The Prove-It Economy is Here | And Most Marketers Aren't Ready

The internet economy is changing with AI and we're not paying enough attention to it. The whole internet economy has been built on attention for 25 years. And at some point very soon, someone is going to ask about you. You individually, not even necessarily your company or your product. They'll ask that too. But about you individually, should I trust you? And they won't ask you. They'll ask an AI. Because you see, the internet economy is changing from I need to get attention. I need to shout. I need to get people to pay attention to me to sell my product. to sell myself as someone who can work, right? To have a resume, to get out there on LinkedIn, whatever. All of that was an attention economy. Now, we are moving to an interpretation economy where the whole web is filtered through what an AI thinks about you. And so, when I say people will ask an AI whether they should trust you, I do mean that very literally for you, the individual, as much as I mean it for marketers, right? And if you're a marketer, this is not just for you as a person. This is for your whole career, right? like your whole career is shifting from how do you get attention from people to how do you get people to understand the question they need to ask an AI in order to transact and I want to talk about the implications of that both for us as individuals and for the marketing role because I see them as essentially twin problems they're the same kind of problem in both cases whether we're seeking new jobs or whether we are seeking to market products we are dealing with a massive transition from the attention economy to the interpretation economy. The attention economy we're all familiar with. It's been on the internet for 25 years. It's how Google built the business it has today. Ads, right? You sell ads to people with eyeballs who click on things. That is how it works. And then you get their attention. Then you go into the rest of the funnel. You go into the demand side all the way down to the bottom. People convert. Whether they're buying Nike shoes or whether they're buying a giant five or six figure SAS contract, it all works the same way. But not now. Not anymore. Let me give you a specific example to make this real. I bought a sound system recently. The people who are in charge of marketing sound systems, I don't think they had anything to do with what sound system I picked. And that should terrify them because I did my entire shopping experience by chatting with my AI. I chatted with both Claude and Chad GPT. And I was able to get a much much better experience from my point of view as a consumer because I did that. I was able to give them the room dimensions that I was working with. I was able to give a budget. I was able to give, you know, this is how I like my sound to be, whether it's warm or whether it's cool. All of that nerdy stuff. And I got exactly the set of recommendations that I thought was appropriate. Now, to be honest with you, as a consumer, because this is the interpretation economy, all of this was interpreted from AI. I don't know if I got the truly best option. I just know I got what seemed like the best option because AI told me. And that difference is worth trillions of dollars to marketers and to us as individuals because we need to be in that consideration set. If if if you're being asked, hey, how do we position ourselves for an AI future? The first question you need to ask yourself is how do I position either the product I'm responsible for, the company I'm responsible for, founding, marketing, building, or myself as a candidate for a future where AI is going to be the first port of call, the first place we stop to ask. I want to suggest to you that thinking about the interpretation economy and how we position ourselves has real practical implications for the money we spend and the work we do when we put ourselves out there on the internet. And I'm going to give you an example. Again, we're going to keep following the thread, individuals, and we're going to follow marketers. In both cases, I see the same preoccupation with what I would call back office motions as a way to AI automate instead of focusing on where the leverage lies. Back office automations like it's like marketers who say, I came in and now I'm going to automate a bunch of my work with chat GPT or I'm going to automate a bunch of it with Claude. I use co-work now to be I got to be honest with you, those are table stakes in 2026. You do have to do them. You're not going to get a lot of credit for it. But if you want credit, you have to think about how you position your entire product and offering for an interpretation economy. That's where marketing is starting to go. And I'm going to give you some specific pointers there. If you're an individual, it's sort of similar. You're not going to get credit for I customize my resume and my cover letter and I I customize where I search to exactly the targeted job search results that match to my keywords. That's table stakes. Now, everybody does that and that's all AI back office automation. You have to think about the interpretation economy. How are you being filtered through AI? Let's start with lessons learned for marketers and get into lessons learned for individuals as we think about the implications of this interpretation economy together. The first lesson learned for marketers is this. You need to have and care about a truth layer for your company that enables you to reliably distribute high quality, highfidelity data about your product to agents on the internet. And by agents, I mean chat GPT is literally sending a search agent when you ask a query. So I don't mean just agents that nerds use. I mean everybody has agents whether they know they have them or not. now and you need to be ready for that. And part of how you get ready for that is you get ready by making sure you understand the map between what customers ask for and talk about and what they mean and how you can have clear, accurate, detailed data about your products that enables an AI to get an opinionated signal on what you offer. So, let's break that down. If you're not opinionated, then you are going to be flattened into the internet average for your category. Let's say you sell shoes. If you're not opinionated about how great your shoes are and why they're great, and if you can't prove it, and you can't structure it out so that any AI agent can read and understand what makes them great in a way that feels provable and reliable, so the agent trusts it, then you're going to get flattened into just another shoe seller. You need to be able to say this is our special spring system for our heels and this helps the runners to unlock more energy on every step and this and that and and that used to be positioned as mostly marketing language for humans and you could get away with these emotional claims. Agents don't work that way. That's why I call this a truth layer. Agents need you to prove it. Agents need you to prove it. And so you have to come back and say this is the actual data behind this. This is how this works. This is the way we save energy on every step. This is the material that this shoe is made of that enables us to deliver on this. And I say that as an example because I think it helps us to understand the difference between the emotional marketing that we're used to doing to people and the kind of factual information and hooks that agents need to get what they want done. That's super super important because if you don't have that agents won't be able to map customer intent onto that purchase and therefore you won't be in the consideration set and now you're in trouble. So for example if as a runner I'm saying hey I want to make sure that I am reducing the the effect of running on my knees and so please think about how I reduce the impact energy of a step. Well, that could map on if you have a good truth layer, if you have good data availability, that could map on to the running shoe I just described, but only if that truth layer is there and it's translated into web pages where the agent can effectively extract the DOM out of the web page, extract a data set out of the web page, it's highly readable, maybe it's a JSON schema, whatever you want, and is able to ingest it and say, "Okay, I understand what this product is and what it does." But I don't see marketers doing that. Almost all marketers say, "That's too technical. That's not for me." But that's where the leverage for marketing lies today. If you want marketing on the internet to work, you have to care about that level of technical detail because the agents are doing the shopping. And that that is going to be more and more the case. Whether people know they're using agents or not, whether they tell you in surveys they're using agents or not, agents are doing the shopping. And for individuals, this also works the same way. We talked about a prove it economy when I talked about the talent board project that I launched. It's still live. It's still up. can go and uh grab it if you're a part of the Substack community. A and the whole idea is that we have to be able to prove that you have AI skills to be a part of the AI talent market. In other words, vetting and proof matters a lot to get reliable roles in the AI economy. And it's not just about dressing up your LinkedIn. It's about being able to say, "Yes, I actually do have the ability to put together an MLOps pipeline, or I do have the ability to stand up an entire agent development pipeline, or I do have the ability to communicate intent so clearly that if a pipeline is there, I can build stuff in production code as a PM or as a designer." And yes, that is actually happening. I've seen that happen. The last time I saw that happen was yesterday. It really does happen. And so if if you want to show you have those skills, you need a place to demonstrate it. So that that that's what the talent board project is about. It's not the only way to solve that problem, but I wanted to make a tangible step towards solving it for individuals because I think that it's the same kind of problem. You need a place where you can demonstrate, prove, and show that you actually have the skills that you say you have. And so if the AI is asked for its opinion, you have the data there to actually explain to an agent what you do in detail. You have your own personal truth layer. And we talk about the importance of underlying data a lot on this channel because if you don't have the data that makes an agent work well, regardless of the task, it's it's just not going to work for you, right? But we talk about it mostly internally. We talk about it as you have to have your data available internally to do the work you want to do with agents. That's all true, but it turns out you have to have it on the broader internet too if you want to interpret and gather information correctly. And so if you're marketing your product on the broader internet, you need that data. If you're marketing yourself, you need a truth layer. You need that data. That's the idea behind the talent board. There are other people doing other versions that are also cool. Uh but I want to call out that if you don't think that way, you are going to be flatfooted. you're going to be caught off balance when the market starts to shift more and more and more toward an AI interpreted internet. And it is happening so fast. We don't see it all the time because you will still see people right now go on the web to transact and purchase. But we feel it. We feel it when there's a thousand job applications and nothing happens. We feel it when you just have to work harder and harder and harder as a marketer to get the same level of attention because it just feels like there's less juice in the orange to squeeze. This is not an economy that is growing. When when you talk about the attention economy on the internet, the internet economy as a whole is doing very well. But the attention part of it, that is not where the future lies. The future lies in the interpretation economy, in the ability of AI to interpret what is on the net. And that means you have to have the data available to do it. Now, if you ask me, well, what's the alternative or what are the other options? I will tell you there's two basic ways to purchase. The first way is to have what I described. Have ask the AI to interpret for you and then the AI transacts. Whether you're buying shoes or you're looking for candidates, ask the AI to interpret. And yes, there are literally hiring managers trading prompts back and forth looking for top candidates. So if you don't think this happens for jobs, it totally happens for jobs. Same thing. This is why I'm telling you it's the same thing. Now the other option is effectively creating brand loyalty. How do you create brand loyalty so strong in a human that they ask by name for the product? But my previous sound system screwed this up, right? They could have seated brand loyalty so well that I only asked for them by name. They didn't. I didn't stick with them for the next purchase. It didn't seem like it worked. That's what you need to do to shape human intent. that is really really powerful for you in the age of AI because I I'm a big believer I know I talk about the internet all the time but I'm a big believer in the offline economy there is going to be a lot of in real life events whether they're jobseeking events or whether they're marketing events that are essentially about seating prompts you're basically trying to be remembered so that when you are asking an AI later you can say oh yeah you know I met this guy Nate uh let me just sort of ask about him on the internet and then and then you've seated the prompt or hey I remember going to this IRL event for this particular coffee company. Let me ask for that coffee company by name and by brand and then that turns into a purchase where the agent is bound or constrained. Even though you're asking the AI, it's not going to give you 15 other coffees. It's just going to pick the one because you asked for it by name. And so when we talk about the future, that's one of the things I want to call out is that we are talking about how you create value offline that then translates online. Now, if you're feeling like this is all a lot, right, I have put together a toolkit on Substack that will help with all this. It will help with developing the the truth layer that you need. Um, it will help with helping you understand your current position, how you audit and understand your current position. It's going to help you whether you're a candidate actually or whether you're a marketer. And I'm going to have separate packs for that because otherwise it just gets it gets messy because I want you to have a place to go deeper on this. I want you to have a place where you can think about the interpretation economy and have a toolkit to take action on it so you can at least get started in that direction. Okay, let's talk a little bit more about offline marketing and what that means. Now, the obvious risk in this argument is that it can make brand sound much less important than it was before. If agents are going to read everything, then you know the emotional work, the connecting with humans doesn't matter as much as it used to. I disagree. I think the opposite is true. Human memory becomes more precious as more of the transaction is mediated. If you have less of it, it's worth more. The agent may do the comparison and retrieve the options and summarize the trade-offs, but the human still applies the preference. In other words, if the the sound system company that I have been buying from had decided to build an emotional connection with me instead of not doing so, I would not have gone to my agent and said, "Hey, find all the sound system options I have and let me just get the one that's best." I wouldn't have even gone there. I would have said, "I trust this sound system company. They're the ones I'm going with, and I'm not going to go with anybody else." And then it would have been very simple, very clear they would have earned the purchase. But they did not do that. And so I went and did a wide-ranging search and I came back with what was agentleible and what worked for me. So human memory becomes more precious as more of the transaction is mediated. And this is why I think the two sides have to reinforce each other. There's humanfacing work that requires memory and preference and trust in language. There's agent-f facing work that creates clarity and structure and evidence and retrievability. If your human brand says one thing and your agent readable reality says another, the company gets weaker in both directions. If your story is memorable, but your product truth layer is incoherent, the agent is going to flatten you out. If your PR materials are well structured, but your brand creates no memory, you may show up in comparisons, but the human is not going to choose you. The best marketers will refuse to choose between the two internets. They'll choose both. And this is why AI washing is so dangerous for us as individuals when we're trying to get out into the market and also for brands. If you're working at a company that feels behind, the pressure to sound AI native is intense. If you are trying to get up to speed on being AI native and you see that AI native roles are paying more the pressure to AI wash yourself is intense. I get it. Please be clear about what you can do about what your product can do so that the interpretation layer allows you to stand out because one of the risks when you AI wash is that you AI wash you you deceive people who don't know better and they buy from you and are disappointed or they hire you and are disappointed and then it's just a bad situation all around. The better alternative is to lean into what you know, follow your curiosity, drive and build on your expertise and deliver that. And you know what that exact chain that is what you need to do from a product perspective to make your product more interpretable on the internet. And it's also the same thing you need to do as an individual if you want to truly make a transition to an AI native positioning on the net that agents and humans can both understand. We are in a two internet economy and it affects products and people the same way. Underline that three times. The job is to find the honest wedge that gives you or your company or your product a future. The wedge on the internet that is distinctive for you. So what does this mean? If you're choosing a role or building a team or trying to understand whether your current company gets it, I would look very carefully at whether marketing is allowed to touch the surfaces that matter. And everybody should ask this. Does marketing touch the website? Does it touch pricing clarity? Does it touch claims or launch process or sales collateral or support facing explanations? Or is marketing only allowed to decorate decisions that are made elsewhere? That distinction matters a whole lot. Now, if marketing is just a content factory, it's a very weak role. If the company says it wants AI marketing but only means make more assets with fewer people, it's a weak strategy. If the company wants to AI wash but not clarify, you're in trouble. What you need is to have a role that gives you the ability to make the company more legible to agents, more memorable to humans and more operationally capable internally so that you are close to product and sales and customer reality and technical teams and everyone at the company should care that marketing can do that because we all benefit when that's the case. And yes, if you're a human, you have to be that marketer for yourself and you have to do those things for yourself. I'm not saying you have to be an engineer to do all of this. That's part of why I built built these toolkits. It's one of my convictions at core is that the future belongs to people who are willing to be a little bit technical but who are not engineers. And that is true here as much as anywhere else. You can make yourself or your product agent legible without being overly technical about it. So marketing is becoming more technical, but it is not becoming so technical that we're going to have a bunch of engineers doing it. That's not what I mean. It's just going to require you to understand a little bit more about AI than you do today. And that's why I've spent a lot of time in this video explaining the different internets we're moving to, why agents work the way they do, and how they interpret information, and why the agentic information flow from that truth layer you're building has to be different than the attention layer you're driving for humans that everybody went to marketing school for, and that frankly everybody was told to do when they were job candidates. Because job candidate motions are like based on 25 years of marketing history. Market yourself. Go out there. Put yourself out there. Get on LinkedIn. Make those posts. Same motion for the attention economy, not the interpretation economy. I want to close by asking you to think about where you're spending your energy. Whether you're a candidate or a marketer, whether you're a founder trying to position your company or a product person trying to build something that gets out there, do you position your energy where you are just trying to get humans to pay attention to you? Or are you positioning your energy in two deliberate buckets of bets? Bet one is how do I make myself deeply memorable to people? And think about that. That is going to be offline as much as online. economy especially as it relates to driving online purchasing online hiring etc etc networking events in real life are a big deal postco but also at the same time are you positioning yourself for the agentic interpretation economy and what I mean by that is are you providing a level of differentiated detail driven by a clean truth layer that enables you to actually compellingly convincingly explain to an AI agent that is doing some random search. What you are doing and and why that matters, what your product does, what your company does, what you the candidate, do is it there at a level of detail where an agent can read it and understand it and say, "Aha, this is going to survive compression because wow, this is this is spicy. This is original. This has opinions." And that's the thing that are we are missing, right? We are missing the ability to have opinions that agents can translate forward into documents, summaries, tradeoff conversations that humans will see. Do not approach the agent economy and say our answer is to be the candidate for everyone, the product for everyone, the company for everyone and not have opinions. You got to have opinions. The value of having opinions is even more important now because agents need to see how you are clearly differentiated. If they can't, you're not going anywhere. You're not finding the job. You're you're not finding a a clear distribution channel as people look to purchase your product agentically. The interpretation layer will average you out. Don't be afraid to have opinions. Don't be afraid to be specific because that is how you stand out in a world where everyone is using their new AI tools in marketing or elsewhere, maybe their new AI resume tool to sound like AI slot and people will read through that and agents will average that out and that's why it's becoming more noise and less signal. I hope this has been helpful. You can always grab more details on the Substack. And if you're interested in going beyond marketing, you're like, Nate, what about the rest of the job families? Don't worry, we are going to be doing more of these videos getting into product, getting into CEOs, getting into seuite, getting into founding, getting into engineering, getting into customer success, right? There's lots of places we need to go, but I hope you picked up the common thread here. Do not think that when you walk into a role with a marketing lens on that the whole point is back office automation. That is like walking in as a candidate and saying, "My whole job is to have AI write my resumes." It's table stakes. It's table stakes. Think more strategically. Think about where the internet is going and how you need to orient your company around that.