Do AI Wrappers Even Have a Moat?

0:00It's like the rainbow sprinkles on top of the ice cream, which for my daughter, that's what does it. It's the rainbow sprinkles. Any ice cream is just immediately better in a sense, right? Welcome to the Talking AI Podcast, where we talk AI with both experts in the field and early adopters. I'm your host, Matt Page, and we're here to demystify AI for you so you can get some value from it. Let's talk some AI. AI rappers have no moat. They're not defensible, and ChatGBT is just going to continue to eliminate them with each new release. Maybe not. Maybe we're thinking about this in the wrong context.

0:38And here with me to discuss that is Clay Ostrom, founder of Map and Fire, also an expert in positioning. So we're going to get into some of that on positioning for AI products with this crazy AI washing we have going on. But he's actually built an AI native solution in support of his business, which I love that use case and how people are starting to do that. dare I say it's an AI rapper. However, there's real value being provided to his business, to his customers, but welcome to the show, Clay. Great to be here. And I am eagerly awaiting to challenge some of your upfront setting up there. Yeah, I agree.

1:15There's a lot of variation and diversity around some of the apps being created right now. Yeah. And it's funny, I was 100 % on the bandwagon of AI rappers have no defensibility. We even did a podcast on this exact topic last season where no defensibility. They're just going to get eliminated with each new release of ChatGPT. Don't even bother building these solutions that are just basically AI wrappers. And then you see tools like Cursor, the fastest one to 100 million. See all these unique tools popping up within very niche use cases. And it's this light bulb just hitting my head. And I'm curious your take here.

1:56And then we'll actually get into what you've built because I want to use it as an example for how folks should be thinking about it. But humans are lazy. We are lazy. And we're just creatures of habit. That's how we're built. We work smarter, not harder, right? So when we have something that makes things easy, is very domain-specific to our use case, specific to our pain point, we use it. And go back to how SaaS blew up. all sass is basically you could do any sass product in excel but you don't because it's a pain in the butt but curious what's your thoughts on that are you in the camp of ai rappers bad ai rappers good or somewhere in between well i mean like i said it's there's so many flavors of this and i think so many levels of what you might consider to be a rapper and obviously we can dig into this more and I would love to talk I can talk more specifically again about what I built and how I I don't necessarily view as a rapper I think there's more going on there so the short answer is yes I think if you're just a rapper it's going to be hard it's going to be hard to defend that but if you can bake in a little bit more of your specific point of view and kind of how you approach stuff and also just like you said the specificity of a specific use case a specific industry i think there can be some defensibility around that well let's get into that what in your context and we can kind of like brainstorm on this live i know for people listening they'll find some value in this but what makes something valuable or not an ai wrapper just that bit extra that you're talking about, what attributes does a solution have that make it still something worth doing, I guess?

3:48Yeah. I mean, so to me, again, it is like, are you using the response back just sort of like face value as is, or are you doing something else with that data? Are you taking pieces of data, analyzing it, configuring it in some special way, applying some kind of, you know, special sauce of sorts like to it. That was exactly what I was thinking about when I built, you know, our app. It was like, I don't want this to be just a wrapper. I don't want this to just be as simple as we plug in a prompt and you get a response back. I want to be putting our stamp on this and applying some of our own thinking behind the scenes around how this works and how we present and configure the data.

4:31I think about it a lot. And the comparison I make often is with the hrefs sim rushes of the world where yeah they're going out and just scraping data right like there are tons and tons of data but they're also applying their own smarts on top of that they're taking the data and they're coming up with scoring systems and you know additional analysis that is beyond just what you would get from you know a google search you know which obviously anybody could go out and just do some of that work themselves, but they're getting all additional value from these tools. So, so basically it's like the rainbow sprinkles on top of the ice cream, which for my daughter, that's, that's what does it.

5:13It's the rainbow sprinkles. Any ice cream is just immediately better in a sense, right? Yeah. Yeah. I mean, again, I would say it's at least rainbow sprinkles. Maybe it's also chopped nuts and some whipped cream. Like it's a combination of stuff, you know, going on, But yeah. Quick break in the pod. If you're listening to this podcast, chances are you've been thinking about how to actually use AI inside your business. And that's exactly why we built the AI Opportunity Finder. It's a free tool that helps you uncover high impact, tailored AI use cases based on your business, your goals, your pain points, and your industry.

5:49No fluff, no generic use cases, just real ideas that fit your business and the ranked by ROI potential. It takes about three minutes to run and it's like having your own personal AI strategist for free. If you want to try it for free, check out the link in the show notes or go to hatchworks.com backslash AI dash opportunity dash finder. The biggest thing to me, you hinted at it, is leveraging proprietary data. Because as we all know, AI has literally been trained on the entire internet. It has everything. But what it doesn't have is your proprietary data that it just can't access via the internet.

6:27So to your point earlier, you know, being able to leverage that in a unique way that only you possess, I think is like one of the most immediate ways when people are building solutions. Can I have that? That equals differentiation. That equals defensibility in a sense. And I also think there's this, I'd be curious your take here, because I have this like perspective that incumbents are going to be experiencing the biggest, most impactful innovators dilemma of our time, of ever, in my opinion, versus new entrants in the market. Because new entrants, I mean, there's a case where you could say you don't need as many humans to build a business.

7:09And they're, what is it, the constraints or the whatever the Steve Jobs quote is, constraints drive innovation, right? So they're going to innovate with less people, leveraging the latest and greatest tools and AI and everything there. And incumbents, they have massive teams. They have lots of bureaucracy, lots of process. But what they do have is lots of data as well. So I know I said a lot there, like thoughts on the data side, thoughts on the incumbent battle between new interests and incumbents. Who's going to win there? What do incumbents need to do to actually win? Yeah. Well, that's I think and that's the other piece of the moat puzzle for me is the data accumulation side, like you're saying.

7:52And part of, to me, that's the other piece of what I wanted to do, again, using our app as a reference point, was building up a catalog, a library, a database of all the analysis that we've done so that when you come in, you have instant access to that. You're not having to do a bunch of stuff from scratch. You can just get all that stuff right away. But by building up that analysis, which again is based off of our own, you know, semi kind of proprietary, you know, way of going about things, it does start to create something that you simply can't get anywhere else. You know, and I think that's the approach that people need to be taking with these kinds of apps.

8:31Yeah. And I think the other angle that I keep leaning in on to is domain specificity. Yeah. Right. Can you focus in on a niche that's underserved and then whatever unique workflow process, whatever it is tied to that, I think that's the other angle too. And what's interesting now is anybody now can build something leveraging AI. We'll get into this in a bit. I know your solution, you leverage more of kind of no code tools, but I was playing with this morning, N8N or Naiten, I don't know how you pronounce it, but it's like, it's, you know, the new version of Zapier, you know, equivalent to make.com and just playing around with it saying, okay, what can we automate in our business?

9:15We have a ton of these like manual processes. And instead of me just like poking around in the tool, I asked ChatGPT, right? And I had this like video I saw on YouTube or something. I was like, okay, how do I do this? And I gave it the image. I gave it the link to the video. And I swear to God, it literally took me through step by step how to do everything. And it got complex. And what I learned, Google has a, you can create custom search engines. So I pointed this one to LinkedIn, right? And it had a, you can do, they have a web scraping tool. So now it's going towards LinkedIn. It's scraping LinkedIn.

9:51so you don't get in trouble with, you know, the plugins and stuff that LinkedIn hates. It scrapes LinkedIn. And then I did all these like web hooks, hooked it up to Google Sheets. And it's like, I can put in some criteria and it's doing a search in LinkedIn and returning back people. And then the next step's like, okay, let me hook that to AI and then what we offer and then just kind of automate that whole outbound thing. But it's like that companion piece of leveraging AI in the building process, whether it's like showing you how or just like all the way to the extreme of cursor or replet or lovable and it just builds the thing custom for you yeah i know i was playing with lovable recently and like you said the app i have is a no code app with ai in the background and i was playing with lovable which maybe a lot of people know now is truly like a text entry to create an app kind of, you know, solution.

10:46And it was pretty wild. I have to say, I mean, I have a CS background. So, you know, as you're watching it, write pages and pages of code for you to build a solution. It was pretty, pretty crazy to see what it could do in a very short period. And it just keeps getting better is the crazy thing. And that's the other piece. And then we're going to get into the tool. Cause I want to like actually go through it. There's, you can build with AI and then you can integrate AI into your solutions. It's kind of those two paths in a sense, but the other piece, yeah, it's almost like you can build for an audience of one, right?

11:19You don't have to have this huge plan to say, oh, I got to scale this thing to a million users to get an ROI on it. That's like no longer the case in a sense. And I had another point on the moat and rappers and I forgot it, but we'll come back to it. But let's get into - Well, before you, yeah, before you jump into that, I was just going to say, you know, to that point about an audience of one, I think it's really interesting, like Going back to your earlier point about how many resources you need or don't need these days to create something. I think it was Sam Altman who said something like, we're going to see the first, I don't know if it was a$100 million company built by a solo premier.

11:57No, I think you had a B in there. Maybe it was a B in there. You're right. It had to be a B. It wouldn't be sexy if it wasn't a B. But I really think that's where we're headed in a lot of ways. It really, the resources needed to build something are obviously dropping dramatically. And it's going to come down more and more to marketing and positioning and brand and being able to, you know, get whatever audience you are trying to get, whether that's one or hundreds of thousands, you know, it's going to come down to your ability to execute on that. And that's a great transition. And then there's the whole agentic side of it, which is like a whole nother conversation.

12:32But that's essentially what you do. You're kind of like a positioning expert, Map and Fire. That's your agency and you built this tool in support of it. So, hey, I think set some context, like what do y 'all do? And if you want to, like even just pull up the tool. And for those listening on just, you know, Apple or Spotify, we will be very detailed in our explanation of what's happening. Yeah. So in Map and Fire, we do a mix of customer research work and we do a lot of positioning, messaging work for brands. So I'm spending basically all of my time thinking about this stuff. I'm thinking about like, how can brands set themselves apart, especially right now with competitive, as we're saying, like tools, instantly accessible can be built in a day.

13:17Competition levels going through the roof. How do you separate yourself? It's hard and it's only going to get harder. Yeah, exactly. But yeah, for us, it's a mix of, you know, the customer research piece, I think is always going to be critical. Like, obviously you need to understand who you're trying to serve, what their needs are and, you know, and also understand the competitive space where you can separate yourself, you know? And I think the tool we built is one piece of the puzzle to simplify some of those things. I'm still of the belief, and this maybe taps into some of your agentic side of the equation, but I still think we need humans.

13:52We still need people to be able to kind of pull all these pieces together and make sense of it all. But at least we can get data more accessible and make it more efficient to get it. Yeah. And maybe we hit that point now. And then maybe you should be able to share and start pulling the thing up as we go. Oh, sure. Yeah. But I agree with you to this extent. I think the role of the human changes. I think we go from the doer to the orchestrator leveraging AI, right? And I think the point makes a lot of sense where AI makes you better. AI enables you. that's 100 % true. I do think though, because I feel like there's some folks that say, oh, you know, there's not going to be any job disruption.

14:33I think we may be naive in thinking that because I think some jobs will go away, but I think what will happen is new opportunities and jobs will emerge. So I think for all of us, all the people listening, the more you learn into AI, learning it and leveraging in what you're doing, the more, you know, okay, back to both the differentiation the better differentiated and defensible you are going to be in a sense that's kind of my hot take on it yeah i don't i mean i don't even think it's you know a super hot take i mean i think you're right i think over time more and more stuff will be doable with ai and other tools and what the humans need to do or where we're really good is going to continue to kind of narrow down.

15:22But it's still, and again, maybe I'm naive, but it still feels far away. Some of it, you know, and at the same time, I feel like every morning I wake up and I'm like, oh my God, they can do this now. Man, that's a real thing though. Cause I keep going through these moments of, yeah, okay. We still have, we still have some things that we kind of own. And then the next thing comes out and I'm like, holy shit. Yeah. You can do that now. It's every day. There's just something coming out. I know. And then I'll talk to people who are not deep in the space and they're just kind of living their life and have no kind of real idea of what's going on.

15:58So, you know, it feels very immediate and very close to us. And some of it is. And then I think for. Because we live in it every day. The larger bubble out there, it's still kind of a buzzword. It's a, you know, oh, that's interesting, but this doesn't impact my life in any way. And maybe those people just get surprised one morning. Well, you know, that's a great analogy. And I promise we'll get to the toy here in a second. But I think incumbents and companies are going to go through that same thing. I think it's going to be like, oh, you know, it's fine. We're using AI. It's good. You know, we're good.

16:37We're. And then all of a sudden there'll be a new company startup, whatever it is that does the thing they do in just a whole, wholly different way, totally different business model, totally different cost structure, margin structure, everything. And then it goes from, okay, we have our brand equity and people just, you know, cause humans don't like change. They will do anything they can to avoid change. but then I feel like with the big incumbents it'll be like a dam breaking and once the dam breaks like you just can't you can't move fast enough and it's kind of what you're saying at the personal level in a sense too it's you know it'll creep up on a lot of people and companies I think yeah no I think that's a great point I think even we think of these brands as being so sophisticated and on top of everything but they're also massive they move incredibly slow the you get the harder it is to adapt and change.

17:31And I think there will absolutely be a lot of these big companies that just get heavily disrupted in a short period of time. I mean, we've already seen one version of that with just with OpenAI, ChatGPT, and Google. I think people are now shifting behaviors. I mean, behaviors that have been now baked into us for decades of like, Google is just the way to do it, to find information and it's not anymore. That is a great example. And I have, I've almost fully switched to chat GBT perplexity or something for almost all of my searches. The only ones I don't do, I still haven't completely switched over is if they're like location specific, like restaurant.

18:16And to be honest, you know, chat GBT and perplexity can do those great. I think I'm still just very familiar with, you know, how it shows the map and I can click on the details and all that, but everything else. And it's in another point too, on that it's a different way you approach stuff. Right. Cause with Google you search, you go, then have to research through all the different links and whatever they are, blogs and things like that. But now you can kind of use it as a partner with, you know, name your AI tool, chat, GBT or whatever. Like the other day I was having issues with my garage door opener and I just put it on voice mode with video and had it walk me through the steps there.

18:54I had done the same thing with like things on my computer, but it's a different like use case where I would have had to call somebody and get help. You can now just go to chat GBT. Yeah. And I mean, and to your point about people's behaviors changing slowly, are we being resistant to change? That's a great example of how in many ways that's just a more intuitive use case or like a more for us, like we're used to having to say, oh, I've got to go to my phone or I've got to go to my computer. I've got to type in a keyword that will help get me to part of the answer. And now it's, I can say in plain English, what I'm trying to do.

19:30And it will give me, you know, the steps to actually accomplish what I'm trying to do. So exactly. So walk us through this. So this is almost, think of it. It's like a tool that you use internally. It's also kind of a cool, like smart Legion tool as well, but I love the name, like smoke ladder with the map and fire. And I'm assuming like smoke signals kind of like where you're going and whatnot. Yeah. That's yeah. That was the inspiration was, you know, thinking about when you get your positioning locked in, it's like a smoke signal. You're putting up, you know, something that's high above the competitive noise that people can spot from far away and, you know, draw them into your brand.

20:05So that's how I think about, or at least that's kind of like the analogy with positioning. But yeah, so again, I know some people are just listening to this, but yeah, just looking at the homepage of the app and there's like a basically a search bar. I wanted to create an app. You could just kind of jump in and get started right away. That was also important to me. Just making it very easy. I used an example here. I've been, I don't know if you're familiar with this tool called Tela. It's like a video recording tool. I've been using a bunch for content. It's been super helpful for me. I'm a big Descript fan.

20:39I've used Descript a lot too. And I ended up switching to Tella because it had some better configurability around like can do side by side video when you're doing app demos and things like that. It makes it makes it very seamless to adjust those things. So anyway, jumping into the actual app, it's designed to analyze two major things. So Tela is the company we're going to investigate. That's the one we're going to run through this exercise, right? Exactly. So we'll do a quick kind of analysis with Tela. And the app SmokeLadder is basically designed to look at your messaging and your positioning and help you analyze those things, figure out where there's opportunities, improve it, all that good stuff.

21:21So this first screen here is basically an overview. It's giving you just kind of like a quick rundown of some of the messaging analysis and pushing points. For those watching, it's got like scoring in here. Back to your point earlier, it's taking like data insights, but then it's adding a layer of, you know, kind of metrics and things on top of it in a sense. Exactly. Yeah. So there's a few different kind of steps with this. The first piece is the message clarity, which this will come in and just kind of look at your messaging. And then it just sort of tells you, are you hitting certain marks, certain criteria that you want to make sure you have, you know, kind of represented in your messaging.

22:01So who our target customer is, what's our category, what's our offering, like, how are we different? And then one of my favorite things down here at the bottom is it'll do a quick check around like industry jargon and like vague words that you're using you know i think we're all kind of allergic now to you know to a lot of the jargon stuff so it can kind of help highlight some of those things and then going to what you're just saying a second ago this positioning section is where we do some of that you know sort of proprietary analysis so we're collecting these data points looking at the brand across all these different points of business value, but then doing some additional calculations on the back end to, to help bubble that up to a simplified score.

22:48No, that's great. That's awesome. So let me ask you this. So what Clay doesn't know is I got him on the podcast to get some free consulting for Atrix AI. But this kind of transitions into your positioning expert. Everybody's talking about AI right now. I'm curious your thought. And let me break it down for how we're thinking about it and just your thoughts. So we'd build custom software solutions. That's kind of how we grew up. That's our, was our original company back in 2016 when, you know, this whole idea of generative AI wasn't even a dream yet. It took what, till the next year when the AI transformer paper came out by Google.

23:27And we were very product focused company. But since this inflection point happened, we kind of have gone all into it. And, you know, but the thing is every other company like us says the same exact thing. They're all AI power. They have AI services, even if it's all just smoke and mirrors back to our smoking ladder tool here, they're still going to say it. So the customer is kind of left thinking, okay, well, do you really do this or not? So one thing that we've kind of built is our proprietary framework methodology for building with AI. It's still evolving and, you know, we're going to actually start bringing it to life in the market very soon.

24:06But essentially, and it's just been as of late too, it's really changed in terms of how you build, right? We're used to having these huge teams with the scrum master, the BA, the QA person, tons of developers. It's just very difficult to manage. But back to that analogy of moving the person to the orchestrator, AI is the executor. You know, it's a totally different approach. You can build so much faster and just, you know, it doesn't matter what language you're using. You don't have to have the human that knows the deep syntax of this language, you almost need a different type of engineer, which we're kind of framing as an agentic engineer that has deep architectural understanding.

24:45They're in the loop of AI building. They can spot when it's kind of going off the rails. And the stuff we're starting to do, we're starting with more like greenfield startup type projects, internal use cases, but it's just night and day what we can build. So like, I'll step back there. Like, how do we build our awareness and make a name for ourselves in this extremely crowded category of, you know, ultimately you call it services at the end of the day. Yeah. Well, just to finish this out, and I think this really ties in with exactly what you're asking. I'm going to just show you really quickly how we kind of approach differentiation which i think is kind of at the heart of what you're talking about here you're you know it's like there's us and then there's a million other providers and like how do we kind of stand out just as an example here i'm gonna so the other you know screen recording tool kind of like the market leaders loom i like that yeah so this gives you a chance to look so what we're looking at now is like an overlay of the points of value with tele versus the points of value with loom and where is each brand strong like where and where are you weak what are the things that you're really emphasizing and what are the things that your competitors really emphasizing yeah and trying to find those gaps watching this basically has an overlay visually across these different attributes one line is tele one is loom and then And this gives you context for where they're strong, where they're weak.

26:19Exactly. It gives you guidance for where to dig in, I guess. Yeah. And then backing that up, there's qualitative descriptions around why, you know, they got the scores that they did and why they're strong or weak. And down here at the bottom, we actually pull out like, where are the biggest differences? So if I'm Tela, where do I currently have the biggest separation from Loom? Like, you know, if I'm trying to get a tiny piece of their market share. and like I said I actually use tele and you know we call out here like marketability configurability as being like their strengths and I would say that's absolutely true I think what they're really good at is creating marketable content whereas looms kind of more of a communication tool in mind for like internal videos and things anyway so the reason I wanted to show this was I think it ties to your point which is we are more and more getting into a space where like the devil's in the details, like it's going to come down to, is there one or two very specific points of emphasis that your brand can have that you can really lean into that separates you in some meaningful way from a competitor?

27:25Yeah. So I got the next, and you may already have this built into the tool, but next feature enhancement. So I was playing with deep research the other day because it's now on the plus plan. So us, us regular people playing 20 bucks a month got access. Yeah. But I saw somebody, I think it was Dan Sanchez on LinkedIn. He put out this prompt. I was like, oh, that's cool. And it was basically like competitive. It's like a sentiment type of thing. But what was neat is it was going out and searching Reddit, G2, Clutch, things like that. So then you're actually getting to see real customers or past customers talking about the brand.

28:04but I don't know if you already have that in there, but that could be like the next thing where it's, okay, I see how you're publicly facing, but how do people, your customers really think about you in a sense? Yeah, no, that's definitely like the next piece that's not really built into this right now. This is really kind of focused on how are you presenting yourself to the world, which to me, I guess the customer stuff could almost be like related to your delivery in a sense too like the actual product itself and how good it really is which is kind of different right yeah exactly but i think you're absolutely right i think that's kind of in many ways the other side of the equation is how do your customers interpret what you do or you know why you think they're good or bad or whatever um so yeah so right now we're kind of focused more on the how are you presenting yourself which to me is the most direct line we have to your strategy it's like in a lot of brands don't do a great job of actually articulating their strategy and that's what this is kind of meant to call out so whether you're talking to a new client trying to get up to speed on their space and their category quickly you can kind of use some of this or you're actually like working on your like you guys working on your own positioning so well here's like another i've had this idea we're starting to build it out but a big thing when we're building something custom is the discovery side of it.

29:25And a lot of times we'll do like strategic engagements where you're trying to like assess what's going on in the org. Where are the areas for opportunity? I've done these engagements and they're so painful because you've got to go find the right people to talk to. You then got to a get them to answer their email or phone call and then schedule their time that actually works for their calendar, your calendar. You got to prep for the interview. You got to do the interview. Then you got to go watch, rewatch the interview to get the insights. Do that by however many times you do, and it's exhausting, but there's tools now.