Name: Group Coaching Session - Week 3 (Motion Creative Strategy Bootcamp)
Uploaded: 2026-04-02T19:17:17.133Z
Duration: 1 h 27 min 24 s
Description: Group Coaching Session - Week 3 (Motion Creative Strategy Bootcamp)

Transcript for "Group Coaching Session - Week 3 (Motion Creative Strategy Bootcamp)": Hi, everyone. So fun. Our our girl Melissa backstage sent us on stage and goes, have fun. We're here again for another awkward week six of realness. And this time, we're joined by a new person this week, Naomi. We'll go through intros in a second here, but let me go ahead and share my screen and get us started. I have a couple of things to share today. So first, welcome to week three Thursday call, Thursday crew. We're developing tiny bit of a rivalry between Tuesdays and Thursdays. I kinda love it. I don't wanna encourage it, but I kinda do. So first, I wanted to go through some of our community wins. You guys are still blowing us up all over LinkedIn, all over x, all over Instagram. I wanna encourage that behavior. We love to see it. I love to see everybody sharing your enthusiasm, what you've learned, your takeaways. It's been great. Also, the Slack Slack channels continue to blow up. It's insane. I spend a couple hours in there every single day, and I still feel bad I can't get to everyone. Also, this week's session with Dara was so good. You guys broke out your sunglasses, and we had our first little tiny sliver of AI hate in the comments also, and we are just getting started. So I just wanted to highlight that. Menno, I know I saw I saw you in the chat. Yesterday, I think, was your first day as a creative strategist. Congratulations. Quite a few of you have said you're you're in job interviews. You're getting jobs as creative strategists. You're building your portfolios. Congratulations. We love to see that too. And then last, I did wanna highlight. I thought this was, like, a really great Slack message that I saw the other day where one person asked, like, is motion, like, b to b? Like like, what is the difference between b to b and d to c? People are asking all this to all the time. And then Patrick replied and said, it's always human to human. And I just really liked his perspective of, like, at the end of the day, it doesn't matter who you are and who you're selling to. You're selling to a human. So, yeah, I just wanted to highlight that. Let's go into our intros. If we haven't met yet, I'm Alicia. I'm a creative strategist at Motion. I do a little bit of paid, a little bit of organic. You might have seen me on TikTok, and I also work really closely with our product and engineering team to make Motion better. Jake and Sarah and Eric will join us in following weeks, and you might have met them too. But for now, I'm gonna pass it off to Joanna for her intro. Go for it. Hi. I'm Joanna. You can call me Joe. I have been a creative strategist for, like, seven or so years. It's been a long haul. Previously, HexClad and Birddogs and TubeScience, and now I consult. Yeah. And how about you, Naomi? You're new. Yes. Hello. Yes. I'm at the I'm the creative strategist here at Space Goods, but I've been in the field for about five years now in kind of wearing many hats in startup world, but mainly in creative and content. Amazing. So we're gonna have some content for you today. First, I wanted to mention that, yes, everything, including this deck and all of the resources we're gonna show today will be shared with you guys in Slack and and an email. So don't worry about that. I know we're gonna get some questions anyway. And just a reminder, there is if you look in the chat, look up. And to the right of the word chat, there's a docs tab. There's some stuff in there for you as well. So just letting you know we will be giving you everything, and it'll be in Slack. Okay. So I wanted to do something a little bit different this week because the homework was a bit more abstract and personal. So I wanted to do a little bit of a recap of everything that Dara talked about on Tuesday. Saw I a lot of people who were like, woah. Woah. Woah. You're going too fast. Can you slow down a little bit? Totally fair. Dara is just like an absolute, like, fire hose, as we like to say, of value. So I just wanted to recap that really fast and show you guys how what she said on Tuesday really resonated with me and my experience with that. So she really talked about how to go from research to creative decisions. So, like, how do you do all of this research and then turn it into ads? Because as she said, the job of a creative strategist isn't to drown in ideas and insights. It's to know which ones to bet on. So the first thing that she talked about with us was personas, which is a huge unlock. So, particularly, she talked about identifying the gap between who your ads are targeting right now, like who you're actually speaking to, and who is actually buying by comparing your ad library to your customer reviews and finding that gap and that opportunity in the middle. So she recommended two different ways to do this. One was to throw your Facebook ads library into Claude and have it analyze your ads. You can also look inside of inspo in Motion. And that's a great way if you are just doing research on behalf of a brand or you don't have access to actual creative data. If you do, though, and if you have a data source connected inside of Motion, you're like me, you will notice that there is an intended audience tag inside of your Motion dashboard. A little fun fact is that I wrote the prompts for this intended audience tag. So I did a ton of research last year and a lot of testing. Wanted to tell you a fun story about identifying this persona gap. So when we launched this intended audience tag inside of Motion's account, we noticed that most of our creative was being tagged as creative strategists. So aka the LLM, the AI was saying, hey. This ad is speaking to creative strategists, as you can see here, the majority of them. And our CEO, Reza, went, Alicia, I think your prompt is wrong because there's, like, two ads that are talking to media buyers. And where are the founders? And where are the growth managers? And where are the d to c leaders? Like, truly, this should be wrong. And I was like, Reza, the CEO of Motion, it is not wrong. Your ads are primarily speaking to creative strategists. And that was a huge gap that none of us had really realized at the time. So I have since done a whole persona analysis, which I can talk about how I went through that process with you guys as well. One interesting thing about motion, and I hope I don't get fired for saying this, is that the thing about targeting or speaking to a creative strategist is that you can be an intro creative strategist, and you can be an expert creative strategist. So that persona is vastly different. So speaking to people through roles is actually not hugely beneficial to us per se. But the way that we like to speak to people is through their sophistication level. So if you're really new to creative strategy, if you're an expert at it, and also through pain points. So I created these two matrixes when I did my persona analysis. And then what I was able to do, and let me show you this in motion, is I uploaded my matrixes into Runneth, which is the AI chat inside of motion, and I asked it to do a gap analysis for me. And it was pretty harsh, the same way that I was when I talked to Reza, where it said the novice profile is almost completely absent. The sophisticated profile has zero representation, and the unsophisticated profile is covered but one dimensional, aka creative strategist. So if you're able to do that persona analysis, you can throw it into motion and compare with your ads as well. So I just wanted to share my experience doing this exercise. Okay. The next thing that Dara talked about was the idea that you need to do a diagnosis, because if you don't have a diagnosis, you don't have a strategy. And what a diagnosis can be can be simmered down to is a one to two sentence answer about why are you making this ad. Because if you don't have a reason about why you're making this ad except for that I think this would be a good idea, that's not a strong enough reason. So there's two things that we wanna share with you today. One is Joanna has made this amazing creative strategy one pager to help you work through that diagnosis. And two, I have shared this ten minute event that I did almost two years ago with Motion that is still pretty relevant today, which was just going over the basic fundamental creative strategy workflow and, like, how you go from, like, briefing to pipeline to analysis to context documentation. And one of the things that I mentioned in here two years ago was that in every single one of my briefs, I include a sentence that I called context about why you're making that ad. So a good like way to practice this of like implementing that diagnosis into your workflow is just included in every single brief. Why are you making this ad? It's just like a reminder to yourself that you have to have a reason. And then you can even go a little bit a level deeper and write yourself a formal hypothesis, which I also really like to do. But, yeah, just something I recommend. And of course, we will link you to both of these documents as well. Okay. And the last thing that Dara was talking about was the creative road map and how the sheer act of writing your ideas down and prioritizing them will put you ahead of most teams. And I think Dara was super correct. I've talked to so many different teams who are just totally going by like, on vibes. They have just Google Docs everywhere, things in Notion, things in Airtable, things in ClickUp, things in they're all over the place. This creative road map will keep you and your team organized and on point in a central location, and it's so necessary. So she did also share this creative road map template, and we'll link that for you guys as well. And then what this really is able to help you do is that it helps you identify the ideas that you have that have the most evidence for performance and rank them. So if you have all of your ideas in one central place, you're able to organize them and decide what to do next based on evidence, and that's what the skill was about this week. Like, that's the that's the thing that we really want you guys to get good at is going from this kind of mess on the left of all of the things that you could do, looking for evidence to turn that list into something like this of the things that are the highest opportunity that you should be doing. So these are this is a slide that Dara shared on her Tuesday presentation also about how to rank evidence, and you'll notice, like, things like because our competitors did it or because I think it's a good idea is all the way at the bottom because that is not a strong enough diagnosis. But pressure testing past performance and a new format for proven messaging, that's, like, a really good diagnosis. That's a really good reason to make that ad. So, yeah, that's what the road map was all about. Okay. Today, we're gonna cover Naomi is gonna go into how to conduct a creative retro to find those evidence based opportunities. Joanna is gonna talk about how to split your budget based on creative testing goals for road mapping. I'm gonna talk about how to find new personas and angles if you're new and have no reviews. And then if we have some time, we're also gonna talk about how to gut check AI. So a lot of this is based out of the questions that we saw coming up in the comments pretty frequently on Tuesday. And, hopefully, this helps you guys connect everything that Dara was talking about on Tuesday and see some, like, real life examples of how we put this in practice. So with that being said, we have our first student. Please raise your hand. I believe you volunteered to go on stage with us. We might have an awkward thirty seconds while we figure that out. But Patrick had a question that we would love to answer. Hi, Oh my gosh. everyone. It worked. Hi, Patrick. Hi. Hey. Yeah. Nice to meet you. Greetings from France, Germany. Yep. It's it's in the evening now. And, yeah, my question is about the holy Andromeda update. and testing because there is so much confuse confusion also when I hear the people talking in link in LinkedIn or something. What is possible in terms of an iteration? What is similar? What is the same for for the meta algorithm. So how do you approach creative testing right now? Because it changes almost every week. Mhmm. I think there's two two ways to think about creative testing. Well, two two ones that I would, like, group them into first. One is, like, what is one individual creative test, and then how do you take a learning from a creative test and iterate on that? I wanna throw this to Joanna first and see if she has a hot take for us about creative testing and the and Andromeda. Put on the spot. I mean, it's are you talking about creative testing within the ad set or within the ecosystem? For example, when you want to test a new angle or something. Because for me, from the logic point, Adromeda doesn't make sense because when when I have different headlines talking to different personas, it should be different, right, from the logic of Andromeda, but it's it's the same. It's because it's it's just the headline change. And the last days, I read that even when you switch the creator and leave the script so same script, different creator, it's the same for Andromeda, for for the for the AI. So I don't know what what is, yeah, what is the similarity for meta for for the Meta AI now. That is a great question. I know that they've come out with an a recent I'm gonna try to pull up my document because I definitely cheat sheeted this. I don't know. Alicia, do you have that handy? Because I know that that's been circling around a bit. Maybe we can screen share that and. talk through it. I don't have it handy, but if you can find it, that'd be great. I think the one thing I do wanna comment on is I think you said something about a different creator being creatively similar. I've actually heard the opposite that a different creator like, you want something to be very visually different from that first frame. So swapping out the creator is actually a great way to achieve visual diversity. But, of course, if you're speaking to the same person, then it's gonna target the same person. Yeah. I think it's. also about like, you can't have the same script and then just have a bunch of people say it. And I think that's where a lot of people may get confused about, oh, but it's a different person, and they're in a different room. But if they're saying the exact same script, that's where Andromeda is going to, you know, punish you. But I will say that I think the important thing is that, you know, if you are trying to build off a script that you know works, you are not sending that exact script to a new creator. You are making sure that you have the the reasons why that that ad worked. You really analyze why that ad worked, and you're taking those learnings and applying them to your scale of that with another creator. Mhmm. So they're getting the the gist of it and the important strategic parts, but this person is also putting their own spin on it because they are a different person. It may still be within the same persona, but you know, if there's a different personality and there's a different vibe and there's a different flow. And so I think that's where people can get tripped up of just sending out the top performing script, and you need to make sure that when you are scaling a top performing script that it isn't coming across as robotic with just a different head. Mhmm. And when you don't talk about, like, UGC ad, like, for example, voice over ad or something like that, then you don't have different personalities. It's the same script. Is is it the same approach then? I think I probably wouldn't have the same script for, like, a, like, a voice over video. Like, you should really be writing your scripts to match the format. So, like, if you're talking about, like, maybe, like, a b like, a cinematic b roll or, a a q and a or an unboxing or a demo or a testimonial, all of those really shouldn't follow the same script. You should match it to the format. And then also think about, like, each format also targets people at different stages of their decision journey. So I think at the end of the day, for me, it's like it's constructing an ad in a vacuum. Like, what would make this ad good? And if there's a testing component, like the person that you're speaking to, the messaging angle, the format that you're testing, pick that one thing that is the same, but don't make it like, don't just, like, copy paste it over, if that makes any sense. But I wanted to, yeah, like, let go ahead, Joanna. No way. Yeah. You want me to add on to that? No. I'm just seeing in the comments here that everyone's like, Dara, disagreed with you. I think that there's I think that there's, like, a happy medium here. It's I I think it's a little bit less tangible than we'd all like. I would love to be able to say to you, like, black, white, yes, no, here is the formula. I think we're all still figuring it out a bit, and I do think it also. really depends I'm sorry. What? Oh, I. just heard something. So, Yeah. Go on. yes, I mean, I'm not saying that Dara's wrong by any means. I'm not saying that I'm necessarily, like, the be all end all on this. I think we're all still really figuring it out. And I think that's okay because these things change so, so fast. And I think that everyone thinks that there's this cheat code to understanding Andromeda, and there's not. You have to test within your account. You have to see what is turning up in your account as, you know, Andromeda is can as cannibalizing this because if they think it's too similar. I think you all just have to kind of take the take the advice and then test through it in your own account. Mhmm. There is no bible. Yeah. And I think also sorry. Just on, like, piggybacking off the idea of, like, using the same script. Like, we do this quite a lot where, like, if we see signs that maybe this this hook is working, can we iterate that in a different format? Can we make it into static? Can we actually have it appeal to a different audience? I think with Andromeda, people think, like, oh, no. We have to do, like, five or 10 different kind of completely different creatives anymore. But the shift here is actually you're having to target micro personas. You're having to target specific people with your ads instead of, like, making different tweaks and copy because, like, having something that says, is it worth it to buy this? Or this is now £29.90. It's not going to really move the needle. You're gonna have to think that level deeper of saying, okay. Like, if you're a busy mom, what actually resonates better? Is it, like, 9AM? This is your solution. And then another one for, like, a function like, a, you know, busy professional, you're kind of thinking about your next pay rise or something like that, and you're gonna have to embed that copy in it. And what kind of visual appeals to those audiences are different? So I think was it Dara or Sarah that was talking about generational research, which I think is very relevant here. Like, it depends on your persona as well. Like, who you're targeting as a mom might appeal to different types of formats as well, and that's another way to iterate a message and trying to go deeper into a hypothesis of a persona that you're trying to target. Mhmm. And do you test, like, different structures from the same script? Because you could test different first frames, for example, to to test which is the best first frame with the best hook rate, for example. And but I think this is right now, this is this too similar, right, when I just change the hook visually also and leave everything else the same? Or when I put out some elements, like, I want to test if a social proof makes it better or not. I think with, like, Andromeda AI, they're not it's not as smart, and I'm gonna be roasted by saying this. But I don't think AI is that smart that it would tell you, like, your whole body is the same. Like, it would only probably look at, like, the first three seconds and see whether that's distinctly different. So, like, how we are testing is that, like, you know, we'll have one concept and three iterations of the same like, we'll have different copy, but we'll also have different distinctly different formats for the three seconds, and they'll go into the same body script. So that's. perfectly fine. So in terms of, like, having that kind of distinct different like, it will be one POV hook, for example, and then another one that's like a head talking head POV talking head hook, and that will be very dis like, distinctly different enough for Andromeda to think they're different kind of cop creative, if that makes sense. So you you still test, like, different hooks. with, the same helps with. the same body. Yeah. Yeah. Okay. I like when I think about testing hooks now, it's definitely not like the I'm just gonna change the visual or I'm just gonna change the headline. I really like you're changing the entry point to the sales sequence. So for example, you have one that is a first person recommendation of, like, you know, my wife finally got me clothes that I actually like. And the next one could be, more of a statement or a, don't do this or, like, you know, big clothing brands don't want you to know this or something like that. So I think you really have to think about sort of like, I wouldn't be testing two very similar headlines coming, like, coming at it from the same direction. You're still going into the same sequence, but maybe one of them is a recommendation. One of them is an expert, like or one of them is a personal experience. One of them is more negative. You know what I'm saying? Like, they're Hook have. to have I'm sorry. Yeah. What? Mhmm. Yeah. Like, hook hook types, like. Not just moving the words in the. headline, but making sure. that that. Yeah. headline is a is a different style of of opener, if that if. that makes sense. Nice. Okay. I wanted to add on too because a few people, yeah, mentioned the Dara. I wanted to re bring Dara's slide back up and show that, like, she showed us that pooping every day was her same hook, like, same concept that she was iterating on, but you can see how differently she applied that. And when you think about, like, pooping every day is such a broad hook and such a broad idea that really could speak to so many different people, and she's using every other element of the ad. The person who's in the ad, the supporting text, the visuals, the format, like, every single thing about that ad, like, adds all of that essential, I wanna say, like, flavor to the concept of of pooping every day. So that she's able to, like, pick this as a pillar, but the execution is so vastly different. So it's so I think we said, like, don't copy the same script, but you can certainly take this, like, broad idea that works for you and iterate on that in a very diverse way, if that makes sense. But she wouldn't just copy, like, the the image with the with the product and then the headline, and she wouldn't change it because the the product doesn't change. The offer doesn't change. But but maybe this product solves a different problem also. Yep. So true. Yeah. Yeah. She wouldn't, like, take this ad and just swap out the headline and keep it that way. Yeah. Yeah. Yep. Hopefully, is. relatively help helpful. I think at the end of the day, it's not an easy question. Yeah. It doesn't. it doesn't make sense 100%, this whole andromeda. thing. For sure. Because I think this is the biggest struggle right now in the industry. Like, I truly think that this is the question that we are all struggling with the most, and I think it's a lot of trial and error. I really do. Yep. Cool. Okay. Thank you so much. you. Yeah. Appreciate it so much. Thank you for being our second guinea pig to come up on stage. Really appreciate that. Okay. Awesome. Next question I wanna ask is when you have so many personas from creative road mapping and insights from research, we have to prioritize them based on evidence. So does evidence include your own data or also data from competitors? And is there a strict rule not to test any new idea that isn't backed by evidence? I think the first thing I'm gonna say to this question is, is there a strict rule not to test any new idea that isn't backed by evidence? I don't think there's any strict rules about anything. We always say, like, you don't wanna just test something because you like it. But if you have a gut instinct, if you if you've done all that research and you feel compelled to test something, I think as long as you have the budget for it, you can test it. There should always be, like, a reason why you're testing it, but there's no hard and fast rules about anything. And, yes, the evidence can include your own data and data from competitors. I think that's part of the research is competitor research. Don't just test something because your competitor did it. Like, a lot of people, what they do is they look at their competitors, they say they tested this ad, so I'm gonna test this ad. That's not a good way to do competitor research. A good way to do competitor research is to identify, like, patterns and try to figure out what exactly they are testing. Are they testing speaking to a particular person? Are they testing trying to use a certain format? Are they testing a new visual format? What is it about their creative testing mechanism that you can learn from and take and, like, spin up in your own way. So there is a way to use the data from your competitors effectively, and it's you can totally take data and research from all different aspects of your creative research. I will. say just as a as a little story about this. Back at Hexcloud, one of my strategists, Paige, I call her Dennis Rodman because she always has completely wacky out of the box ideas that are usually way more organic. She's a a creator and an actress and, like, definitely comes from comes at performance with this background of less, you know, data heavy, like, this works. I don't like it, but it works. Let's pop them all together. She comes from a from a much more organic background. And so when she would have her wild out of the box ideas, first of all, I encourage them, but, you know, obviously, you wanna have a percentage. You should have some wild out of the box ideas. You can invent a new format. Like, that's the whole point. We should not just be recycling everybody's format. Like, somebody's gotta invent a new one, guys. So there's that. But then also what I would do is when we're if she has a wild idea or way of doing things, I would say, awesome. Love it. We're gonna have the Dennis Rodman version v one, and then we're gonna have the boring Joanna version v two that uses best practices, and feels more like an ad. And I don't care who wins. Right? We all win. Doesn't matter. I don't need to be right. But what is so good about it is, like, it's very much of a hedge your bets, take a risk. If you are in Vegas, you are not gonna put it all on red. Maybe you will. I'm not that person. You're gonna wanna hedge your bets. And so I think the whole the whole way to think about it is new ideas are phenomenal. Hedge your bets. Your account, your testing should not be too heavy in untested ideas and should not be too heavy in overly tested ideas. You have to have a nice balance so you can keep pushing the needle. And you always wanna be the person and the brand who starts a new trend, not the one who's following, so it's a really good feeling. Mhmm. Yeah. I think there's always been this idea of, like, how much of your creative testing should be iterations and net new ideas. And I think I heard, like, seventy thirty, eighty twenty. I always looked at it like, if our account is in a really good place and we have the flexibility to take some bigger risks, then I'll do that. Now is the time to do it. If our account is not in a great place and we really need a winner right now, I might be focused a little bit more on the things that have stronger evidence and that I can measure more specifically. I also saw someone ask a question about, like, if you don't like an ad, why would you run it? And I think this is something that, like, Dara got a little bit hype about on Tuesday was, like, somebody said I didn't like this, and I said, I don't care. At the end of the day, if it doesn't matter if you like it or not. If it makes money, it makes money, and that's the game. Right? You wanna find a top performer. You don't wanna find an ad that you like. And that's another reason why I think looking at competitors and inspo can be a trap for marketers is because you look at you look at an ad and you go, oh, man. I love that ad. But you're a marketer. You're not the person who's not pooping every day. You know? So you need to be looking at that ad and being like, I'm a person who's not pooping every day. I resonate with this ad. That's the game. Right? Yeah. Okay. Amazing. I'm gonna throw this to Naomi to go through her creative retro, which I think is highly linked to a lot of the things that we've spoken about today. So take it over, Naomi. Thank you. Let me quickly share. So as it is April 2, this is March's retro, which was very good timing. It's a lot of information, so I'll zoom in to each area and how we usually run these. So a bit of background with Space Goods, we have an internal team of creative that work on kind of internal shoots and internal edits for ads, but we also have additional outsource resources. So, like, an agency that helps us as well as well as our influencer team that works with kind of influencers and white listing content. So this is a really good place where all of us come together, and we kind of go through all the data, and we brainstorm for the next month and agree on, like, a focus to kind of anchor our ideas for for the next month. So how I usually like to start is we'll go into the top 10 ads. So these are always structured structured by spend. Spend is king. So we look at, like, what's our top spenders for the last month, and then we'll go through soft metrics. So I've kinda pulled out the ones that we usually look at. So it's hold rate, click through rate, thumb stopping, conference score usually. And we kind of take a look on a kind of zoom out a little bit and see what kind of is the kind of key repetitions here. I know it's so small. Sorry about this. I just realized how tiny it is on on the motion. But yeah. So, like, some things to call out, we'll put it in Post it notes. So, like, cortisol as a theme is new for us. So we're like, you know, this one took up most of the spend. And then also kind of holding up things like what is the highest hold rate, why did this one perform well. Thinking about iterations, this was actually an iteration of this ad, which is kind of the same exact same script, actually. But we reiterated, refilmed it with a different kind of actor and creator and maybe different formats as well. So this one's more of like a talking head, where this one, we had, like, a kitchen setup and kind of having the comparison right in front of you. So, like, we understood that, oh, okay. We actually improved on the ad that we had, but this one still had the highest thumb stop at so in terms of hold rate. But this one still had the highest thumb stop rate. So we kind of went into a little bit of a brainstorm, and this is more of a collaborative collaborative session where we kind of then think about, like, okay. What can we iterate next? Or, like, why did this one do well? This was the only one that had a split screen as the first three second visual. So I always screenshot the first three second visual here so that we can kind of really understand, like, what was the thumb stop analysis here. And then also, like, in comparison, you know, this one is is kind of showing two things at once, and that tend to do well in the previous months, but then it kind of went down, and then it's coming back again. So we're kind of discussing how can we iterate maybe this one with that. So yeah. So then we go into creative insights. So here's where motion kinda shines the most for us. We have all our creatives renamed in a very tight naming convention system. So we kind of do essentially our kind of human based analysis first where we kind of every every creative that we got we create, we do our first run of analysis saying, like, okay. Which persona this is targeting? What's the benefit that we're trying to what what we're trying to convey in this ad? And then we have it all kind of named and tagged. And then through motion, we get to kind of have all these results. So, like, understanding on our personas, so these are kind of what we've named our personas as, like, functional optimizers, mindful achievers, focus hackers. What's interesting here is that mindful achievers are actually our number one persona or, like, what we think our customer group is majority is the majority of. But functional optimizers have kind of risen in the ranks in the last month or two, so we kind of went into a bit of a deeper analysis with Motion AI tags. So understanding messaging angles as well and how they differ. So, like, here, you can see, like, functional optimizers improve sleep mood and stress is, like, a major angle that's working for us versus, like, what you can see in Mindful Achievers and Focus Hackers is, like, discount and trial. So here, we kind of made the deduction that maybe, okay, it's kind of looking at quite bottom of funnel heavy in Mindful Achievers. Should we kind of look into more top of funnel ads for Mindful Achievers for this month? And then focus hackers, you can see it's, like, one angle that's really doing most of the heavy lifting versus the rest. So then we kind of did a little bit of a brainstorm here. Like, can we iterate on this? Like, what other micro personas that we aren't we aren't thinking about in focus Hackers that we should kind of dive deeper in? And I know there's different terminologies, and I probably mixed it up as well by calling it micro personas, but we call it niche customers as well. But they're basically the same thing. Right? It's talking about if you've got a busy mom as your, like, main persona, you've got different types of busy mom. You've got a yoga mom. You've got a mom that goes that's a working professional as well. You've got a stay at home mom. You've got a remote worker mom. You've got many types of moms. So how can you go even deeper into one personality or per persona? So we call them niche customers. And then what we do here is that we compare it without with the AI tags. So this is, the AI targeted audience with motion. And here is just kind of a good sense check for us to see whether we're actually missing out on a specific type of persona or type of people that we're actually targeting already with our ads and maybe have a discussion around, like, is this actually a bigger kind of question a bigger pool of people that we thought, and we should maybe go go harder on iterating on them. So, like, here, could see social and event goers, for example, is one that's really, like, one that we probably didn't even think about versus the ones that we have. And is it, again, worth it to to iterate on that for now? Maybe not. Maybe it's like, oh, we should maybe think about neurodiverse individuals first. So this kind of gives us, like, a really good gauge of, like, just a a wider eagle's eye view of what we should focus on. And then going back to core benefits as well, this is what we think about, like, what our product actually serves for the for our audience. And we we this is also, again, tagged by us. So we kind of think the way we tag this is always because with videos, you kind of go into many benefits and stuff. So we always think about the first benefit called in the video because we know our attention spans are not not long enough to watch the full video usually. So we're always thinking about the first first core benefit that's called out. So then even when we iterate, we can say, like, okay. Well, if energy is the main core benefit that's doing the most spend, like, can we iterate this video that talks about energy right at the end but has a really good thumb stop rate? Can we maybe reiterate the script in a way that's we're swapping out energy to the to the start? And maybe that kind of appeals to the audience first or sooner, and they would probably stay longer or convert. And then we go into a little bit of customer insights. So I like to change this up every month because I don't want it to get boring for for our team. So sometimes I would pull in data from, I don't know, a kind of a call a conference that I went to or something that I've learned or like, last month, we were we were thinking about time based purchases. So, like, we looked into data around our customers on when they bought our products and seeing whether, like, you know, would that kind of affect people's mindset. So, like, most of our products were bought being at night. So people are probably doomscrolling, in their bed, tired, not bothered about the next morning, but then all our ads are talking about that morning moment and, like, how good it is and how refreshed they are and everything. So we were thinking about, oh, should we maybe not talk about, like, having a better sleep cycle, like, well and kind of getting ready for the morning in a way that's more natural for the customers that are buying it in that moment? So this this month, especially, we were looking into more of a brand dip because we were really confusing ourselves, I suppose, on whether we are coffee replacement or health product. And I don't think it's either or. I think we're just kind of understanding, like, how we're kind of positioning ourselves and whether we should kind of lean into one or the other. And we kind of discussed around putting in claw into Claude, not to mention AI too much. But we've kind of chucked in our post purchase survey data and our kind of reviews data and understanding what motivators and barriers people are kind of addressing when they bought us. And we've kind of tried to see what what is the main things that came out and just kind of align on what we might have to address, maybe forwarding it on to NPD team as well. So, like, sweetness sensitivity, is that something that we should maybe talk less of, or maybe we should call out more flavors that we have because we've got, like, multiple seven I think seven to eight flavors now. So maybe we should address that more. And then this is just another additional, like, other things that I sometimes call out depending on kind of efficiencies and stuff. It's just, like, how we kind of split out sources and stuff and and kind of call how we talk about, like, ads in total. So, like, how many ads the ad performance in the in this oh my god. I can't speak. The ad performance in February and what are the top ads, but then also what are the top ads that have launched in February. Because sometimes ads do take a while to kind of scale and to kind of rock it off. So it's also good to understand the the the comparison there. Like, okay. This ad is doing really well this month. That's a new up and comer, essentially. So, like, maybe we should have pay more attention on how to iterate that or, like, kind of get ahead of the wheel of knowing, like, okay. If gut health is coming up again as a topic, we can kind of prepare ourselves for the next month to type make sure that we have enough content in pipeline for it. And then this is kind of tapping into what Alicia was saying about, like, looking at competitors but in a in a more structured way and how we kind of look at examples. So I always kind of start with, like, okay. Here are our long standing top performers that have been around for a while. What are the key messages? Like, what is it? Another proof proof message there. But what are the key messages that are, like, consistently coming up again and again for for submit space goods? And then I kinda structure it with formats to iterate with. So, like, what what formats that we might have not tried or that we can iterate with our best performance with. And then we go into, like, the two core pillars that we wanna focus on for the next month. So, like, this month, for example, we're looking at anxiety and jitters as one kind of pillar or, like, key core benefit, and then the other one being focus and productivity. So then some some ads make sense more for one versus the other, so we kind of break go into it, kind of share the ideas. This is more collaborative, so I'll give time for the team. I'll probably usually send this retro, like, three days in advance so they have time to actually watch everything and then add the ideas here, and then we kind of do a sharing session from it. Yeah. And that's it. That's That's the we're aboard. I'm gonna stop sharing. I saw a few questions come in, so I'm gonna ask them on behalf of the people in the comments. One of them actually, my question one of my questions was you said something about people usually buy it at night, but you're talking about the morning moment. How did you determine people usually buy it at night? Is that through, like, Shopify data? Yeah. Interesting. Okay. Cool. So you're looking at, like, when people are actually buying the product typically. How do you is there a way that you're mapping that to how do you know that people are seeing the ads and buying at night and not seeing it somewhere else? That's a good question because I know, like, you know, there's always that data set of, like, what, seven seven the frequency of seven, isn't it, before people buy? Yeah. But I think if they are purchasing, it will probably be more of the bottom fund. Like, it won't work for all the ads, but I think it's kind of you guys think of the bottom funnel ads, like the ones that are discount heavy, etcetera. Maybe we need to kind of lean into more of that moment of when they're buying. It's usually maybe, like, a discount that's ready to go or, like, that's that's ending soon and stuff. Mhmm. But I agree it won't it won't work for everything, Yeah. and you still need to have a a diverse set of creators and stuff. But yeah. Maybe that's a maybe that's a good good example of, like, you have data. It's not necessarily, like there's not a clear line written from one to the other. It's just a hypothesis that you're kind of, like, putting together based on seeing two things that are you can't guarantee the correlation, but it's something you're gonna test anyway. Right? Yeah. And I think that third pill that customer insights I said, like, I I always change it every month. Like, I don't want to, like, always look at at data of, like, when people are buying, but I wanna change it up. Like, this month, we're looking at, like, are we a health product or coffee replacement? Or the next month could be, like, are we are we scale like, you know, are are flavors the right flavors for us? Like, what kind of flavors that we should shout about the most, like, one's bringing in more sales. If it's coffee flavor without sweetener, then maybe we should target we should make our drinks black instead of with milk or something like that. I always try to, like, have something different that's kind of a bit more of a spark sparking a conversation. Mhmm. And is that an example of prioritization? Because you just listed, what, like, 10 different things you probably could have tested, but you you probably had to pick the one or two that month that you felt had the most evidence or that you felt most compelling to test that month. Right? Yeah. I think with this month, especially, because we we've launched in retail recently, and we're kind of deciding, like, you know, we in in some retail stores, we're in we're in health aisles versus coffee aisles. And then also with this time period, like, budgets quite as it budgets isn't everyone's budgeting, everyone's saving. So the the the mindset of saving money is quite in forefront of everyone. Mhmm. Well, not every again, hypothesis. But yeah. So we're thinking, like, okay. Should we kind of consider marketing more more into the health supplement angle of, like, saying that this is more kind of beneficial for you rather than saying a coffee replacement because, obviously, coffee instant coffees are cheaper than us, and, like, we're got we've got so much added benefits. Mhmm. Yeah. It's it is definitely, like, rooted in this time of the month and other kind of signals that we're seeing. Yeah. Amazing. Lots of people are asking you how long did that that research take you? How long did that that retro take you? We I've been doing it monthly for yeah. Ever since I started. Like, I would usually put it together the week before because you kinda need to be quite on it, like like because it's a wrap up of the month, isn't it? So it's everything's kind of quite quick, I suppose. And with, like, Claude, with Motion, it makes it quite simple. I don't know whether that's the right answer, Yeah. but yeah. Did it take you, like was it, like, a full day deep dive? Yeah. I would say maybe a day, but I would do it usually across a few days because I kinda wanna sit with it, kinda. go back into it, especially with the brainstorming and the kind of the the focus for the next month. I would probably wanna dive into the data as well. Like, we're a subscription brand. So some of the data that I would always look at is, like, retention numbers and kind of retention quality. So without tagging system, we can also kind of track our customers in the fact that, like, okay. Which customer is actually staying for the longest? And that would also tell us, like, okay. What's the highest quality customer for us is, like, this angle, for example, and then we should maybe focus more on that one. But yeah. Yeah. I think also, not to jump in, but to jump in. I think also, like, yes, this is kind of intimidating with how much work this is. And what I find realistically when you're on a team is that depending on how well staffed your team is, we all know they're not well staffed, Mhmm. you often feel like you don't have time to do something this intense. And I think that the way Naomi is talking about it, doing it monthly is really important because I've been on Teams where we try to do it weekly. We try to do it biweekly. That's a joke. Like like, that's like that's like if you were given a document on, like, how to be a creative strategist. Like, that would be what it says. In reality, ain't nobody got time for that. So, like, what usually ends up happening is that you should be touching base with your data every single day. Don't look at, like, that day necessarily because that often depending on the time of day, whatever. You wanna look at, like, probably, like, a week to two weeks, but keep an eye on your data every day. New ideas and new strategies are gonna come to you all the time. And I've set, like, weekly mini retro meetings for my team. So, like, on Mondays, we all come together, and we half ass it. And it's not really half assing, but sometimes putting together that beautiful thing that she put together is not worth the time unless you do it in very specific time periods. Like, monthly is good. It is unrealistic for you to do this weekly. And if you try to do it, you will cancel that meeting every single Monday. So really just try to do it on the fly, know the basics, have, like, a checklist of what you're kind what you really want to keep an eye on, but don't pressure yourself to do these massive deep dives every week. You do not have time, period. Let's. all be real. I like that. I like to think of my Miro board as, like, a living, breathing document, which is such a corporate TikTok y phrase. But, like, the way that I think about it is something that's constantly open. I have motion in my Miro board open at all times, and I'm constantly, like, adding and reaching into it. And then, yeah, monthly, I'll just kinda, like, go through. I'll, like, set aside that time block to, like, really put some intention behind it. But for the most part, I'm just shoving stuff in there, and it's not very organized until I make it organized. Yeah. And one last thing is what is is the importance of sprints, like, to Joanna's point. Like, there's I know some teams, like, struggle to be like, oh, we have to create creative and launch it every week and da da da da. But, like, we we try and be very strict with our sprints. Like, we only do a new in house shoot one like, once a month or every other month. Mhmm. And we spent time, like, really planning that out because, like, you know, we know our resources. We know what we can create in a week. There's no point trying to rush into, like, creating things just for the sake of creating things. So I think it's important to also push back and have that proper system in place. Like, it doesn't have to be, like, how you create content. You can do it, like, biweekly sprints. But, like, I think having a sprint is important to give yourself some space to breathe and actually do that research properly. Mhmm. Yep. For sure. Okay. I think what I'm gonna do is I'm just gonna throw it right to Joanna for your part talking about your budget doc, like how to align your creative road map with your budget, how to devote your budget to different initiatives and to your what did we call them? Diagnoses. So you go ahead and take that away from here. Awesome. Let's see if a grown woman can screen share. Hey. Alright. I can zoom in more. How are we feeling about this size font? I'm not actually looking at the comments. Great. We're gonna assume that it's great. Okay. So this is building off of what everything that Dara said on Tuesday. And I like to to think of how you're prioritizing what you're going to test when. I like to call them pillars. So a pillar is a learning objective for the quarter. It's the question you are committing creative budget and production time to answering. So this can be a persona, where do young professionals actually convert, a format you wanna test, maybe does elevated, the elevated videos work? Can you have it outperform AGC? Does it have a place in your in your ecosystem, or a moment you need to capture? Say you have an a tent pole event, like a Mother's Day sale. You have to focus on that. So your intent is to test the most important things that your account needs to learn now and what is most likely to work now. So I like to think of this as if you are going to test, a fitness persona in November, we are eating in November. That is not fitness time. What is fitness time is January, February when everyone gets those gym memberships for five seconds, throws their money away, forgets about it, and cancels it in a few months. That is when you should be testing a fitness persona. So you need to make sure that you are thinking seasonally, thinking about the trigger moments, thinking about when a particular, persona is going to purchase because not every persona has the same schedule. So step one, pick your pillars. This is gonna be a first pass because oftentimes there is, like, a champagne, taste, beer budget situation. And, so you're gonna wanna try to pick three to five pillars based on your diagnosis of the account, based on everything that Dara has taught us on Tuesday. So this is the first draft. You are going to reality check this later because I guarantee you, you're gonna wanna test more than you can afford. So once again, to just recap what she said about diagnoses, it's one to two sentences explaining why you're testing what you're testing. So examples would be all of our top performing content is focusing on one persona. We need to open up a new audience. Or all of our top performers are bottom of the funnel, CPAs are rising, so this quarter we really need to focus more on top of the funnel to refill the pipeline. So let's say q two pillar examples could be Mother's Day sale, elevated videos, and young professionals, like I said earlier. So for this quarter, I'm focusing on converting the young professional demo, testing it to more elevated videos like hype reels or ASMR, and carving out a chunk of my spend and my time for Mother's Day because I feel like what happens with a lot of teams is you forget about those tentpole events that you are committed to creating those sale assets, and then you get really excited about your evergreen road map. And then you realize that you have overcommitted yourself because sales take a lot of time. So definitely be realistic about this. So, again, like I said, this is a starting point. We will trim this as we go along. So this is when you reality check your budget. Your budget is your meta prospecting spend, which is how many dollars are going into testing new things versus scaling what's already working. So just a little cheat sheet. Testing is finding new angles and proving them to the point of statistical significance. Scaling is keeping winners alive and stable while you look for that next thing. So everybody wants to know how much do I put in testing? How much do I put in scaling? Depends on so many things. There is no final answer. What it really, really depends on is your account health. Where are you right now? Are you in a situation where you have a ton of evergreen content that's crushing it and you just really need to focus on scaling it and making more and more? Are you in a situation where nothing is working? Are you in a sale period? They all that is going to determine what your split should be, not your you know, what someone said on LinkedIn, unless I said it on LinkedIn. I'm kidding. Okay. So I would say, like, your default here is probably going to be around $70.30, $60.40. This is most quarters. So this is how you are funding your learning across the three pillars that you've chosen while keeping a solid chunk of your spend dedicated to winners. Again, these are not bible. These are honestly a little bit of vibes. These are kind of just keeping you honest. This is not like, please don't sit down and count your ads and make sure that, like, I have to do 2.5 ads for this. No. It's okay. Just key it's it's the when you're bowling, it's those the the thingies, the buff the bumpers. Hey. There we go. So let's say you're in a situation where you're a new brand, you're a new product, you're opening up a new audience, or the account feels stale. That's when you're gonna put much more of your money towards testing. You don't really have that much to scale. And I think a lot of people also try to scale things that are working because they're working the best compared to other stuff in your account. But if they are not great, they are just better than your other stuff, please don't overly scale them. That is a waste of your money. And it's very tempting because you wanna feel like you have a winner and you're doing something with it, but, like, how much of a winner is it? So let's say you are proven. You have a stable, account. You have all these these winners that you wanna scale. So you're gonna put more of your money towards scaling, and you're gonna still keep that testing flow going. You never wanna not be testing. That is really scary. Please always be testing even if it's a small amount of your budget. You always need to be expanding your audiences and your messaging. Now if you're in a promo window, let's be realistic here. You have a lot of ads that you need to do for your sale. So this is maybe more of, like, a fifty fifty. Evergreen ads are still very, very valuable during your sale period. They are still doing a lot of work. Often, when I've had sale periods, our evergreen ads that already have momentum and stuff, they are and they've already been tested, they're doing a lot of the work. Your sale ads are really bottom of the funnel. Just, like, do it do it now. So, and then but then if it's, like, a huge promo, like Black Friday or holiday, probably gonna end up being, like, more 90% on sale ads and really cut back on those evergreens. But, again, it depends on the size of your your promo and, you know, how much your revenue depends on q four, if it's, like, 80% of your revenue or not. Okay. Now you also need to think about reality checking your bandwidth because we all like to work ourselves to death, and it's not sustainable. So bandwidth is your team's time and production capacity. How many concepts can you actually execute this quarter, and which ones need a shoot versus which ones you can turn around fast? We like to call them heavy lift and light lift. So when you are testing into a pillar, this is not one concept. If you say I tested gifting, and I say, how did you test gifting? And you say, did one UGC where a woman in her twenties talked about gifting. I go, okay. Great. You tested one concept. You did not test one pillar. A pillar needs multiple creatives, multiple parts of the funnel, multiple formats. So not just, like, static and video, but, like, UGC or elevator. Like, you have to really think about the creative diversity that you're applying to the count as a whole and use that creative diversity for each pillar. That is how you know that it's a fair test, and you wanna be able to walk away at the end of the quarter confidently saying, we tested this, and here are our learnings. You can't do that if you get super spread out all over the place and try to test a billion things. You will walk away with hunches and vibes and thoughts and feelings. So each pillar needs a minimum of three to four distinct concepts to get a fair test. So that means that if you only have capacity for six concepts total, you should not be running five pillars, you should be running two. So that's when you have to kill some babies and trim your pillar list if your bandwidth doesn't support it. Also, this is something I really want everyone to keep in mind. When you are thinking quarterly, the end of the quarter doesn't fall off a cliff. There's another quarter that follows it, unfortunately. And you really have to think about the fact that, let's say, you know, q two ends June, but you have a July 4 sale four days into July. So you need to think about my June bandwidth bandwidth is also going to be preparing for July, for q three. So all these quarters are connected because I've done this before, and you wake up on, like, the first day of the new quarter, and you're like, well, f me. I crushed last quarter, and now I'm screwed. Sorry, guys. I just I really can't talk professionally. I tried so hard. I don't try. I lied. I don't try. Okay. Plan. Step four. This is where I said, the creative diversity per per pillar. I kinda break it up here. I really jumped to that too early. Apologies. And then this is how I put together just a little table here for you to kinda think about. This is my product. This is this pillar for this product. Here is my hypothesis about why I am testing this pillar. And then here are the three formats I am going to use as a minimum to test this pillar. So you wanna make sure you are noting the part of the funnel. You also need to think about your production needs. So making sure that you are not doing three concepts for one pillar that all need major production. You should be thinking about scaling or adapting from other pillars that you know work. You should be thinking about where are some shortcuts and making sure that you're allocating your resources wisely and not getting overexcited. She's done. Thank you. Everyone's loving how you keep it real. Okay. There's one question that keeps coming up a couple of times, which is a bit of a high level question. But what is the difference between a pillar, a concept, and an angle? Ah, okay. A pillar is a big old bucket. Well, I mean, it's a pillar, but it's a big old bucket. Right? A pillar is let let's say, like, it's pillar is a persona, young professionals. A concept is I'm gonna do a UGC with a young professional speaking straight to camera, talking about why whatever product I am making up in my head is good for young professionals. An angle is what is you know, honestly, but that's when you get into the ad set. What is the angle that I am going to come at this concept? So maybe that angle is, I've tried everything and everything's failed. Now I found this. Maybe your angle is, I am an expert in this, and I'm gonna recommend this to you because I know what I'm talking about. Maybe your angle is is I don't know. There's other angles out there. So, really, it's pillar is big, concept is medium, angle is small. Mhmm. I love that. And then Tina's asking about, like, what is a big idea? Is there is there something that you refer to as a big idea? Is that a term that you've heard heard of before? Did I say that? Nope. She asked about it. Big idea. I have lots of big ideas. They're fun. I mean, Yeah. I don't I don't think we, like, we use that term frequently as, like, an exact creative strategy term, if that makes sense. We hear, like, pillar content concept, angle, format. Those are the ones. And I think what we need is to add to that glossary of terms of what those actually mean. Okay. I'm gonna ask you a couple of questions that were actually student questions that are related to this document. The first one is when you're creating this road map, are you creating for the brand something for each individual SKU that the brand is selling or something for each individual persona that the brand is selling? Like, how do you how do you decide how to define those things? Wee. Here we go. Two by product. Very important. Every product has its own value props, has its own angles, has its own personas. So for example, when I met, let's say, doctor Squatch. Right? They sell soap and they sell deodorant. When I was working on that account, there's a lot of adaptation across there, but those are two different products, and you have to treat them differently. They both need as much love as your brand and inventory has determined. So, you know, if it's your hero product, that should be you know, you should be allocating more spend to your hero product. Mhmm. I think that definitely the pillars go underneath products, but make sure that you are remembering that you don't need to make ads for every product. I've seen people do that. It will kill you. And it's not the best way to spend your money. You really have these products that are intro products that hook you in, and then there are products that you scroll around and you find yourself. Those are the ones that maybe aren't as profitable or you don't have as much inventory in or they're not seasonal. You really don't need to you need to think about what are your priority products and then drill down your pillars under there. Mhmm. And then if you only sell one product, you just pretend that product doesn't exist and start from pillars? Yeah. Yeah. And then your life is so much easier. So congratulations. It is easier in one sense. In another sense, you're only selling one product. one time. Hopefully, you have a subscription. But for the purposes of of my five minutes here, much easier. 100%. I guess it's another nuance that we we just won't even dip our toes into. Okay. And then one one other question that came from Larissa that I had in this deck was, what if we really don't have a lot of budget to spend and one particular pillar or persona is working? Is it still should we still make an intention to, like, set aside some budget test new personas, or should we lean into that one that's working? I think definitely lean into the one that is working, but you are never going to actually truly be able to scale your company. unless you are unlocking new audiences. Mhmm. So I think that the most important thing for you to think about is likelihood or likelihood of value of that audience or persona and size potential size of that persona. Mhmm. So if because you have that one persona that is working so well, you have to make sure that, yes, you you are gonna allocate some of your budget to testing, but you better be testing the absolute most potentially profitable moving the needle, new persona or new direction. So you have to just be very, very picky. It cannot be your personal taste. It cannot be, like we say, vibes. It has to be very much what has the biggest potential to blow up the account. Mhmm. Otherwise, when you run out of that persona, the one that's really working for you, when your creative starts to fatigue, when you've kind of hit the you've sent ads to, like, everybody in that demographic, When stuff starts to fail, you have you then have to start from scratch with your research. So you would always make sure that you're one step ahead of creative fatiguing. Mhmm. Because when it fatigues, you do not wanna be caught with your pants down. You wanna have put your pants on beforehand. and be ready for the fatigue. It's probably a good good opportunity to use that gap analysis that Dara was talking about and find your. highest opportunity persona that you haven't been speaking to. Right? And, like, focus on that one first. Absolutely. Yeah. Okay. Amazing. Okay. Well, you wanna get off, my. And there is there is a question I feel like I can I can answer a little bit? Pillars can definitely be repeated. You should be retesting stuff. Like, seasons change, years change. All We thought 2022 was the worst year ever. Then 2023 came, it was worse. 2024 came, it all got worse. So you do need to continue to test personas. The people are not the person in 2022 is not the same person as in 2026. So definitely come back around and always be retesting. And then once you found a pillar that works, it goes into your repertoire, and it gets repeated as long as it's seasonally appropriate. So that just goes into your scaling, budget of now we are scaling six pillars instead of four pillars that we had before this quarter. Mhmm. Amazing. Okay. Cool. Thanks, Joanna. Appreciate it. Alright. Can I go back to sharing my screen now? Unshare, please. Get off my stage as as you said an hour ago to me. Okay. Amazing. Okay. The next thing I'm gonna cover this time is a question that we got a lot on Tuesday during Dara's session, which is if you're a new brand, if you haven't been collecting reviews, if you just don't have reviews at all, what do you do? So I've come up with a workflow that's a bit of a workaround for you to help find personas if you have no reviews. Now the first thing I'm gonna cover is for b to b, if you don't collect reviews, and Motion is a great example of this, it's not like we collect reviews and, like, post them on our website because we don't sell a product. We we sell a service. So I get this question all the time. If you don't have reviews, what do you do because you sell to businesses? I had someone much smarter than me on my team named Faye who, collects a lot of data from customer calls. She hooked me up in Claude Cowork with some kind of connector that goes through Hubstot HubSpot, that goes through Gong, that goes through Notion. Anyway, I'm able to query all of our sales calls and our customer success calls so that I can reference customer conversations and do the same research, but with conversations instead of reviews. So if you have the ability to access that type of data, if you record your sales calls, if you record your customer support calls, if you record your onboarding calls, if you record, like, anything at all, use that as an equivalent to your reviews and ask similar questions, or you're just gonna have to expand into other places. So maybe there are LinkedIn messages. Maybe there are Slack communities. Maybe there are Facebook groups. Maybe there are blogs or forums. Maybe you have to dig into Reddit. You're gonna have to go somewhere else for that information. But with that being said, if you do sell a product and you only have, like, 50 reviews, really, what what you should have learned from this whole week and from a lot of these different sessions is that reviews are absolute pure gold, and you need them. So if you don't already have a post purchase ask for reviews and a post purchase survey, that's very, very valuable. And that's something we're setting up, like, today, if you don't already have it so that you can encourage people to leave those reviews knowing that they are going to serve you very, very well in the future. That being said, if you do if you don't have reviews for now and you're just starting out as a brand, there is a fix. You can borrow from your ecosystem. So what I like to say is that your customers already exist. They're just buying from other people in other places. And so you can go to those other people, those other places, those other products, and you can borrow from their data. It's not gonna be a direct one to one, though, which is why the first thing that I recommend you do is do your research about your own product first so that you have something you can measure yourself against. So before doing anything else, understand what makes your product, your brand different from anyone else on the market so that you can look at everybody else's reviews, look at everybody else's data through the lens of your own brand and product, and identify those gaps effectively. Otherwise, you're just gonna copy exactly what they're doing for their product, but you're not selling their product. You're selling your product. So really deeply understand what you're selling and why you're selling it and who you're selling it to first and create that document to inform all of your research. So I do have a prompt for that. I will share some prompts after the fact, but that is the first document you're gonna wanna need. Then the next thing you're gonna wanna do is you're gonna wanna find a direct competitor. So for this particular example, I'm just gonna skip ahead here. I invented a product called Mute, which are ear like earplugs. But instead of, like if you thought of if you've heard of Loop earplugs first of all, I don't know who hasn't heard of Loop earplugs. They've targeted every person under the sun who needs earplugs for any given reason. But if I came up with a new brand called Mute and I really wanted to make these earphones earphones, headphones, earplugs, earplugs. I don't even know what product I'm selling. For people specifically with sensory sensitivity and that's it, then I'm gonna compare myself to Loop. But first, I'm gonna understand what makes my product different. So my direct competitor is Loop. That's a great resource. The next one is an Amazon comparable. Go to Amazon, type in sensory sensitive earplugs or whatever it is you sell. Find what people are going to Amazon to buy. Because the thing about Amazon is when people go to Amazon, they're not gonna type in loop earplugs. But when they type in sensory sensitive earplugs, earplugs for concerts, earplugs for sleep, They're gonna find loop earplugs. They're going there to solve a problem, and it's brand agnostic. So the reviews that you're gonna see on Amazon are very different. These are people who came here to solve a problem. They are not precious about your brand. They don't care at all. They want it to show up on their doorstep for cheap, and then they want it to work. The So reviews you're gonna get there are just have, like, this different flavor to them, which can be very informative because those people are gonna be very honest about the gaps that other products are not filling that you could fill. And then the other thing is you could go find adjacent brands. So for my made up ear ear dear god. Earplugs. I should have picked a different product. For my made up earplugs, for people who are are sensory sensitive, they might have also purchased something like a weighted blanket, a hatch alarm clock, and, like, a fidget product. So if this is the type of person that I'm speaking to, if I understand what other types of products they might have purchased for a similar problem or circumstance or use case that they have, I can also go to look at those brands. And then what what I've done is I've written a series of prompts. So I have my document about my brand that defines exactly what product I'm selling, who I'm selling it to, and what makes it different, and what our, like, intention is behind behind selling this product. I went and scraped a CSV of reviews from the Loop website, which I did with Claude. I also scraped reviews from three Amazon listings for other earplugs that came up when I searched sensory sensitive earplugs. And then I also scraped reviews for these other products. And then what I did was I went to Claude, and I used some prompts to basically get some information. So the first thing I did was I gave it the brand document that I created. Here, I'll scroll to the top of the conversation. I gave it this brand document that I created. This looks like a like a very complicated I hate when I look at things like this. I it really looked like a PDF before, but this is the brand document that I made. These are the the reviews that I got from Luke. I gave it to Claude, and then I gave it a prompt that basically says, here's information about my brand. Here's some reviews from competitors. Find me personas, who's actually buying the product, micro moments, specific situations or daily moments that show up repeatedly, triggers that made them buy emotional sentiment, language, and framing, but most importantly, gaps and opportunities that that tell me what my product fills that this product does not. And then add ideas and what it spat out for me. And people have asked me before, like, how did you make Claude do that? It, like, gets very visual. I truly don't know the magic of Claude. I didn't tell it to do it this way, so it might not show up like this. But it makes it very visual. And then what I did was just I said, can you make this an a a file that I can download? So now I have this whole document here that outlines. We've got some personas, so, like, the neurodivergent daily warrior, the misophonia sufferer, the sensory overload parent. We've got some micro moments here. We've got some triggers. We've got an emotional arc. We've got some language and framing, and then we've got some gaps. So in this particular instance, it identified that fit failure is a problem. So I ordered loop earplugs once, and I think it came with, like, three different sizes. But for some people, that's not enough. They they still, like, fall out of their ears or they feel too tight. Sleep is an unmet use case. So I think this is because when I went and I grabbed the reviews from this one particular type of loop earplug, it wasn't the sleep one, so there weren't as many reviews about that. There's no tip refill system. That was kind of interesting. When you buy the loop earplugs, you can't buy new tips, apparently. Or maybe you can, and I just didn't see that in the reviews. But that's what people wanted. Clinical credibility is absent. Mode differentiation, skepticism. So, like, you know, how, like, loop says that you can, like, switch into different modes. People are very skeptical of that. They really only care about that one use case. So there's lots of good information here. And then what I did was I did the exact same thing, but for the Amazon comparables. I gave it, again, my brand information, the reviews from the Amazon comparables, and it did a very similar thing. So it found, this time, it made it, like, very pretty. Again, I don't know why Claude does this. I don't tell it to. It just gets visual. It's very magical and fun. But it identified, like, pain points, objections, transformations, gave me all of this valuable information. And then the last one I did was the exact same thing with other products. And, again, I'll share these prompts with you guys so you can grab these. And it gave me another document. And then what I did was I just put all of these documents back into Claude, and I told it, here are the three gap analyses that I did. Here is the information about my brand. Consolidate this. Identify all different kinds of patterns and spit me out the the strongest ad concepts that I could work on today. So then it made me this final document. So it did this, like, cross source pattern map. So it mapped all of the patterns between all of that information that I gave it, and it identified that the recurring personas are neurodivergent daily professionals, side sleepers, quietly overwhelmed people, and caregivers. These are the recurring micro moments. So people in an open plan office, this sounds like an absolute nightmare to me. Imagine being in an open plan office and being like, you know, like, you're sent like, sensitive to everybody else around you. That's quite interesting. Grocery store, I've totally felt that one before. 3AM wake ups, family dinner. So these are all of their micro moments. We've got emotional states. We've got that recurring language. They fall out, regulate my nervous system. I was skeptical. Recurring purchase triggers. We've got some differentiators. So, like, FIT is the differentiator. Medical grade certification is a differentiator. And then it told me, like, who to prioritize and what ad concepts to start with. So I specifically asked for problem aware ad concepts, and this is what it gave me. So now I have like this ad concept for a neurodivergent daily professional. So as you can see, like, all I had was information about my made up brand, and I went and I used all of the information on the Internet that I could find, and it gave me something to start with. So again, this isn't like a one to one comparison. Of course, you have to take everything with a grain of salt, but this is a great place to start. So, yeah, just wanted to show that workflow and how I would have done that with no reviews. Okay. Lastly, just before we go, you wanna talk about gut checking AI? A few people asked, how do you make sure that AI is giving you an actually good outcome and not hallucinating and making stuff up? So, Joanna, you had a document for us. Did you wanna start with a couple of tips about this? Sure. I know that the documents are the key to everyone's heart, so, I'm now document girly. Just give the people what they want. Okay. So let me share really fast. I'm just gonna, oh, I can't share. I lied. you go. Now I can okay. See? You got me back. I feel better now. Okay. Oh, she can do it. She can do it. Here we go. Look at her go. And then it's like, my budget, my taxes pop up. Yeah. Okay. So let me zoom in a bit for you guys. That was too much. Here we go. Okay. So I made this little little doc doc. Thank you, Claude, for being so pretty. This is kind of a recap of what we talked what Dara talked about on Tuesday of, like, how we used to do it, how AI does it. I'm not gonna go into that. But what I am gonna go into right now is ways that you can check to see if the bullshit, you know, tester. So the first thing is ask for receipts. You want to make sure that their assumptions are coming from real quotes. And, also, I love when they I'll say, like, give me the receipts, and it will it will give you the source link. And you can check yourself and be like, alright. You misinterpreted this. This was the onion, you know, whatever. And, if it can't produce it, then this is not the persona is not grounded, and it's just hoax. The second is checking the sample size. I have gotten burned by this before where, you know, back in my chat GPT days before I learned to move to Claude, they you know, I'd throw in my ad account info, and it'd be like, this is your top performing persona. It's dominating the account. It needs to be everything. And I go, ma'am, how much spend was on that ad set that you're so jazzed about? And they're like, $200. And I'm like, that's not statistically significant. You're very wrong. So you wanna make sure that they are pulling from a statistically significant, you know, spend or, number of of reviews or anything like that. Check a second AI. Paste the same raw material into, like, chat or Gemini. Ask for the same output and see what comes out. You know, if they're all agreeing, probably correct. If they disagree, you need to start going in there and figure out as the human, you know, where where things are going wrong. The next is checking for a trigger. So a real person like Dara was saying has a trigger the moment that pushed them to start looking for a solution. So cares about health is your demographic, but just found out her kid has eczema and CVS doesn't carry stuff is the trigger. So you need to make sure that are you just glossing over and saying generalizations, or are you basing this like, found a review where this mom talked about this eczema CVS situation? And the next that I also have gotten very tripped up on because I know, like, at bird dogs, it would start, like, spewing out information for pants, and I'm like, bro, I'm talking about quarter zips. And so you wanna check for SKU blending. You wanna know that it is focusing on the reviews and the information for that particular SKU, not for the brand in general, not getting confused because in your last chat, you were talking about a different product. So you really wanna understand that as well. So now what everyone really wants is the prompts. So for each of these, check, you know, points, there is a prompt here that could help you paste it right in and call them on their bullshit. The end. Amazing. Great. Naomi, did you have any other anything else to add about AI gut checking? How do you how do you know that AI is giving you good outputs? Yeah. I think that I kinda mentioned it briefly in the retro as well. Like, we we use our naming conventions, and we've we we tag them ourselves. So we do our first gut check, and then, like, we kind of cross reference that with how AI kind of sees our ads as well. So, like, going back to the Andromeda, it's like you're you're having to use AI to check AI, but you're still talking to a human in the end. So you're kind of having to, like, cross reference that, but also use your gut to kind of make sure that you're you're saying the right thing. Mhmm. Yeah. Yeah. I think there's one thing that we we chatted about a couple days ago as well, or maybe it was yesterday. They all blur together. But, like, I was saying, my hot take is that when you're going through an AI workflow, any workflow really before you AI automate it, it's worth doing it manually to build up that kind of, like, gut instinct and knowledge of your own branded products so that when AI gives you an output, like Joanna said, she's like, you're talking about pants, but I asked you about quarter zips. Or I asked you how much spend did this go behind, and you're telling me $200, but we spent 200 k a month. That's not significant. Like, you need kinda, like, that understanding, that fundamental understanding of going through the process manually first that when AI gives you an output, you can identify, like, little little things that are like, maybe this isn't correct or know what questions to ask. Right? So that's, like, my biggest tip is just just know your job and and be good at it so that you are able to to hold AI accountable instead of AI, you know, you being at the mercy of AI. You know? Yeah. Oh, lots of people were asking about your dog. This is perfect timing. Oh my god. He's even cuter in person. Not in person. I know. He also knows he's cute, and he manipulates me to death. Yes. Oh my gosh. So cute. And his name is Monty. Yes. Lots of people were asking. It's after mister Burns from the Simpsons. He is mister Montgomery Burns. Yes. Amazing. Okay. Great. Well, the last thing that I wanted to cover here because a few people asked about AI tags. So selfishly, because I, you know, work at Motion, feel responsible for letting you guys know how to use AI tags. So that will be the last thing that I'm gonna show you guys really fast. Welcome, Monty, to my tutorial about AI tags. Okay. I'm gonna screen share real fast here. I've got Motion open. This is our demo account. So if you've been in Motion before, if you have a data source connected, we are inside of Motion Analytics, and you go up here to create a report. And if I just create a top performing report, you can see these are all of my top performing ads. And right now, there are no AI tags in here, but you can see that we have naming conventions. So what what a naming convention is is people will specifically curate these names for their ads so that they can filter for items. So for example, this one says product image with text, offer first banner, feature bet feature benefit point out, urgency, everyday SPF protection. So by indicating the contents of this ad in the name, I'm able to filter for things. So I can come up here and say, filter, add name, lifestyle, or or SPF, or what or what have you. So the whole point of naming conventions is just so that you can get those creative insights and track those things that are not data. Like, it's not a thumb stop rate. It's what your angle or what your format or what your product was. So you put that in your naming convention. The thing with AI tags, is that AI will look at your ad and will do some of that work for you. So we have these AI tag categories that we've automatically tagged all of your ads with. So we have asset type, which is like UGC video or image, lifestyle image, visual format, which is, like, us versus them, testimonial, unboxing. And then this is the one that Dara was showing off, which is intended audience. So this, just to be very clear, is what we have a prompt that looks at the ad and says, who do you think this ad is speaking to? So this cannot guarantee that this is the person who Meta is necessarily targeting, but it can certainly illuminate you, I think Dara said, to who you are actually speaking to in that ad. If you made this ad and you thought that you were targeting, I don't know, like, people with mature skin or something, and the LLM says skincare enthusiasts, it's just kind of like it's not meant to be, like, a hard and fast rule. It's just kind of an interesting way to look at your ads and identify gaps that you wouldn't have seen before. So like I said earlier on in this presentation, the Motion team thought that they were making a lot more ads for media buyers. And then when they looked at all of these, they realized most of our ads are tagged with creative strategists. We're clearly not speaking to the people that we thought we were speaking to, so we need to reevaluate our strategy and how we're speaking to these people. There's other AI tags as well, like messaging theme, seasonality, offer type, hook tactic, that type of thing. The other type of report that you can make that you saw me make was a comparative report. So these are your top ads that you can sort by things like spend or first frame retention or whatever your primary KPI. Whatever you like, you can sort top ads, but you can also make a comparative report, and that's the one that I was looking at where you can just group all of your ads by one of those tags. So let me show you what that looks like. Also, thank you, motion gods, for not putting me through what poor Evan went through the first week and having motion work for me. So this is great. So this is what this has done is grouped all of my ads by the intended audience tags. So that I can see a 103 of my ads are speaking to budget conscious shoppers, a 139 are speaking to skincare enthusiasts, 235 are speaking to millennial women. But then I can also see, like, wow. Almost, like, twice as many ads are speaking to millennial women, but we haven't spent nearly as much on this particular audience that we're speaking to. Why is that? Like, is this not a scalable audience for us? So you can start to make some assumptions when you compare this type of information with data. So that's how you find the AI tags in here. Hopefully, it's very helpful for you. And, yeah, and you can group by other things too. Like, were talking about, like, hook tactics or for example, we're talking about hooks in the first lesson. So, like, urgency hooks, social proof hooks, question hooks, sales hooks. Yeah. But that's how you find AI tags, and that's how you use them. I hope that's helpful. Yeah. A quick one there, Alicia, just to build on what you what I usually like to do as well is, like, combine the two. So, like, having our naming conventions of, like, for example, our functional optimizers. I'll filter out functional optimizers and look at AI messaging angles. So that's why I. screenshot it in retro Yeah. and come back like, understand the differences as well, so adding those filtered nuances. But, also, you can go into, like, a wider stance as well, like, looking at intended audiences, looking at, okay, top spenders, maybe the creative strategist from motion, and then going into the another kind of, I guess, top ads or whatever and having that as the filter. And. then you can see you can look into the messaging angles for that. So it's kind of, like, doing that double filter, if that makes sense. Yeah. For sure. Yeah. Because you can hook up your naming conventions inside of Motion as well and be able to filter for them as well. I think I think it's next week or maybe the week after that, we're gonna do a little bit deep of a deeper dive into analysis. So we'll teach you a little bit more about how to use Motion. I know some people still find this quite fast and a lot to absorb. I totally get it. But, yeah, if you have any questions, please feel free to reach out in the Slack community, and maybe I can film some Zoom videos as well. But yeah. Okay. I think that's everything for this week. If you guys have any leftover questions that we didn't get to answer, there's an ask the coaches channel inside of Slack. That's a great place. If you have a question that's unrelated to homework and particularly related to the session, please go drop that in there. And then we all spend some time this afternoon answering those for you. Thank you so much for showing up to another Thursday session, and we'll see you again next week. Really appreciate it. And, yes, ever all of the docs will go to Slack. I promise. Okay? Thanks, Thanks. guys. Bye.