How AS Beauty Scaled from $500 to $50,000 in Spend

How AS Beauty Scaled from $500 to $50,000 in Spend Speaker 1: You were kind of one of the first ones to test out AppLovin. Speaker 2: By Monday of Labor Day weekend, we were spending like $50,000 a day. So we went from like $500,000 at the beginning of August to like $50,000. We doubled every day. Speaker 1: Pretty much. Speaker 2: Through the weekend, then we got more comfortable. Like once we got something, we were like $5,000, $10,000, $20,000, $40,000. And I think over like Black Friday, Cyber Monday, AppLovin became almost like 30% of our budget. It was wild. As we spent more money, revenue went up, new customers went up. So we felt confident as we kept pushing it that all the other metrics held. Speaker 1: Welcome back to a new episode of Chew on This. Today we're going to be diving into our special series where we're talking about diversifying marketing channels. Brought to you by AppLovin. And we've been graced with the presence of Scott Kramer, who is the VP of Growth at the AS Beauty Group, which is a house of brands that have notable brands like Laura Geller and brands that you've seen probably in your own home. And Scott's done an incredible job doing literally what we're here to talk about, which is literally diversifying across marketing channels. And I think one cool piece there, Scott, is You were kind of one of the first ones to test out AppLovin and is actually one of the reasons why a lot of us get to test it out. So, you know, we'll get into that today, into the episode, but for the few people who don't know you, maybe a little bit of background, a little bit about all the things you manage and what you roll up to. Speaker 2: Yeah, well, appreciate it. Nice to be here and yeah, always good to hang out with you guys. Yeah, I've been at ASPD a little over three years now and I started we're mostly just Facebook and Google and I mean from my agency days I saw that just diversifying ad mix was It was really important. So I'm their VP of growth. We've got four brands in our portfolio. Laura Geller, Cover FX, Julep Beauty, and Bliss Skincare. So across the gamut of cosmetics, skincare, and all different sizes. So our businesses range from kind of doing small eight figures to over nine figures a year, just in dot-com alone. We also have a little bit diversified sales mix. So we do also Amazon. So we do some wholesale. TikTok shops has been big for us. And then retail is something that we're always kind of playing with and looking at. But the nice thing is we're always trying to grow at our own pace and speed. So diversifying, marketing, all those kind of things that we're talking about today are very relevant in how we've been able to grow the business in a healthy, sustainable way. Speaker 1: First of all, the fact that you have so many different brands to work on, I'm sure there's obviously great benefits of learning and being able to hypertest and be able to move quickly, but then there's also an element of just So much management, right? I'm very curious on how you kind of go about when it comes to just acquisition and thinking about it across the different brands you have. How do you approach that? Because a lot of people's dream is to build a portcullis. You know, you do it once, then you do it multiple times. But give us a little bit of color into what a portcullis looks like. Speaker 2: Yeah, it's definitely not not easy. I mean, I think that's where coming from an agency background, it's very much like that's I think the fun part of you're not working on one thing. So because something worked for another brand, maybe it's gonna work for the other. Obviously, that's the dream is like, hey, if we see success here, how do we take it to all the brands, but also kind of helps us diversify our risk a little bit of like, hey, if we want to really test something, but we're afraid to test it on our largest brand, because it could have a big impact to the business. Let's test it somewhere smaller or vice versa. Hey, this is going to have bigger wins and it could work across everywhere. So that's kind of the approach a little bit, but yeah, the day-to-day, obviously earlier on when all of our business were a lot smaller is easier to kind of manage. We're kind of starting to figure that out. We've actually just started to silo out a team that's head of, I forget if we're calling it innovation or acceleration, like an acceleration kind of hub of like how do we have a team that's really focused on, because that's the hardest part where it's like, Like anything, when you have 90-90% of your business focused in one area, how do you think about that other 5%? I think that's also the struggle that we faced early on with multiple different channels when you're spending 90% of your budget on meta and you're like hey how do I make TikTok work and it's like okay everyone just takes their best creative and throws it there and similarly like you don't have the time to put all your energy into these smaller areas so now we're kind of like starting to really think about you look at the L'Oreal's, the Estee Lauder's where they've got MAC, Bumble, all these brands that are like They're all independent, but I think that's also the problem. It's like, if those businesses aren't talking to each other, if they're testing AppLovin and seeing success, how is the Bumble team talking to the Mac team or something? Like, what does that look like? So I think they're still learning, so I think that's also interesting. Like, the larger holdcos are looking at how the smaller ones are doing it, how they'd be nimble. Is there a digital team that's across all of it? So I think we're, as we continue to grow and build more brands, I think that's gonna be the evolution of it, but yeah. It's hard, can't always bet a thousand, so I think we're trying to figure out what is the next brand that is going to be the one that we really want to focus on, or is it another acquisition down the line too? Speaker 1: Absolutely. Speaker 3: Makes sense. I guess the question I have in mind is you have so many brands that are pretty much targeting the same people for the most part, right? How are you guys Kind of separating out your marketing, you know, strategies and coming up with a creative strategy for each brand, for each product. Because even when we're doing it for our brand, right, we're trying to hit as many different demographics as possible, different personas, different types of messaging. I'm sure it's difficult to do that, but then across multiple brands to the point where I'm sure you're competing against yourself at some point, right? Speaker 2: A little bit, I think that's like, I want to touch on two things there because I think that's, everyone always thinks like, hey, if I start a whole co, I already have this customer that could sell this other product to. It just never seems to work. I mean, you've probably experienced it and like, even if you're in like the same category, oh, these are beauty products that sell other beauty products to these same people. They just don't resonate the same way. We've seen more success now with skincare and beauty a little bit because like, Obviously, that's the progression. Okay, we've got a good beauty line. Do we launch skincare products? And we've thought, do we launch skincare products on that brand? Do we launch separate ones? I think that's where we're kind of seeing what that helps us for our own product evolution. But yeah, I think that the hardest part is kind of just trying to like sell one item to another person Even though like you think that's gonna work, but it just never seems to I think that's I mean how beauty's grown so much lately there's it's not as much like a product like agnostic thing where someone's like I'm only gonna buy like a If you love aloe yoga, like I'm only gonna buy aloe. I don't touch Lululemon. I don't touch Bilt. I don't touch these other brands. But with beauty, it's like people are coming and testing a lot of different things. So there's loyalty in certain areas. Like people love their foundation. They love their mascara. They love their blush. But they're trying them from different brands and I think that's also I mean if you look what Sephora has done I talked to so many women that love Going to support just to get the free gifts and the free like GW peas that they get just so they could test new things I think that also Plays into the nice thing with that the category where people are liking to test and try a lot of different varieties so we can all kind of like We play together in the same space. Our brands, even our competitor brands, we share a lot of notes and talk to them. I think our meta group that we work with in Disruptors actually was four different beauty brands that I was close with all the different heads of. And so we were able to kind of share notes and have different powwows as we were kind of thinking about international expansion. Even as we were talking with AppLovin, it's like, We were discussing with them like what are they seeing and that's where I think where we really kind of made it almost as like internal competition early on of like the success we were seeing and we were able to talk to other people that were seeing success like oh if they're doing that maybe we should scale. I think that's what just made it a fun exciting like Q3 and Q4 with AppLovin. Speaker 1: You talked a little bit about AppLovin already. I think it'd be awesome to go back to the days you kind of got into it, right? Just maybe first with like your mental model, right? I love when you said like, hey, we're spending 90% on this one platform. Now, how do you even think about trying a new one? Forget like trying one like a TikTok where it's kind of around people are getting success and whatnot, so you're more apt to it. AppLovin, again, you guys were literally one of the first to beta their new product that's built for ecommerce now. But how did you even get to the mental point of like, yeah, I should go and put time towards this and try this? I think a lot of people are initially even just stuck there. Speaker 2: I mean, to be honest, I had no idea what to expect. We were approached from one of our partners, Disco, and they had set up a partnership that they're working on with AppLovin and trying to bring it into e-commerce. They're like, hey, they're doing a funded pilot. We do a lot of funded tests through Meta, TikTok, all these channels. I kind of thought nothing of it. I'm like, you know what, what's the worst that can happen? They're going to pay for some ad credits and went through onboarding. They're like, you need to place this pixel, do this and that. We need your 10 best creatives. So I worked with our creative team and like they need to be under a minute. So we just went into the meta account. I was like, hey, pick the best ads. And we sent it over in a spreadsheet and they didn't think anything else up. And I was like, okay, you know what? And then a couple weeks went by, I was like, did we launch? Like, what's going on? I was like, maybe I'll start to see some sales. Because we have so much going on, it was like, this is exciting, but again, this is like probably June or so. And then July, the AppLovin team was like, hitting us up, like, oh, we're seeing success. Things look good. We're like... Okay, we don't really see anything. We have no idea. And then we go and we look like UTM. Like, oh, there is some AppLovin sales coming. Like, oh, cool. Like, how much do you spend? We didn't even know how much they were spending a day. So we just literally knew nothing. It was just on autopilot. Then they started talking to us. And then I think July, we were getting on regular meetings. And at this point, to be honest, I had no idea what AppLovin was. Speaker 1: That's insane. Speaker 2: I thought it was like another display channel. It meant nothing to me. And then they finally sent me a screen recording of what my ad looked like. In someone playing like Candy Crush or playing a game I was like, oh that's what it is. Speaker 1: We've seen that reaction a lot now. Speaker 2: And now it's becoming an issue. It's like, okay, mobile game ads. Okay, I get it. It's played in between my games. And I think also because so many of us that, I mean I'm not a huge gamer, but obviously on commutes you see everyone doing it on the train and whatnot. I play a little bit on planes and here and there. And you're like, okay, and you're just so used to getting other app games that it was something you never even thought about as a placement. And so that was like July, and then by August, they're like, hey, like, what do you guys think about really ramping this up? And we were still only spending like $200 a day, maybe $500. So again, as compared to our meta budgets and everything else, it was hard to really see an impact. But they're like, hey, we're seeing good ROAS, and we use triple whale pretty closely for measuring kind of like a first click attribution. And we started like, Looking there like, you know what, the numbers look pretty good. ROAS is comparable to a meta. Like, let's give it a shot. Let's start to scale up. And so then I think through August, we started to spend more. We went another couple hundred dollars a day. And that's kind of our testing theory. It's like, let's spend a little bit and let's push the budget and kind of like see that ceiling. Let's keep pushing it until we kind of like hit that ceiling. Speaker 1: Is it fair to say you typically are just doubling that previous span? Speaker 2: More or less. I think we started like 200, 500. I think we went up In like increments that we felt comfortable with it was like 500 to like 750 to a thousand and then it was like 1,2001,500 and then when we got there I think we're like hey like This is looking really good. Let's try going a little more aggressive. We would wait a couple days because we wanted to see like, is it holding? Do the numbers hold true? Look at like seven day lookbacks. And I think we're up to like $2,000 a day. And this is where I was talking to some other people in the Disruptors team. I was like, what are you guys seeing? What's going on over there? Like hitting people up that I knew because I think there's still maybe 10, 20, maybe less than 100 people on it. So fortunately we had a cohort of people that we could talk with. But the crazy part was then it was like we're going into Labor Day. And Labor Day was like a big sale for us. And I talked to this other company and they were like, oh yeah, we're already gonna spend like $5,000. And they were like, oh wow, they're spending more than we are. Okay, I was like, how much are you expecting to spend this weekend? He's like, I don't know, maybe 10,000, 20,000. We're like blown away by that. Like this is impossible. Fun competition like well if they're spending that much maybe we can spend more. I think we like just started ramping it up also like obviously the numbers held it wasn't just kind of like an ego like oh we can spend more, let's go bigger. Obviously we wanted to make sure we're like ROAS looked good and everything. So me and the team were in the accounts and through Labor Day weekend, we had Labor Day Creative in there. And again, it's very similar to our meta ads and everything. And through that month of August, all the feedback we're getting from the team is they're updating the models, they're improving the algorithm. And as they're getting more data, it was helping inform and just the performance of the platform in general was getting better. And I think by Monday of Labor Day weekend, we were spending $50,000 a day. So we went from $500 at the beginning of August to $50,000 on Monday of Labor Day. Speaker 1: We doubled every day. Speaker 2: We got more comfortable. Once we got something we were like $5,000, $10,000, $20,000, $40,000, and then it was like, oh, $50,000 maybe. I don't know. For some of us marketers, we like round numbers. It feels good to hit milestones. We were really happy. Then obviously after the Labor Day sale, we pulled back. The whole time of August when things started to work, I think, or maybe not, it was probably in September where I was like, this light bulb moment that like, hey, like, if this is really as good as it is, like Q4 could be massive for this platform. Because in my mind, again, it's like, we know Meta's great, Q4 is for all e-commerce businesses, holiday sales periods are huge, but Meta gets really expensive. And I was like, I was talking to that AppLovin team, and we were working with Tony and Bo over there, and Kathy's been great, like, You can't say enough great things about the AppLovin group. And they're like, we're like, what is Q4? What is November? And they're like, I mean, maybe we see a little bit of an uptick. I think we're seeing like, maybe CPMs go up 10%. I was like, oh, that's it? Okay, this could be like our little like secret back pocket. Like, And so that's where I mean by the time like November rolled around, I think over like Black Friday, Cyber Monday, I think AppLovin became almost like 30% of our budget. It was wild. And so it's a huge thing to have to help us offset. Obviously Meta did amazing. We scaled a ton of money on Meta and grew our budgets. I think October, November usually just like double budgets in general, but just more of that share went to AppLovin. So it's not that like I think it's the thing that I talk to people like, oh, who's hurting? Like, did Meta, did you pull back on Meta? It's like, well, we grew everything. But I think it was just very much incremental ads. I think that's where it's like, we unfortunately didn't get to have the time to do true incremental testing or that. But I think like, The nice thing is as we spent more money, revenue went up, new customers went up, so we felt confident as we kept pushing it that all the other metrics held. I think that's also the hard part, especially for us at our size. We needed to make those big bets to really see it pay off, but the metrics were looking good in Triple Y and other channels we were testing, so we felt very confident the whole way. Speaker 3: I love that. Even when we tested AppLovin, it was taking our best performers and just moving it to AppLovin and coming up with a few end cards, right, and seeing what worked. I mean, at $50,000 a day, how does the creative process or what does that look like in terms of, well, are you making creative specifically for AppLovin? I know the end cards have to be buried. I think that's where it drives a lot of the performance here, but I'm curious on the creative strategy that goes into, well, we have a process for meta. Now that we've actually seen something substantial here, how do you kind of change the strategy a little bit? Speaker 2: Thankfully we've got a really great creative strategy and kind of like paid creative team and internalized works on that and then also the AppLovin team. They gave us a lot of feedback on what is for end cards. They built the first end cards for us also and so we worked closely with them originally on like Because we didn't even know. It's like, what is a good end card for this platform? I think that was also something that was like, oh, typically it's like we would do end cards at the end of your meta or TikTok video. And it just kind of like lives there. Speaker 3: And for those who don't know what an end card is, it's basically when you see an ad on, let's just say, you know, you're on Candy Crush, see an ad, you play the video, you could probably skip the video. But after you skip, you get like an ad for the brand or the product. So there's like that last Try to like get somebody to click on the ad that is what the end card is and so that's why I think it's very important. Yeah, right. Speaker 2: I think it's something to get to so I think that's also what's just so unique about AppLovin. It's like you're getting someone to watch for like 15 to 30 seconds and then you get they skip it and they get another ad and then they get another. So they're really getting a lot of that information for the same cost as maybe some another kind of placement. Speaker 3: Yeah, you can't scroll past it. Speaker 2: People need their lives. Speaker 1: Exactly. Speaker 2: This is what I think is just so unique. Everyone talks about what made Meta so good and what's making AppLovin so good. I think it's the way that they're able to weave ads just into something so natively. People scroll through ads all day long on Facebook and Instagram. It doesn't feel like an ad. I think AppLovin, for whatever reason, I think just the way that freemium games kind of rolled out, that people just got comfortable. To correlate to like TV pre Tivo and DVR. It's like you can't skip over you're stuck there, but you know what that's life I'm gonna watch this commercial and then I'm gonna get through I'm gonna get back to my show I think so many people have that just stuck with them of when they're playing games It's like it's not even a bad experience. Just like that's what it is. It's like I'm gonna see an ad It's expected and I'm gonna expect it. I have to wait. Okay, when is that little thing pop up? But while I'm waiting, I'm gonna like watch it and engage with it. So I think that's where it's just like I I mean, I don't know, Meta and all these other channels like years ago would say you have five seconds to stop someone from scrolling. Then it was three seconds. Now it's one second. Now it's like you've less than a second to stop someone when they're scrolling. And now all of a sudden there's a channel where like, wait, they have to watch this? It's amazing. Speaker 1: I love how you kind of talked about the process of How are you guys scaled? I'd love to maybe hit on a few of like debunking of some of the myths or not myths but like some of the roadblocks that people do come across and like a few that I've heard I'm sure you've heard some from people but like the one I hear a lot is like oh I don't think my customer plays games and like the ones that may I don't know if they'll If while they're playing a game, they're going to go and click on an ad and buy my, you know, nail kit at home, right? Technically, the response that can apply to all platforms, right? Like, you're scrolling on Instagram, you're not really in the intention to buy but like, How would you kind of answer that piece of like where AppLovin's inventory sits in a place that kind of feels a little bit like you're playing games, like your probably lowest intent in some level of mindset, but Clearly, performance is saying the opposite thing. What's your response? Speaker 2: Yeah, I mean, I just keep equating it to like meta. I mean, I started running meta ads at scale in 2016. I was like, wait, this actually works for brands? And we would talk to clients all the time that were like, I don't buy in it. So I think that's always the tough part. It's like we ingrain the actions of our customers in what we do, but we can't even do it. Even if we think that we act like our customer, I think you just got to look at the data and see that it shows that there's so many people that are learning and they're kind of finding out brands through ads on all different channels. I think also the interesting thing is like the way that it pops up. So people interact with stuff differently where it's like, you know what, I click on it and they just let it save. And it just sits there as another like Safari window on their phone or another Chrome browser. And then maybe when they're done with the game, they'll come back to it. Other people, I've heard, they take a screenshot when they see an ad. So there's different ways that we just don't think about that people are doing it that's maybe not the way that we do it. I mean, again, I'm like, I don't buy anything on an ad. Everyone has that experience that they say, well, I don't do it, so my customer doesn't. But I think that's where you got to kind of understand. Also, I mean, to the audience type. We took the same stereotype like, oh, it must be a bunch of gamers. We're not advertising to gamers. And it's like, wait a second, the people that actually are playing these type of games are our core demo. Our core demo is women over 40. So it's an older generation that is at home. When we remember people that have played FarmVille, it's like your mom, your grandma, similar. And then we also use our post-purchase survey. So we use NoCommerce for that and we do a question in that of like, did you come in from a mobile app game? So we get like really specific and then we ask them what game. Speaker 1: Instead of just asking where did you come from, you prompt with... Speaker 2: Well, it's where did you come from, but that answer type is, is it a mobile app? So we have, I think we've got like 20 or 30 things we've From TV commercial, Meta, TikTok, Instagram, all those kind of channels and then in the mobile app game and then the follow-up question of like what was the game? And a lot of times it's these like mental games. So it's like word scapes and Scrabble. So again, it kind of aligns with like, okay, that is our audience. I mean, I think it's like not to say that maybe there aren't other audiences also on there. I know there's like apparel brands and men's brands that people have been seeing success with. So I think it's more than just like that older female demographic. But the same thing like you're saying, like years ago when TikTok was coming out, everyone's like, oh, I'm an older demographic. Gen Z is not my customer. I'm not running TikTok ads. Not to say that that's the only group. Maybe there is a cap to how much you can spend at scale, but it doesn't mean let me not use that channel because I don't think my customer's there. Speaker 1: Great, great response. Speaker 3: That's so interesting. Just to understand what games your target demo is playing, I think even Changing the ad creative to kind of match what they're doing so that it flows with what they're doing. I think we were talking to Nick Shackelford, right? And they're pushing a lot for Breeze. Being able to like tap into somebody who's playing a game and it's like, well, if you're playing a game, you might as well chill out as well and have a drink, right? Or, you know, even for us, something I've been wanting to go back and test is like, you're probably playing games like right before bed and we have like a nighttime aid, right? And so it's like mixing and matching that creative messaging, I think is so important. And I think a lot of people forget that you do have to have some sort of marketing strategy and not just like, Hey, this worked on Facebook. This is gonna work here. And it's like you have to remember that there's somebody on the other side of that phone doing something and you're interrupting their day. It's like, how do you make that less intrusive, you know, with the right message? Speaker 2: Yeah, I think that that's always true. It's like, I think everyone was able to do so well early on with their metacreative and now it's like, what does phase two of that look like? How are we developing ads just for it? And I come back to the non-skippable. It's like, we were spending so much time creating ads to get someone to stop for a second. And so our ads are so good at just like getting a lot of information to someone in a short period of time. So when you give them 15 to 30 seconds of that, no wonder it's gonna do great. Because it's like, hey, like, I'm just trying to get you to watch for one second, two second, three second, and then you just kind of like, put all that together into this. I think that's where everyone that kind of just launched with MetaCreative similar like, or opposite to like, hey, take your MetaCreative and throw it on TikTok. Oh, it doesn't work. I think it worked well on AppLovin because it's like you're really good at creating digestible content in a short period of time. So you're stacking a lot of that together. But to the point of like how are we developing for the next one? I think we're working with AppLovin team on like is there a gamification like some of these like word puzzle games and it spells out Laura Geller. It spells out beauty or like blush. I'm like is that gonna work? I don't know. Is it just something to make us feel like, hey, we're marketers, we're gonna do stuff that we think is right. I think that's the whole thing, is just keep trying and testing stuff. But yeah, it is interesting, I think, playing with end cards. There's just so many more variables to it all. Speaker 3: One thing you guys said was, you're obviously spending a lot on meta, TV even, AppLovin, and I'm sure other channels, right, like TikTok, Google. I think one of the biggest Questions that people have is going from one platform to the next, and to the next, and to the next, right? And actually proving out incrementality. Because every single ad platform is going to want to take credit for some sale, right? I know Facebook is the worst at it. So how are you guys, I guess, like, what is your strategy to make sure that, hey, when we do deploy cash here and, you know, Snapchat is saying, well, we're driving a 5X return, How are you deciding that that additional dollar is incremental? Speaker 2: Yeah, I think that's for us where triple L and first click attribution has been just monumental for us. When I first started with the team, when triple L was first coming out three years ago, I was like, hey, this is going to help us. We're still coming out of iOS weirdness of meta. It was delayed and modeled reporting. We're like, hey, this is going to help us tell more apples to apples. Just the old school mentality of me, I was like, I want to get back to like first click and be able to understand delayed attribution and not have it just kind of like appear in ads manager. And then also to look at like meta TikTok and other channels together and even the playing field, like we spent on Pinterest. Pinterest, they've got like a, they've got three metrics. It's like click view and engagement. So they sometimes email like a 30-30-30 window. Speaker 3: It's ridiculous. Speaker 2: So I think that's where, and you'll see like these six, eight, ten rows, but if you were just like, oh, trust the platform, like spend there. I think that's where it's like, how do we create some kind of even playing field with the attribution tools? And now we're really expanding it to like MMM tools and really thinking about beyond just the click aspect, like what is that? Like you're saying, like people just viewing it, maybe that has an impact on it also. So how do we better understand? But obviously it's been a journey for us to get to this point. I think obviously at our size, we're doing it a lot different than a lot of people, but I think many people can get really far using a click attribution tool. First click more than last click, but again, like early on, I was just looking at do I even believe in this at all when I knew nothing about it. I was just looking at last click as UTM. So again, you can kind of create your own proxies. Again, each business is different. If you're talking a high AOV company, maybe there's more consideration paths. Maybe you need to look at how did you hear about us. The tough part, I think, how did you hear about us, mobile app game. I haven't perfected what the right way to ask that question is, so I know there have been studies looking at how you answer those questions. I know people that are very much like meta and TikTok, and they're like, hey, the attribution in my triple L North room doesn't look as good, and platform looks okay, but my how did it hear best, everyone's saying TikTok, so let me scale on TikTok. That's worked for some businesses, but I think there's just different ways to kind of cut it based on your business. But yeah, for us, Look at that first quote attribution. Also look at new customer. We look at new customer ROAS, new visitors. So I think that's a big one. I know we'll probably get into also. Yeah, that's a big concern people have. So really making sure you're watching the metrics that matter to your business to see that if it's gonna help grow incrementality. Speaker 3: I think what you just pointed out was also like a concern people had, right, is where if you're trying to drive net new visitors, right, the point of diversifying is to drive net new visitors you know, get out of whatever audience meta is, you know, circling, right? So I guess the question here is like, what are those targets? And how do you set those targets for each channel to understand? Yes, this is doing what I want it to do. So I guess what's the what's the tactical solution there of like, really understanding your KPIs and setting them? Speaker 2: Yeah, I think it's also creating the baselines. If we know meta is where a lion's share of our budget is and that's doing well and we're happy with that ROAS, that metric, and it's matching what we're comfortable with where the business is, let's use that as a proxy and a baseline and can we get to that or get close to it. So I think that's where we're looking at those metrics as benchmarks and then also like obviously TikTok has some that we're seeing. So it's very much looking at what we could in triple whale and kind of analyzing that. I think that's where the new visit percentage got really important to us as we were like, how do we make sure we're getting new people to site? We know that we can't do exclusions like we can on essentially every other channel, which I think is the scary part for so many people. Like, how is this not just going to be retargeting the bottom funnel? I think that's something we've been working with AppLovin team and they're in the process of kind of like rolling out new things to kind of answer it. So I think that's also where we started. We were actually on, I guess it was more like traffic campaigns were the original campaign type. Then they rolled us to CPP, so Cost Per Purchase, and then they're like, hey, ROAS is the new thing. And so we were testing all of them in tangent, and even from the early days, like, with the AppLovin team, they're like, you should really just run one campaign. We're like, no, I don't know. We run a lot of stuff. We've got to test things. We've got to diversify. I don't want to put $50,000 a day into one campaign. I want to diversify. That doesn't work. Or different product funnels. As we tested all that stuff, And it did well during the sale phase. I think at those moments, anything works. But coming out of Black Friday, it was some of the worst period that we had with AppLovin. It was because we were testing CPP campaigns. We were testing ROAS campaigns. We had a sale campaign. We had one for our foundation, one for our kits. And I also think back to the early meta days again. When you have that audience overlap, you're like, oh, if you have three lookalikes running or you're overlapping on interest, it's like, it doesn't, it hurts performance overall. It's not like one does well and the others don't. It just kind of like breaks everything. So that's what it took us time to realize. We're like, oh, we should probably just listen to the AppLovin team. And so then in February, we're like, let's consolidate just to the one ROAS campaign. And we saw, funny enough, new visitor numbers came back up, all the metrics we really cared about. So it was kind of like, It was hard as a marketer that wants to like do all these different funnels and kind of like build that way, but that's how the platform today is used the best way. And so we've consolidated just to ROAS and our focus there. And similarly, we have it on one of our other brands and did a similar thing. Let's just consolidate to one and we saw that new visitor, the metrics that we got concerned about. So we saw new visitor percentage drop. New customer looked good, but again, it's like, is this just going after Retargeting past customers, people that have been to the website, and then we actually did see that improve. Again, it's not to the same percentage as meta. I don't think it's ever going to be, without being able to exclude, but it's at a point that we feel comfortable with it, driving incrementality. And it's also like, if it's not there, we need new customer ROAS to be that much better. So let's say it's like 20% worse in new visits, we need the ROAS to be 20% better. So we kind of need to balance it, so we're not going to run it the same way. Speaker 3: Makes sense. Speaker 1: I'm going to jump into a little bit more around just like your entire stack today in terms of the different channels, right? So I'm assuming is meta still obviously a big lion's share and then like can you maybe break out how you're just on a percentage basis how you've diversified across the rest? Speaker 2: Yeah, yeah, I mean like when I started I was like meta in TikTok or meta in Google really and then we expanded to TikTok and we saw some good wins. I think TikTok's definitely grown more for us lately with TikTok shops. So we're seeing a lot that's big in beauty and skincare and cosmetics. It's something we want to make sure we're there and growing. And then TV was a big breakout for us about a year and a half, two years ago. But it wasn't until we got to a certain size. Like I think there's a size and place. I'm not gonna tell everyone like Hey, you need to diversify. I think so many businesses, especially if you're sub 10 million, even if you're sub 20 million, I think you get very far just on meta. And then kind of now is it time to diversify to some of these other channels. But that's not to say it's not worth testing AppLovin early on, because again, we're saying like, what is the lift to test a channel like that? Testing a channel like TikTok may be a lot more complicated, like we're saying, diversifying creative. As long as TV is a whole other ballgame of like, Try to spend the media. I know there's a lot of new technologies coming out to make it easier to get into TV, but it's a top of funnel. CTV is a whole new thing that we're seeing more success with. So again, it's like consumer behaviors as they catch up. Are those the next areas? For us I think as we kind of scaled we saw that TV was a great upper funnel opportunity for us. I think that's where we're really been happy with that helping create meta as it's still our largest channel but this like mid to upper funnel area and then kind of like we don't really want any like bottom funnel. We try not having that. That's what email, SMS, all those areas are for. We have abandoned card flows so no reason to try spending too much money bottom funnel. We're very focused how do we get as much new people in. Speaker 1: Awesome. And on that same note, I know you've been working extensively with the AppLovin team. Any tidbits that you're like super excited about going into the fact that, you know, they'll be coming up on a year of really rolling this out, testing this out. So, so early, right? Because you can, if you go back to like meta days of year one, it's like, it's everything is still so new. You know, I love the fact that they have like a resource library and like literally put like winners. For you to go and copy and it's not just like a Facebook ad library we kind of have to guess but there's some things that I think they're doing that's really cool anything from your Lens and the fact that you have the privilege to be able to obviously even be deeper Anything that you feel like is gonna be some of the exciting things that they're working on Yeah, I mean, I think that's a great point. Speaker 2: We're talking on creative. We could have gone deeper on that for hours But I think you have the resource library of creative to show you hit you like the weekly wins. To get inspo and you can say, hey, we made it on the board, or you see everyone else is kind of testing similar styles. I think it's like we're all kind of learning together, but outside of Facebook ad library, you're taking a guess, like, oh, let me see what AdMobby's doing. I heard they're crushing it on the podcast. Let me see what ads they're running. But yeah, I think it's that. I mean, I don't know if Cathy will maybe slap me on the wrist or maybe this will get cut, but I think a lot of stuff they're looking to roll out is in some of these new customer optimizations that they're trying to test. I think that's where they're hearing the feedback from us and they're looking and rolling out tests and testing some of that. There's also the TrueSelfServe platform I know they're working on is coming out. I think that's also the exciting thing, like all these channels, like we give feedback to the metas, the TikToks, and they're big corporations. They move slow. I think it's been amazing to see even talking to internal people at AppLovin, like how excited they have seen internal teams, that they're all getting on board. Speaker 1: Because they have a whole other business to worry about. It's like, you know, been around for decades. It's like, no, it's really cool. Speaker 2: Yeah. Speaker 3: Aside from like just the marketing channels in general, how do you see creative evolving, right? If you asked me this question like two, three years ago, it's like, oh UGC is everything, right? Hire a creator. Script it out, film it, send it back, you know, run it. And I feel like the creative trends are probably starting on TikTok and then kind of making its way elsewhere. But I've noticed a big shift in what creators were working and no longer working now to what everybody's doing, you know, with podcast style ads and street interviews. And I feel like there's just these constant trends that come and go. How are you guys thinking about you know strategy when it comes to well obviously we want to capitalize on trends but Understand it may not be an evergreen strategy. How do you guys think about just creative? Speaker 2: Yeah, I think it's such a tough one. It's always evolving. I mean, you think about UGC creative, it's like, do we go our entire ad account on UGC creative? I think we also do a lot of stills and still images still do well and kind of like product flat ways and kind of like kind of explain or kind of like value optimization for that and up value positioning on price and kind of call outs. Even like you think like ugly ads and some of these things like how do you just like try some of these different trends but not get overwhelmed. I think UGC is it's here to stay. I think it's just it hits people in that way that it's like it's native it feels like they're seeing a friend or someone they know but how do you kind of make it work for you and for your business. That's not to say don't spend any time on high-res stuff. Again it's like we're testing some of that again and again also because we have a lot of that we're shooting for TV and even testing more like this high fidelity for AppLovin maybe because it's like, hey, is this like a traditional TV commercial? Does that work? I think that's the beauty of creative. It's like you guys have probably seen all the time. The ad you think is going to do best never seems to win and the worst ad can do the best. I think our hardest part that we had with our creative team over the last year is giving them the data and the why. They kept being like, well, why is that ad better than this ad? And it's like this, they want that exact silver bullet answer. It's hard to quantify, but it's not like, oh, this is shot on a red background, this is on a blue background. People like red, that's why the red one won. It's not as kind of like black and white. So I think that's where giving them as much information so they can kind of make their own inferences from it. I think that's where if we take only like my opinions then our creative is only going to look one way. So I think that's where it's how do we kind of like test it. We also go into it with different like brief mentalities. Like we want to brief for different kind of like creative styles for it. So it's like you know what if we're gonna do just bad matches. So this is gonna be a whole set of creatives focused on the objective of you know what my foundation doesn't match well with my skin. So it's kind of like siloing into buckets like that. And then are there trends that make sense with that to play with it. I mean you probably saw was it like the guys that would run up to people's car like What do you do for a living? It's like, oh, you should buy my t-shirt or something. It's kind of cute. It's maybe a good brand moment. Is that going to be an ad that scales and they spend a bunch on? Speaker 1: It's just a moment. Speaker 2: Yeah, just a moment. But you never know what moment can stick. The podcast ads are interesting. It's also like, how are you trying to help sell a product by validating it? I think that's where people connect like, oh, podcast, these people must have some platform. They must be important. So I'm going to listen to them. As we used to say a long time ago, it's like people buy products for like a few main reasons. It's like validation from either a friend or family member, a trusted source. Or if the price is just so good that they're compelled to give it a try. Speaker 3: I feel like this year is going to be the year of creative diversity and creative volume, right? Especially for those who are on meta, the new algorithm, right, Andromeda, is basically saying we want as many, you know, pieces of content to match with different people and they're using AI to put the right ad in front of the right person at the right time with the right messaging with all these enhancements and stuff. So it's like Creating this balance of, well, creative diversity means, you know, static and videos and then within that, well, do we have ugly ads versus, you know, the high quality, high polished. And then how do you just do a lot of it all at once, right? So I'm curious, like, what does the team structure look like for you guys and how can brands start to build, you know, this creative flywheel for their own brands? Speaker 2: Yeah, yeah. Andromeda. That is new. Not even new. It rolled out, I guess, in December. They didn't even tell us. It sounds kind of creepy. I don't know if, like, well, it's scary. Like, what does this mean? Yeah, creative diversity and creative volume is a big Importance, but I think it's just like, how does it do a better job serving it? Not as much for our business, but I think to other people's brands where it's like, you know what, I may have this niche audience. I've talked to some people that say, you know what, I'm going to create a bunch of ads around firefighters and you know what maybe it's not gonna scale a bunch but because that creative is the same message you're telling with another or like or crossfitters or gym goers or whatever like this low niche category or like people that are into farming you'd be like I never thought this was a category that like I'm gonna sell this product that you know what people also want it and so I think as they identify with themselves and hopefully that's where Meta's system has gotten better to identify that, hey, this pocket of people that are really into this category, we're also going to be really strong because we know how to serve these style ads that matches that group. So I think that's a better way with beauty and cosmetics. I think it's tougher for us to go after specific categories to think about. So that's where we think about a little more on just telling different narratives around it. So I think that's where we're thinking narratives while other people may be like, hey, I have these niche audiences that maybe I'm not going to Build a ton of ads, but maybe I can start to penetrate a little bit there, a little bit here and diversify a bit in my creative that way. But I mean, I know we haven't touched on the AI element of it. That's taken it to a whole other place. Speaker 3: Yeah, I mean you can literally take one ad and they'll give you like 10 different variations just based off of what they want to put in, right? It's like, oh, we'll change the lighting on this, we'll change the headline on this. For me, it's like, I feel like my views are changing week over week, because I do feel like the more you kind of give in to what they want, they'll reward you, right? I think at a certain level, it's like, if Meta is asking for you to turn on enhancements, and they want you to use the AI, you know, copy and headline and give it a bunch of variations, They're going to prioritize the advertisers that are testing the new features. I don't know if it's a conspiracy theory, but if I don't give them what they're asking for, then my ads are not going to perform. So I got to the point where it's like, sure, turn on all the enhancements. If they don't work, it's fine. We'll move on. Let the AI spit out some headline variations and copy. If one headline matches what somebody is interested in and doesn't match somebody else's, why wouldn't you want to give the variables there? I think people are scared to give in because it's like, oh, meta doesn't know. I do think the way that the algorithm is moving, utilizing AI to process, I mean, there's a reason they're storing this data in warehouses, right? There's a reason why everything's moving towards AI and it's just faster computing process or processing. And so it may just be smarter than we really think, right? And that's why I think it's like, just give into it, test it out, see what happens. Yeah, I mean just let Meta prioritize what you give it. Speaker 2: Yeah, I think that's where it's like the best thing we can do is kind of like set up the controls that we believe, run tests of like what is the right optimization window, all that stuff and then just feeding it more creative in the right parameters that make sense. It's like don't load a hundred creatives into one and kind of like what is the right framework for kind of doing it and then I think it's the diversification. I think that's where we've also kind of changed our mindset a little bit instead of like changing the copy or like the text overlay a little bit. Let's take bigger swings to really see how that moves the needle more than like hey we created this one ad style that we're gonna do four different versions of it and put it in and then be like okay well what is the new headline that works and obviously sometimes You're going to have wins, but I think you need to make them more drastic to really get bigger wins that are going to be more exciting to see those. Those really play out to like, you're saying, feed the system what they're asking for, toggle on all the things. I mean, we saw this years ago with web and shops. It's like when we did web and shops, It wasn't saying that it's gonna spend all the money, but I think just giving it that opportunity, it knows more than we are where that person is. I may say the same thing, like I've never bought from a Facebook shop or an Instagram shop. Should I turn it on? Maybe my customer's there. If that's gonna be how Metta wants to serve and get me the best for us and hit the numbers I'm looking for, it's worth it. Speaker 3: I think there's a lot of tools that we can start utilizing now, especially with ChatGPT and AI image generation. I don't think it's the silver bullet that everyone thinks it is. I do think that it will allow you to achieve diversification in your creative strategy. If you're using like lifestyle photography of your product or you're using renders, right? There's only so much photography that you can do, right? And lifestyle photography is not cheap, right? For brands that are just starting, it's like you can have a photo shoot. You're not going to have photo shoots every week, right? But what you can do is use AI to create a background. Overlay your product so that at least when your potential consumer is looking at this ad, they can resonate a little bit more, right? So it's like, you know, we want to get an image of somebody sleeping and having the bottle on the bedside. You can do that now. I don't have to have a photo shoot for that, right? So I do think the bar is set higher because the ability to produce more is there, but people need to think strategically of like, okay, well, how do I use this tech that's here to, again, diversify what the creatives look like and put out as much as I possibly can? Speaker 2: Yeah, I think that's where it's gonna be the biggest, at least right now, is for the smaller teams that they don't have in-house creative. They don't have in-house designers where like, I could send stuff to my designers and ask them, hey, can you put this product like a sunscreen item? Oh, can you put it on a beach? Can you put it with palm trees in the back? And like they could edit and design that or pull like stock photos. But if I don't have teams like that or I don't want to spend money on a outsourced editor or whatever it may be, it's like you can just go to a ChatGPT or even I think even a bunch of the different like tools and look at it. But I think the video element of it is where it's getting really interesting. People can see it still. Yeah, so you see it. Oh, that's AI. Yeah, that's AI video. It seems a little weird. Speaker 3: I think it might get there. Speaker 2: I mean if you look I think in Japan they have like these massive influencers that are just full AI influencers that are just from the get-go. Yeah, from the get-go that are essentially you can pay the AI influencer to be like, hey, will you do a content video for my product? And it's like. Speaker 1: That's all one day there's like an account that had 170,000 followers and every post and everything about it was a yeah It's insane like that person went traveling and like posted photos Down in Soho, that's just like he just stands there just like films all day long. Speaker 2: Yeah, he's like, uh, it's like a He thinks he's a robot. I forget his name. I don't know. He's just like interacting live, but he's like, I don't know. I'm not a big gamer, but it's like in that first player shooter kind of mentality. Speaker 3: Oh, okay. Speaker 1: Interesting. Speaker 2: Yeah. There's a lot of stuff you can do. Trying to live stream, figure out things, but I think that's also like, just looking at like the younger generation, like as I get older, I'm like, there's so much we can learn from them, how they absorb media, how they look at it. Speaker 3: That's where the trends are starting. Speaker 2: Yeah. Speaker 1: Before we wrap, I have one question just around, you know, you've been super tactical today, so I'm just curious on your thoughts here, but like, is there a couple of like really interesting things you've done this year? Could it be a campaign? I think you also did, you guys had a really cool campaign on the Own Your Age. There are a couple of things that you've done that kind of surprised you or something that's been like really interesting that kind of stands out in your mind as like, huh, didn't know that's gonna work, but that was cool. It could be an ad, anything that you really can have any references. I think it's always cool to have our listeners walk away with an idea that's like, oh wow, didn't think that would work. It's pretty cool. Speaker 2: I'm trying to think back. I mean, obviously the first start of this year has been a little rough, obviously. I think there's so many things that we think about that we're sometimes afraid to. I think we're at the size where we don't want to be the first movers and stuff. Because you know what, it could backfire. But I think that's where brands that are really, you know what, they're going to take a risk. I've seen some people that are like, Right now with the tariffs and everything they're putting the tariff tax as an additional line item in their things like people like hey I'm not raising prices. We're taking a stance. We don't want to pass it along or vice versa people like Hey, you know, it's starting May 1st. Prices are going up. So I think that's where there's... Speaker 1: Buy before then. Speaker 2: Yeah, buy before then. There's just a lot of different tactics. I don't think there's any... I think for us, obviously, last year was a big year for AppLovin. The year before was big for TV. I think international expansion has been big for us. I think we've figured it out and have a good operations standpoint in place. So that's what I want to tell everyone. Go ahead and kind of, hey, US is tough right now. Go spend money internationally because it's hard logistics wise. There's different taxes and all that stuff that comes into play. But I think just kind of understand your data, looking at your business and kind of being able to decipher what is that next risk or bet that I want to bet on and how do I not expose myself too much into that. I think that's where we're in potentially tough times. So it's like how do we just think about carefully what is the risk we feel confident in making that we can then back out of if it doesn't work also. Speaker 3: I love this conversation, by the way. We talked a lot about creative strategy, diversifying channels, incrementality, attribution, everything. This is a media buyer's dream. But one thing, if you could kind of pull out from today's conversation and leave as a takeaway for the viewers and listeners, what would that one thing be that somebody could implement in their business today? Speaker 2: The one thing I think again, it's just getting ingrained in your data. I think understand that because like I go tell everyone, hey, you know what? You should test AppLovin. Give that a shot today. I could also test TikTok shops, get on TV, but it depends on where your business is and what size your business is. It's like. And also like what is the AOV? I think that's where it's like we were in like a mid-range AOV for like TikTok shops works well for us. If you're selling furniture or luxury items that's probably not the place to be. Similarly maybe for another channel. I think it's like what does it need more consideration or research phase. So even telling someone like Set up your account this way because like maybe that you've seen that it takes eight to ten touch points for someone to make a purchase. Maybe it's like structuring my abandoned cart flow or SMS's differently, but I think that's the hardest part. It's like trying to learn from other people is helpful, but I think it's you got to think about and look at the information and kind of like understand like what is it my customer wants and what is the data telling me instead of being like hey I heard Scott tell me I need to go do this so I'm gonna do it and then it doesn't work and they're like Find me on the street. Speaker 3: That's a lie. Speaker 2: Yeah, exactly So I think it's it's the toughest non-answer Unfortunately of like you got to look at it know where your business is and what's right for you in those different phases of growth But I think it's also surrounding yourself with like-minded people that have maybe got to the next level and find out What was it that they tested that they did or what did they change? I think even for us early on is like We started sending more emails and more SMS and everyone's like afraid to do that early on. It was like, well, you know what? Obviously, it's like we're afraid. What are we afraid of? Is it our unsubscribed rates are going to drop? Speaker 3: Yeah. Speaker 2: It's like, well, do we watch them? Did they actually drop? Or is this just kind of like a perpetuated fear we had? So I think there's different areas that you kind of like go in and kind of see where your business needs to grow and where there's opportunity. Speaker 3: If you want more from us, follow us on Twitter, follow us on Instagram, follow us on TikTok, and check out the website chewonthis.io.