a long list of other things to do first… blog posts to write, campaigns to launch, and don’t get me started on the meetings! I’m not alone in this, either. A lot of marketers are just too darned busy to follow up and optimize the stuff they’ve already shipped. According to HubSpot, only 17% of marketers use landing page A/B tests to improve conversion rates.
A small glimpse of my ever-growing to-do list. Ain’t nobody got time for A/B tests.
Sure, running a split test with one or two variants always sounds easy enough. But once you take a closer look at the process, you realize just how complex it can actually be. You need to make sure you have… 
The right duration and sample size.
Taken into account any external factors or validity threats.
Learned how to interpret the results correctly, too.
But—while there will always be a time and place for A/B testing—there’s also now an easier and faster way for marketers to optimize. Smart Traffic is a new Unbounce tool that uses the power of AI and machine learning to get you more conversions. Every day, more marketers are using Smart Traffic to “automagically” optimize their landing pages. But whenever we launch anything new, we like to test it out for ourselves to learn alongside you (and keep you up to speed on what to try next).
Here’s what I learned after taking Smart Traffic for a test drive myself…
Shifting Your Mindset to Optimize with AI
I know many marketers are (perhaps) skeptical when it comes to promises of machine learning, artificial intelligence, or magical “easy” buttons that get them better results. But AI is all around us and it’s already changing the way we do marketing. Landing page optimization is just one more area of the job where you no longer need to do everything yourself manually.
Smart Traffic augments your marketing skills and automatically sends visitors to the landing page variant where they’re most likely to convert (based on how similar page visitors have converted before). It makes routing decisions faster than any human ever could (thank you, AI magic), and “learns” which page variant is a perfect match for each different visitor. This ultimately means no more “champion” variants. Instead, you’re free to create multiple different pages to appeal to different groups of visitors and run ‘em all at once.
This is very different from A/B testing and honestly—it can feel kinda weird at first. You’ve got to trust in the machine learning to figure out what works best and what doesn’t. Data scientists call this the “black box” problem: data goes in, decisions come out, but you never really get the full understanding of what happened in between. 
Smart Traffic is fundamentally different from A/B Testing. You can learn more about how it works here.
For marketers using Smart Traffic, this means shifting your mindset and starting to think about optimization differently. Unlike A/B testing, you’re not looking for those “aha” moments to apply to your next campaign, or a one-size-fits-all “winning” variant. Instead, you’re looking to discover what works best for different subsets of your audience. This gives you unlimited creativity to try out new marketing ideas, makes it easier and less risky for you to optimize, and gives you an average conversion lift of 30% compared to splitting the traffic evenly across multiple variants. (Woah.)
My Experiment with Smart Traffic
I know all this because I recently experimented with variant creation myself to better understand this new AI optimization mindset. I created 15 variants across two separate landing pages using Smart Traffic to discover…
How easy is it to optimize with an AI-powered optimization tool?
Could I quickly set up the tests in Unbounce while still getting those other to-do’s done?
What kind of conversion lift would I see from just a few hours invested?
I took a little bit of my inspiration from Ms. Frizzle on the Magic School Bus. No, not her haircut, her catchphrase: “Take chances, make mistakes, get messy!”
Oh, so that’s where she got all her good ideas.
Creating 15 Variants in Under Two Hours
The beauty of Smart Traffic is there are no limits to how many variants you can create and it automatically starts optimizing in as few as 50 visits. Just hit the “optimize” button and you’re off to the races. Could it really be that simple?
My guinea pigs for this experiment would be two recent campaigns our marketing team had worked on: the ecommerce lookbook and the SaaS optimization guide. The team had created both of these ebook download pages in Unbounce, but we hadn’t been able to return to them and optimize very much in the months since we published.
The original landing pages would serve as my control variants. (Click to see the full pages.)
Before starting, I consulted with Anna Roginska, Growth Marketer at Unbounce, to get her input on how I should create my variants. She advised:
You can take the ‘spaghetti at the wall’ approach, where you create a bunch of variants and just leave them to Smart Traffic to see what happens. It’s that ‘set it and forget it’ mentality. That’s interesting, but when you look at a bowl of spaghetti… There’s a lot of noodles in there. You won’t necessarily get to explain why something is working or not working.
The other approach is to be more strategic and focused. I think there’s a huge benefit to going in with a plan. Create maybe only five variants and give them each a specific purpose. Then, you can see how they perform and create new iterations for different portions of the audience.
I had two landing pages to work with, so I thought I’d give both approaches a try. But with only a few hours scheduled in my calendar to complete all these variants, I needed to move fast.
The “Spaghetti at the Wall” Approach to Variant Creation
On the ecommerce lookbook page, I wanted to spend less time planning and more time creating. Whereas in A/B testing you need a proper test hypothesis and a careful plan for each variant, Smart Traffic lets you get creative and try out new ideas on the fly. Your variants don’t have to be perfect—they just need to be different enough to appeal to new audience segments.
This meant I didn’t have to make any hard or fast choices about which one element to “test” on the landing page. I could create 15 different variants that varied wildly from one another. Some used different colors, some had different headlines, some completely changed up the layout of the page.
This is something you just can’t do in a traditional A/B test where you’re looking to find a “winner” and understand why it “wins.” I had to remind myself I wasn’t looking for that one variant to rule them all (or for that one variant to bring them all and in the darkness bind them). I was looking to increase the chance of conversion for every single visitor. Certain pages were going to work better for certain audiences, and that was totally fine.
I wondered, though: how many variants would be too many? Would the machine learning recognize that some of these were not anything special and just stop sending traffic to them? And how long would it take to get results? With these questions in mind, I checked back on my first set of tests one month later…
Changing up the background color
Usually, color A/B tests are pretty much a waste of time. You need a lot of data to get accurate results, and most marketers don’t actually end up learning anything useful in the end. (Because color by itself means nothing, it always depends on the context of the page.)
That being said, we know there is some legitimate color theory and certain audience segments respond better to certain colors than others. So I thought it might be interesting to switch up the background on this landing page to see what would happen. And color me surprised—these variants are seeing some pretty dramatically different conversion rates:
Pink background – 12.82%
Green background – 21.43%
White background – 21.74%
Black background – 31.71%
One might start to speculate from these conversion rates that darker backgrounds perform better than the lighter backgrounds. But hold your horses, that’s thinking about this as an A/B test again. Here’s why Jordan Dawe, Senior Data Science Developer at Unbounce, says you should be cautious about drawing any conclusions from the conversion rates…
Smart Traffic is not sending visitors randomly—it’s trying to get the best traffic to the best variant. So in this case, it doesn’t mean that a black background will always convert higher than a pink background. There are likely portions of the audience going to each color that would be doing worse on others. Here’s what you can conclude: the color black is preferred by a portion of the traffic that converts highly.
It’s hard to shake that mindset of looking for a “winner” and trying to figure out “why” something is working. But I was starting to accept that different portions of the audience would always respond better to different variants—this was just the first time I’d been able to use AI to automatically serve up the best version.
Making big (and small) changes to the headline
For the next group of variants, I switched up the H1 in both small and big ways to see what effect that would have on the conversion rate. In some cases, this meant just swapping a single adjective (e.g., “jaw-dropping” for “drool-worthy”). In other cases, I went with a completely new line of copy altogether.
Here’s how the variants stacked up against each other:
See 27 Sales-Ready Ecommerce Landing Pages in Our Ultimate Lookbook – 25.81%
See 27 Stunning Ecommerce Landing Pages in Our Ultimate Lookbook – 25.93%
Get Ready to See 27 Jaw-Dropping Ecommerce Landing Page Examples – 28.13%
Get Serious Inspo for Supercharging Your Ecomm Sales – 35%
See 27 Drool-Worthy Ecommerce Landing Pages in Our Ultimate Lookbook – 40%
Again, each variant yielded a different conversion rate. I wondered if I kept testing different variations of the headlines and found one that performed best, could I deactivate all the other headline variants and just go with the “best” one? 
Here’s how Floss Taylor, Data Analyst at Unbounce, responded…
Smart Traffic doesn’t have champion variants. You don’t pick one at the end like you would in an A/B test. Although one variant may appear to be performing poorly, there could be a subset of traffic that it’s ideal for. You’re better off leaving it on long-term so it can work its magic.
Trying out different page layouts and hierarchies
The last set of variants I created messed with the actual structure and hierarchy of the page. I wanted to see if moving things around (or removing sections entirely) would influence the conversion rate. Here’s a sample of some of the experiments…
Removing the Headline – 16.67%
Adding a Double CTA – 21.95%
Moving the Testimonial Up the Page – 27.27%
Nothing too surprising here. And because I had created so many variants, Smart Traffic was taking longer than usual in “Learning Mode” to start giving me a conversion lift. Here’s how Floss Taylor explains it…
Smart Traffic needs approximately 50 visitors to understand which traffic would perform well for each new variant. If you have 15 variants and ~100 visitors per month, you’re going to have a long learning period where Smart Traffic cannot make accurate recommendations. I’d suggest starting off with a lower number of variants, and only adding more once once you have sufficient traffic.
The “Strategic Marketer” Approach
So throwing spaghetti at the wall turned out to be… messy. (New parents beware.) For the SaaS optimization guide page, I wanted to be a bit more strategic. And I actually had a leg up for this one, because Anna Roginska, Growth Marketer at Unbounce, had already started with a Smart Traffic experiment on this page four months ago.
Anna had set up a test between two different variants. One had an image of the ecommerce lookbook as the hero graphic on the page, while the other used the image of conversion expert and author Talia Wolf. Anna says she decided on this second variant because of research she had seen on how photographs of people tend to convert better than products.
I put Talia up front because I knew from other tests I’ve run and research I’ve done. [Photographs of] people tend to convert better. I didn’t know if it would work better in this particular case, but I was able to set up a variant and use Smart Traffic to find out. And it just so happens that the algorithm started sending way more traffic to this variant.
Anna seemed to be onto something, too: her variant was converting at nearly double the rate for a large traffic subset. And while I now know we can’t consider this a “champion” variant like in an A/B test and learn from the results, we could iterate based on her design to target new audience segments.
I created a simple spreadsheet to develop my gameplan. The goal was to create five new versions of the page that would appeal to different visitors based on their attributes:
Reducing the word count to target mobile and “ready to download” visitors
For inspiration on my first variant, I consulted the 2020 Conversion Benchmark Report. The machine learning insights here suggested that SaaS landing pages with lower word counts and easier-to-read copy tend to perform better than their long-winded counterparts.
And while the original version of our download page was easy enough to read, it did have a long, wordy intro with a lot of extra detail. Could I increase our conversion rate for a portion of our audience if just focused on the bare essentials? I was ready to kill some darlings to find out…
Original Long-Form Version – 10%
Low Word Count Version – 21.43%
It seems there’s a segment of our traffic coming to this page who didn’t need to see all that extra info before they decided to fill out the form. I speculated that this variant might also perform better on mobile devices since it would be faster-loading and easier to scroll through. Interesting!
Switching the headline to target different audience segments
Next, I created an additional four page variants to speak to the different pain points and reasons our audience might want to download the guide. (Actually, this is something Talia herself recommends you do in the SaaS optimization guide.) I switched up the headline copy here, as well as some of the supporting text underneath to match. After a month, here’s what the conversion rates look like:
Get Talia’s Guide to Optimize – 19.05%
You Can’t Just Build – 23.08%
Optimization is a Lot of Work – 24%
Not Sure How to Optimize? – 33.33%
Each variant is serving a different segment of the audience, by speaking to the particular reason they want to download the guide most (e.g., maybe they don’t have the time to optimize, or maybe they don’t know how to get started). As Smart Traffic learns more about which variants perform best for which audience segments, we become that much more likely to score a conversion.
What I Learned Running These Smart Traffic Experiments
Smart Traffic absolutely makes optimization easier and faster for marketers who previously never had the time (or experience) to run A/B tests. It took me under two hours to set up and launch these experiments, and we’re already seeing some pretty impressive results just over a month later.
While the ecommerce lookbook page is still optimizing, the SaaS ebook page is showing a 12% lift in conversions compared to evenly splitting traffic among all these variants. And this is after only a month—the algorithm will keep improving to get us even better results over time. (Like a fine wine, or that suspiciously old cheese in my fridge.)
At the same time, I did walk away with a few important lessons learned. If you’re planning to use Smart Traffic to optimize your landing pages, here are some things to keep in mind before you get started:
There are no champion variants – Unlike traditional A/B testing, you won’t be able to point to one landing page variant at the end of your test and call it a winner. The machine learning algorithm automatically routes audiences differently based on their individual attributes, which means you have to be cautious when you’re analyzing the results.
The more variants you create, the longer you’ll wait – While it can be tempting to throw spaghetti at the wall and create dozens of variants for your landing page, this means you’ll also have to wait longer to see what sticks. Try starting out with three to five variations and take a more strategic approach based on research in your industry. (The 2020 Conversion Benchmark Report is a great place to start for some ideas.)
It’s (usually) better to leave low-converting variants active – Because Smart Traffic learns over time and continually improves, you’re typically better off leaving your variants active—even if their conversion rates aren’t all that impressive. The AI takes the risk out of optimization by automatically sending visitors to the page that suits them best. If you turn off variants, you may lose out on some of those conversions altogether.
It can be a lot of fun to get creative with the different page elements and try out new ideas. You just might want to come up with a bit of a plan first and be strategic with your approach. Still, it’s better to experiment and optimize with Smart Traffic (even if you make some mistakes along the way) than to never optimize at all.
(And in case you were worried, yep—I managed to get my to-do list done, too. )
from Digital https://unbounce.com/marketing-ai/smart-traffic-experiments/ via http://www.rssmix.com/
0 notes