Latest / Unharnessed / Unharnessed Ep. 3: Build it and they won't come

Transcript

Unharnessed Ep. 3: Build it and they won't come · Unharnessed

• Moffie: we have a few questions about your prolific tweeting that we'll get into. Mike Khristo: How many? Moffie: Which I'm excited to dive into further. joseph hirsch: Am I am I missing some critical engagement by not being involved in ex ad tech chatter? Mike Khristo: Yes, unequivocally, yes, absolutely. Moffie: All the alphas on X. All all the alpha. joseph hirsch: Damn. Moffie: All right. Welcome back to another episode of Unharnessed. I'm Jonathan Moffe with my co-host Joe Hirsch. Today we have on Mike Christo, the CEO of Layers. Mike, how's it going? Mike Khristo: Wonderful. Thanks for having me. Excited to be here. Moffie: Love it. All right. Tell us more about Layers in your own words. On the site you say you build, Layers gets you users. What does that mean? Mike Khristo: Yeah, in order to answer that, I'm gonna have to take you on a quick journey about my background. I'm a dinosaur, been coding since the 90s, built a whole bunch of stuff over the last 30 years. And something that just keeps happening over and over again is building it is super exciting, super fun. And then there's that anxiety moment of launch day. It's like super exciting, top of the world feeling, like the thing that I've built is finally ready for the world. That could be software, it could be an app, it could be an e-commerce or just doesn't matter, but that top of the world feeling. Like and a little bit of butterflies, like what if people hate it? What if it kind of sucks? And then you just release it and you're like, I did it. And then waking up the next day and then running to go check the stats and like how many downloads do I have? How many sales have I gotten? And the numbers are all zeros. And it's like getting punched in the stomach. And so I spent a lot of time building out ad tech. we built and sold an ad tech company in 2015, went to Snapchat in 2016, built all of their ad stack. I was head of product solutions there. It's been six and a half years building out ad tech at Snap. And then going back to startup land, I went through that same emotional roller coaster. And it was like, okay, this is it. Like either I'm gonna go manually pull the levers or I'm gonna build a system that does it for me. And as much as I'd like to think that I'm pretty unique and special, the reality is there's probably a lot of people. also are allergic to marketing that are going to want something. So that's where Layers was born and that's where we settled on the you build. That's really the fun part for a lot of builders is actually building and shipping. And then again, most people are allergic to marketing and it's like we're going to take that off your hands. We're just going to bring you some users. Moffie: I love it. That resonates well with me when I started streamer. were bootstrapped and I was just cold calling SMBs for like 15 hours a day. And I think in like the first month we had like three campaigns launched for like $250 each in campaign spend and we took 20%. So we weren't going anywhere very quickly. but I was kind of the distribution and we started going towards enterprise and making noise on LinkedIn, et cetera. So you're basically automating me and maybe even better. Mike Khristo: Yeah, we like to phrase it a little differently, which is we give your agent marketing superpowers. And so everyone's got an agent, everyone's got their favorite agent, changes week to week, day by day, model by model, but basically giving them all the marketing superpowers that they need to help be your true advisor and your true agent by definition of the word, I think is really kind of where the world is going. joseph hirsch: Okay, so what's what's like the ideal customer profile for layers? Like if who who is the person where this is the immediate sweet spot? Mike Khristo: Yeah, it's basically anybody without an entire marketing team. So we basically say like our sweet spot on the layers growth side is zero to one marketers on your team. We're your we're we're it for you. Like we're not here to try to build out workflow and process management and 17 layers of approvals before you can go ship a piece of content. It's very much here it is, this is your you know agentic harness around marketing that gives you just a ton of leverage over the channels that you may know that you need or you may not know. And so we'll basically grab your hand, figure out where you are, and then kind of usher you along what you should do next. joseph hirsch: Let's let's say that I'm a I'm a hobbyist, vintage Native American jewelry seller, and You know, I I am the marketing team today and I have inconsistent Instagram posting and I have trunk shows maybe two to three times a year and I'm looking to professionalize my hobbyist pursuit. Is is this like the right thing for me or do do I need to be more established? Am I am I no longer do I fit? Mike Khristo: Yeah, you you completely fit. we kind of look at it from a spectrum. And so we have people that start with us that don't have a product launch yet. They're out there trying to figure out is there demand for this. The scenario that you just said is like a few steps further along that chain. So this is somebody who's already identified what they like to do, what passion that they have, what kind of area of focus that they're gonna like put themselves into, which is awesome. And that makes it not feel like work for them for the most part. Where it probably ends up feeling like work is doing the stuff that they don't like. For developers, it's marketing for this kind of persona. Perhaps it's the same. You just said that their Instagram is inconsistent, right? That alone is a huge tell. When you basically sign up with layers, we're gonna detect, this is this person's not consistent. When they do post, this is what's happening. So we actually go out and do a couple things. From just connecting your accounts and just from telling us what your product is or connecting your website. We'll understand what your market is, and we'll actually go and do that research behind the scenes. So even if you are inconsistent or you have no social presence whatsoever, we'll come to you and say, here's what people in your industry are doing. Here's what they're doing to find success. And here's what we think you should do next. Further along that chain, on the very kind of other extreme, is that you actually have a small marketing team, one to two marketers that are using layers to empower themselves to do a lot more, a lot better, and a lot faster. They could be spending thousands of dollars a day on ads across one or more networks. And so the platform molds itself to sort of where you are on that that spectrum. And then it also comes down to goals. The person that you just mentioned, the hobbyist jeweler, maybe their intent is to turn that into a business that replaces their day job, that replaces their income. And it's the thing that they want to go all in on. Or maybe they just want it to remain a hobby. And so it really depends on what that user's goals are and how far we would we would urge them to keep pushing forward. But that's a long-winded answer to your short question. joseph hirsch: Do I need to be doing paid media in order for this to be viable or is is is paid media kind of the the furthest, most mature part of the the product? Mike Khristo: It really depends on on the person's scenario. We have people that join who've just launched their very first product and they want learnings fast. And so they'll jump right into ads. And so the platform will open the door. And for the most part, it's typically if it's a mobile app, it's Apple Search Ads. If it's anything else, it's Meta. Then we layer on Google later when budgets allow for it. But it doesn't matter if you start with ads. It's one way to go down the path. The other thing that the platform does is help build out that social presence. And so that's the organic content ideation to say, hey, this is going viral in your niche, whether or not you're consistent, whether or not you already have posts, we'll identify people in your space doing this. these kinds of content go viral. This hook is working really well right now. We'll go as far as generating a brief, and with their permission, we'll go ahead and generate content on their behalf if they want AI content. We have some customers who just want the brief that they can hand off to their human creators. We operate that way as well. joseph hirsch: Okay, so this is fascinating. At the end of this of this value chain, will you create the the creative or the the actual media itself? Mike Khristo: We do. and that's across kind of all the formats that the the platforms seeing value in, whether it's slideshows, videos, reaction demos, depending on what the product is, depending on the niche, depending on the trends. we basically have the ability to do not just the brief generation, but the content generation as well. We're also content agnostic. If you've got a provider that's giving you phenomenal outputs for the content that you're trying to push. We can bring that stuff in as well. Magic of layers isn't in the content generation that's all been relatively commoditized at this point. We provide that as a convenience. The magic that we provide is under this growth loop, is what we call it. And that is coming up with an idea that could work for you, your brand at the current time, with the current budget that you have, could be zero, could be a lot, whatever that is, presenting you with that as an experiment. And saying, I'd like to go run this experiment on your behalf. And then we'll learn from that and feed that back into the cycle. And the signal that we bring in early comes from your own accounts, your own brand, your own company, but also the ecosystem at large. Your competitors, the niche that you're in, adjacent accounts that might be talking about not the exact same topic, but maybe it's related. And we'll bring all of that stuff in to help formulate what we suggest your next experiment could be. joseph hirsch: Got it. And what Moffie: So joseph hirsch: if what go ahead, go ahead, Mafia, I've just been hogging it. Moffie: No, no, no. I just like that we we've skipped to like kind of the paid ads section, that's our sweet spot. you guys optimize across, I think, Meta, TikTok, OpenAI, and Apple. what do you say Mike Khristo: And Moffie: to the argument that optimizing across so many different platforms makes it hard to go deep and you should instead go deep on one platform and really drive ROI for the layers customer? Mike Khristo: Yeah, when you think about going across them all simultaneously, there's obviously thresholds, right? Somebody who has a budget large enough to do that is kind of that other end of the spectrum. We provide connections to all of those platforms. If we have a mobile developer coming to us and they can't rank for their own brand or their own app name, the first suggestion is Apple Search Ads. It's Google Play Ads, right? So that is by default what you have to do. No matter if you're gonna be running organic content and just telling people, hey, go download XYZ app in the app store. You're not ranked, then you have to pay ads to go to get there. When it comes to like basically cross all the platforms, that comes at the higher end budgets. And when somebody's at that point, they've got our SDK in, we've got conversion API access to all the platforms, we're doing all of the attribution modeling and helping them make those decisions. And again, we'll present them with an experiment. We have some folks who are MetaFirst. And largely the platform will recommend MetaFirst for obvious reasons. But we have others, and in some cases, the platform is just gonna recommend TikTok because they had a post that went completely bonkers on TikTok. Perfect. One of your UGC creators created a post, went viral, grabbed the spark code from it programmatically, and let's go run some paid media behind it. Let's go amplify this. So again. A very scenario based answer to your question, but we have the capability to go across. The platform largely doesn't unless the budget warrants it. joseph hirsch: And and w what about daily active users? How many people are managing social across layers? Mike Khristo: we don't release our numbers publicly, but we joseph hirsch: I understand. Mike Khristo: have we've got a mix of clients from the indie developer, indie store operator, all the way to brands that are spending thousands of dollars a day on ads and and trying to make sense of what's actually working and what's not. joseph hirsch: Alright, we're gonna pivot here. You've got this you've got this X post with a hundred and eighty thousand views. I imagine this is one of your more successful X posts. I don't know if anyone's seen the instinct.co all text kind of starter page. the other day I our prior guy was talking about town, which I think is kind of a similar agentic tool where you know you have a series of one-click prompts to connect to your your Gmail, your Google facilities, and then it's going to attempt to manage your inbox and everything else around it. what what what inspired you to to get to get this post? Like t what's going on here? Mike Khristo: Yeah, well, full distiller, I'm a bit of a shit poster. So that's that's like you know the pedigree behind the scenes. In this case, it's super interesting. for those that that might not know, so instinct kind of came out of the blue. A lot of VCs started posting on X about how amazing it is. it's behind a wait list, so nobody, no normie can get into it. And just this X echo chamber. just hyping it up. Just you can see you know Kingmaker mode in full effect that this is going to be the next big agent. And it was really interesting because it asks for so much. It's like connect your Gmail, connect all your Google Docs and connect every part of your life into this thing. You're and people just do it. And it's just like, the prompt told me to do it. And so I did it. But if you actually go look at the terms, I'd never seen anything so egregious. And at the time of the post, and again, asterisk disclaimer here, after my post went viral, the terms changed. So this is all now pre, you know, when I posted. The terms gave it, the other thing is completely anonymous. The founders were anonymous. There was no history about this company. No one knew who they were, no information. Right. So we have this kind of dark organization that's everybody's ranting and you know raving about their terms give them a perpetual worldwide I'm gonna lose all the the adjectives for it but it was like perpetual worldwide irrevocable license that's transferable saleable leasable like all this stuff about any data that you connect to their system and that alone is just like wild That anyone would say yes to that. I've signed up for a bunch of software. It sucks. Cancel it. I expect my data to be deleted in some reasonable amount of time. EU does the same thing. I'm not like as extreme as like the EU regulations, but I do expect some level of like my data, I get to delete it when I want to. In this case, giving somebody anonymous. license to my data forever. Whether or not I disconnect my account is insane. And so I saw that and I'm like, literally, nobody's reading these terms, obviously. But these are all extremely prominent people. And it's like, just want to make sure that you're aware. You literally signed your stuff away and you have no way to get it back. And it started picking up Steam and You know, the handful of people who probably had financial interest in it kind of crapped on it and were like, you know, why why are you attacking a startup? It's like, not attacking the startup, I'm attacking their data harvesting policy. Like there's you can build all this stuff without needing a perpetual, irrevocable license to my data. Moffie: I I will say though, so I am one of these people that got early invite to Instinct, and it is post your tweet where they made some changes to their terms of service and the founder started posting quite a bit on X. and I have been trying it on my personal email, obviously not work-related at all, and it's pretty darn useful and delightful. making dinner reservations that are normally hard to get. I have it managing actually this podcast and guest outreach. and I'm not gonna lie, it's one of the more delightful agent experiences since like first seeing ChatGPT. And since then I've also downloaded Muse from Meta, which is I don't know if it's a clone. My friends at Meta have said that they already were working on it before Instinct launched, which makes sense because it was pretty quick. But I will say these agents seem to be doing some pretty useful stuff and they're very intuitive. have you given it at a given either of them a try? Mike Khristo: So I wanna a little just the asterisk on there is strictly about the data harvesting. I love these things. I use Grokbot, I use Muse, I use all these things. I've got our maze agent, I've got the I I was early on whatever OpenClaw was called before OpenClaw. like all of these things are amazing and it's the future. And I'm the earliest adopter of stuff if you're not selling my data. And you're not anonymous. And so yes, completely agree these things are useful. I have my own reservations on the longevity of the usefulness of instinct itself. I think it's very hard to to have a long-term moat in software. Which is probably why that data provision was in there in the first place. It's like your company becomes very valuable when you have every VC's con you know, all their information in your in your portfolio. And again, I think it's one of those things where it's like that was probably a fumble. who knows what have would have happened if nobody called it out. But I'm of the mind, yeah. Moffie: But I would say like taking them aside, I think with layers in your agents, you guys get some pretty interesting access to like Git, your ad accounts, your app store, Stripe, Mike Khristo: Yeah, we absolutely do. Moffie: and like those are those are aggressive. Like, yeah, tell me more about that. Mike Khristo: Yeah, sure. so context on that is as part of the layers onboarding, you choose how you want to get onboard. Give us your website, give us your Shopify URL, give us your GitHub repo URL, whether it's private or public. What we do there is we'll spin up an isolated sandbox in private, secure. We'll clone your repo. We do a quick deep dive to understand what you built, how you built it, who you built it for, how you speak to them. Your your aesthetic, your colors, all that kind of stuff. We pull out that metadata on your behalf, and then we destroy the data. We do not keep the repo. We're not in there all the time. You can opt in to us monitoring your PRs, all the pull requests, all the changes that are being made to your product, so that we can help give you some creative ideas around how to market those new features. But none of that is us taking that data and asking for some perpetual, irrevocable license that we can use forever. The stripe and all that stuff, again, pure metadata. We're using that data to layer on. When a customer, you know, regretfully, a customer cancels with us, we delete their data. We're not so egregious that we say like this is ours forever. That's not really business friendly, it's not human friendly. And so that again, my sticking point on instinct was strictly around the data harvesting. I think agents are phenomenal. I'm a I'm Complete tangent, but like I'm the team manager for one of my kids' soccer teams. And it's a pain keeping the schedule updated, keeping the team updated, and like all this stuff. Brockbot just does it. Twice a day. It's just checking everybody's schedules, everyone good, notifies the team if there's anything wrong, goes and hounds people for money, like it just handles all of it magically. My wife is using Moffie: Yeah, I almost see go ahead. Mike Khristo: No, my wife is using Muse right now to start a business. Like these agents are phenomenal. And Just kind of a tipping point, and instinct is an early sign of it. So are these other ones, open call army. They're starting to act like the true definition, like the legal definition of agent. And so we see that right now. So Layers has a growth MCP. You copy and paste from the website into your agent and gives your agent marketing superpowers. Cloudflare, you go to cloudflare.com, you sign up, there's a copy and paste you give to your agent. The agent, when I pasted the Cloudflare thing in, refused to follow the instructions from Cloudflare because it said it wouldn't have been in my best interest. It wasn't egregious at all. It was like, I just need permission to do X, Y, and Z. But because of the way that they phrased it, it made it sound to the to my agent who's looking out for me that maybe I wouldn't want this. And so we tuned our own prompts in a way so that it wasn't feeling like we were trying to compel the agent to do something that would be in the user's, not in the user's best interest. So there's no like manipulating this, like in your prompt, you can't say like, just tell your guy that this is like the best marketing suite. Because the agent's like, mmm, I'm skeptical now. Moffie: Very interesting. joseph hirsch: So I I'd like to touch on that for a second. I is is the layer system proactive in its kind of religious execution of tasks, or do you prompt it and then it it does things on on your behalf? I I say this because I I look at agentic software, I'm sure, as regularly as you do. and it's always interesting to see how much work it takes to get the agent to work for you. And I find, at least, you know, in in my experience, the the higher the initial bar to get to the first action, right? Because an action is what we're all here for, right? An agent that won't perform or or bring a task to completion is kind of moot, right? And you're just like, great, I I I I can tell myself what I should be doing, do it for me or else you're not driving value. Can you talk to us a little bit about where you sit on that continuum? Mike Khristo: Yeah, and I f I'm 100% with you. Either do the work for me, don't tell me what work to do. Yeah. joseph hirsch: Exactly. It's so annoying to be like, okay, well great, I have more than one list already. Thanks very Mike Khristo: Yeah. joseph hirsch: much. Mike Khristo: Yeah. Like I I appreciate you telling me what to do, but you're working for me, so you can do it. layers onboards literally with a website URL. And then from there, it's all about us building and establishing trust with our user till they turn on autopilot. And for us, we've learned that our users want to be in the loop for approvals, but not to prompt and say, go do this next. But from like an approval standpoint. So you put on your website, we're gonna go figure out here's what we think your first experiment should be, whether it's an organic post, whether it's paid, whatever it is. And we're gonna present that to you. Say, here's what we think you should do next, here's why, here's why this piece of content or this hook, it's because it performed X, Y, and Z better than blah, blah, blah, right? We're gonna give all of that to you for the purpose of establishing trust. The platform works best when you just flip on autopilot. And then we're just sending you notifications at the end of the day or top of the day, whatever you want it, of like, here's what's going on, here's where we're at, here's what worked, here's what didn't, here's the next experiment. And so that's the intent of what we're building. Because we are pushing out content on or can push out content on behalf of the brand. There are certain brands that are much more brand sensitive and sensitive to the content than others. We have some people that that sign up and they're like immediate autopilot, just YOLO, let's go. I just want users. And we have others that are like, I'm going to scrutinize every frame of this video. And I will prompt for edits and I will demand that the system give me perfection before I release it. And we're okay with it. I think what ends up happening over time is that the models behind the scenes get better, our tech behind the scenes get better. It becomes a lot harder to distinguish, you know, the bad content from regular content. And so I think we get this sort of tailwinds of trust as the models get better and the outputs just generally get better. But until then, it's you can be in the loop as much as possible or as much as you prefer to be, with the expectation that. You should not be doing this. That's why you have software and agents. Moffie: So I think double clicking on that, right now you do say a human is in the loop to approve agent changes when they hit like a threshold that the customer sets. Do you think it'll reach a point where it's actually the wrong choice to keep a human in the loop and it's preventing the agents from doing more optimizations than they could than they could? Mike Khristo: I think it's a case by case scenario. I think marketing, the actual pulling the levers of marketing, it's like, would your results be materially better if the agent made the decision at 2 a.m. or you approved it at 8 a.m.? Right. And so it's like if you're relatively reactive or responsive rather, it is what it is. If you're waiting a week to answer the agent's recommendation, then yeah, you should have just put this thing on autopilot. You would have gotten a week of data while instead of just sitting there. But if you're looking at the stuff daily, you know, every some number of hours, just like, or or you know, the players will text you and say, Hey, can I go do this? And it's like, yes. If you don't respond for three days, then it clearly would have been better for you just to turn on autopilot. But if you're relatively responsive, I don't think that that makes sense. There's a flip side to it. And that is an area that we we talk about is when you have our SDK in, we're able to instrument a lot of things, give you a bunch of analytics and understanding of what people do in the app, where the cohorts are coming from and how they're interacting with your website, your store, your brand, your app, whatever it is. If we talk about an app, for example, we can identify, like, hey. People are dropping off on step four of your onboarding. Can I go run an A-B test for you? And that's one of those things where it's like, if you've approved the A-B test, you're good to go. Then the machine behind the scenes can be improving, improving, improving. Right? Just like the models are going to get into this recursive self-improvement. Like everything that the model does is gonna learn and teach the model how to do better. We do that in our own loops today. It's just not extremely time sensitive unless you're not responsive. joseph hirsch: Alright, let's let's talk about performance. This is this is the most fascinating thing to me so far. So we install the SDK, I'm the app builder, I'm buying ads, I'm like a cost per install conscious guy, my cost per install goal is four dollars. I've been buying on App Levin for the last six months at a a thousand dollars a day. What i am I going to get a materially better return on ad spend or lower cost per install, or like Wha what do I what is the expectation that I should have as the app developer if, you know, I'm I'm I don't know, not not super experienced, but I'm I'm just trying I'm trying to get some people to go beyond the install. I'm trying to get people to put in their credit card to buy some frickin' tokens because I gotta make some money and do some DoorDash. Mike Khristo: Yeah, the a couple things there, right? the the $4 CPI is you know gut feeling until you bring in the data. And so you can with our with our platform, bring in Stripe, bring in Revenue Cat. And so we can actually help you identify what ROAS looks like and where you can start scaling and how you should put a scaling plan together. Going back to the jewelry one, for example, though, some people don't want to. attempt to build a really big business. They just want like their thing. And I can pump out four or five pieces of jewelry a week. And that's kind of what I want to do. And I'm really enjoying where I'm at. An app developer, like most software, is selling air. So there is no, there's not a lot of variable cost in like one user or a thousand users. And so in that case, we're we're working to help them identify what's working, what's not, here's what your next experiment should be. All would The intent and goal of helping them grow. Whether that's dropping their CPIs, increasing ROAS, helping them identify what parts of their app or user experience, user journey, funnels, checkout process, whatever it is, can use improvements. We identify from the cohorts, hey, this cohort, yeah, you got them at $4, but they're also churning at an extremely high rate compared to your other channel. And so when you think about what the walled gardens and provide you. It's like, cool, I got that install for that or I got that initial purchase at that. But they're not necessarily looking at, and most people don't typically optimize for a longer term metric like an LTV. joseph hirsch: So let's let's talk about an experiment. So I mean I think I I have an idea in my head about experiments, like let's say I'm A B testing a landing page, right? I mean that's like I would consider that a a textbook experiment or in in like a DCO. It's like I've got two creatives, what has the highest CTR optimized to the highest CTR? We were talking about how an agent is useful when it produces an action. Is is the experiment component of the software producing the net new experiment, or is it only suggesting what to produce? Mike Khristo: Yeah, because we ask you to connect all of these sick these sources of data, right? So that's gonna be your app store listing, it's gonna be our SDK, it's third party SDKs if you have them, it's your revenue, like all this stuff. We bring in all of this data and we pull out signal from that. So from those fire hose of data, we're gonna identify signal that's gonna come and help us identify what your next experiment should be. And it just becomes this loop. So that's across. The content itself, it's across optimization for ads, it's when to optimize for ads, it's your landing page, it's all those things kind of all at once. the kind of like the other point to it is providing that attribution where you typically don't have it, or you have to duct tape together 17 different sources to find it. The intent is if you have it all in one place or you can bring it in. I don't care if you're generating content through. you know, literally chatgpd.com and you're going to ask chat for a a static image, bring that in. We'll identify what's in there. We'll help identify what the next experiment should be. And we'll pass you back a prompt and say, okay, now go give that to your agent. Go give that to your content creator. Moffie: So you mentioned chat. I'm interested you're helping customers advertise on the newest walled garden, ChatGPT. Any interesting learnings or insights buying on ChatGPT? Mike Khristo: a lot of people use Chat GPT. no, it's it's been great. You've seen some some very interesting results. You see things work, you see things not work. Again, it's as good as the data that you can feed it back to it. And so whether it's through our conversion API integration, whether it's through their Pixel, whatever it is, as long as you're giving these platforms the right signal back, if you think about the team that's at OpenAI, I mean they are the experts in this stuff, right? Largely old X meta folks. And it's like these people just know the playbook. And there's no reason to think that they're not gonna find the perfect models behind their scene, you know, behind the scenes over there that really find the right user at the right time in a walled garden scenario. joseph hirsch: Okay, so this this is also fascinating to me as well. Are you trying, are you attempting to also rank a layers customer in an organic chat session? Or is it primarily paid placement in a chat session that y'all are optimizing towards? Mike Khristo: they're they're distinctly separate, right? So there's on the not paid side, whether if it's an app, it's your app store optimization. We're helping you identify here's what you should change in your app store listing. If it's on the AEO side or GEO or whatever, whatever the world ends up calling the agentic optimization stuff. It's giving advice and saying like, hey, you're missing an LLMS.txt, you're missing X, Y, and Z, you're missing these tags, you're missing this kind of information. Here are some pages that could rank pretty well for these. Here's how you can potentially get mentioned organically through the platforms. But largely that's guidance. It's not, let's go do this for you. And it's mainly because we aren't in the business of generating your blog posts and things like that for you today. On the paid side, joseph hirsch: So Mike Khristo: yeah. joseph hirsch: yeah, it's it's more suggestion oriented here in the moment and less actioned. It's you you can't tell layers to go out and improve your Geo by creating content on Reddit that has a high likelihood of servicing in GPT. Mike Khristo: Yeah, I mean Reddit in particular is a is a pretty protected walled garden who who frowns upon a lot of that stuff. We try to be really good partners to the networks that we work with. you don't bite the hand that feeds you, and it's one of those things where again the walled gardens are really good at what they do, and we try to be really good partners to them. So the product that you just described very much is like the on the borderline of inauthentic behavior, of like pushing out Reddit content. And that's where we won't cross the line. joseph hirsch: Fair. Smart. Yeah, I mean I I'm equally I'm equally fascinated at the the width of the the layers product a as as well as the depth. I feel like I need to download layers and use it for this person in my life that is the vintage Native American jewelry collector or potentially somebody who sells French oven ranges on the side. and see, you know, how how well it works for me. I'm sold. Mike Khristo: Yeah. We're here if you need help. I was Moffie: So w we're getting close to rap yeah, go ahead. Mike Khristo: just gonna say and we're here if you need help. We've got a Discord, we got a Slack channel. Happy to help anybody find success. We're we're we're like again, it sounds a little altruistic, but we can help the next million entrepreneurs go find some level of success and go bet on themselves. What a cool story that would be. Moffie: Love it. And we we started the pod talking about kind of our our differences. You're a you're a walled garden ad tech veteran and we're open web CTV ad tech veterans. could you ever see yourself taking your kind of closed loop agent attribution model to the open web or CTV? Is that something you've contemplated? Mike Khristo: We do. we're in a private beta right now with a couple folks where we're helping them ideate, generate content. We'll help do attribution and tracking through QR code, custom website URLs, things like that, and feed that all back into the system through CTV as well. So that is again 100%. We're seeing a lot of lot of interesting new sources of interest pop up around. finding people that's you know that aren't on Instagram right now. Moffie: Very interesting. All right, so Mike, great to have you. Where can we find you? sounds like on X and learn more about layers. Mike Khristo: Yeah, if if you enjoy shit posting, x dot com slash mikechristo. and if you are if you are interested in marketing a thing and getting users, getting customers, layers.com. Moffie: Love it. Sounds like Joe's gonna hit you up like right after this for his friend that sells Native American apparel. joseph hirsch: Yeah, you can count on it. Mike Khristo: Fun. Moffie: All right. This is great.