The New Shelf: Why AI Visibility Has Replaced Traditional Shelf Space for CPG Brands

Gone are the days where retail customer primarily purchase items going into a brick and mortar store, the proverbial shelf space is now directly tied to AI vistibility.

Let's paint the picture in 2026: A shopper picks up her phone and says "find me the best organic suplelment to help with better sleep" and never names a brand. Her AI shopping assistant, whether it is Chat GPT, Claude, Google, or another service entirely, searches the internet, find 5-10 supplements, searches their ingredients, and any third-party endorsements and suggests it. The agent, with no understanding of a "brand" just picks one, with no consideration of a logo, colors, or origin story.

For thirty years, winning in consumer goods meant winning the shelf. Eye-level facings, the endcap, and the right planogram. The fight was for space a shopper could see.

Bu that shelf is moving and the brands that don't earn a place on the new one go invisible. Not gradually, but all at once, the moment an AI answers instead of a search bar.

What AI visibility actually means

AI visibility is whether assistants like ChatGPT, Gemini, Claude, and Perplexity name your brand when a customer asks them to solve a problem. It's the AI-era version of being on the shelf, except the assistant reads the entire shelf first, then hands the shopper one or two options instead of fifty.

A grocery aisle shows dozens of facings at once. An AI answer shows almost none. Analyses of AI-generated answers in 2026 find they name only two or three brands on average before the customer stops reading.

That's the whole shift in one number. Shelf space didn't just go digital, it shrank by an order of magnitude. And unlike the endcap, you can't buy the second facing.

GEO vs SEO: what actually changed

Search engine optimization (SEO) earns a ranked spot in a list of links the customer still has to choose from. Generative engine optimization (GEO) — also called LLM SEO or AI search optimization — earns a place inside the answer the AI gives, where the customer often never sees the alternatives at all.

With traditional SEO on Google, you could buy your way to the top of the page. Whether through a thorough, long-term SEO campaign or paid ads, you could almost guarantee a top spot, or least on the first page.

AI recommendations don't sell that slot. You can't pay your way into the answer ChatGPT gives (maybe one day though). The assistant recommends what it judges to be the clearest, most credible, best-supported solution to the problem in front of it.

The win moved from being found to being chosen. That's a harder game and a more honest one. Google's own 2026 guidance frames generative engine optimization as a layer on top of solid SEO, not a replacement, so the two reward the same foundation: clear, trustworthy, well-structured claims.

Why AI visibility hits multi-market CPG hardest

The qualities that make a claim discoverable by an AI are the same ones that make it defensible to a regulator: specific, substantiated, sourced, consistent.

Vague puffery fails on both fronts. Phrases like, "The best," or"Clinically proven," with nothing behind it the AI agent will skip, or at least not consider. It's the same thing with a regulator: Same weakness, two different judges.

A multi-market brand feels this at scale. One claim, dozens of markets, each with its own rules and its own language. Get it slightly wrong and it doesn't just risk a fine, it becomes the thing the AI quietly stops repeating.

Consider what shipping at scale really means. When a major toy line released Wicked-branded packaging a couple of years ago, a single mistyped link on the box sent buyers to an adult site instead of the film. A human can make that mistake in one second. At a million units, it's a national story, and exactly the kind of trust signal AI systems are built to weigh against you.

How brands get recommended by AI

Three things move the needle on AI visibility:

1. Be specific and substantiated. AI recommends claims it can verify. A claim like "reduces odor for 48 hours, tested across 1,200 users" beats "long-lasting freshness" every time. The same specificity that survives legal review is the specificity that survives the model.
2. Be consistent everywhere. AI assembles a picture of your brand from every mention across the web — your site, retailer pages, reviews, press. Contradict yourself market to market and you blur the very thing the AI is trying to recognize.
3. Get it right before it ships. Models reward accuracy and trust, and both are decided at creation. A claim that's wrong, off-brand, or non-compliant doesn't just bounce back for rework — it teaches the web, and the models reading it, the wrong thing about you.

Where review comes in

Here's the part most teams miss. AI visibility looks like a marketing-channel problem, so it gets handed to an agency to fix after launch. But the raw material — clear, accurate, substantiated, on-brand, compliant claims — is set the moment the asset is created.

That's review. The step everyone treats as the bottleneck is actually the leverage point.

At Puntt, that's the layer we build. Our agents review every asset — packaging, video, social, artwork — against your own brand rules and each market's regulatory requirements, pin each issue to the exact spot, and cite the rule behind it, before anything ships.

We don't replace your regulatory team or your agency, and we're not a GEO dashboard. We make sure what you put into the world is accurate, substantiated, and on-brand the first time — which happens to be the exact foundation AI discovery is built on. Getting the claim right the first time, along with the proper messaging, taglines, and branding creates a consistency across your marketing material that, not only will regulators not penalize, but help with AI visibility.

It's been proven to dramatically reduce the time spent on tedious tasks, like triage, and free up time for marketing teams to be more strategic and thoughtful. It's no longer just about speed, but control. When marketing executives can show up to the GTM all-hands confident in their product claims and sourcing, it build confidence with leadership and allows for creative to push the boundaries on campaign that can actually move the needle.

How to improve your AI visibility: start Monday

1. Audit yourself inside the answer. Open ChatGPT, Perplexity, and Gemini. Ask the way a customer would — "best [your category] for [specific need]" — and see whether you appear. Omnibound data indicates that AI-driven search changes are causing up to 83% of queries to end without a click, making non-cited brands invisible to a significant portion of traffic. Furthermore, research highlights that 94% of B2B buyers now utilize AI assistants, leaving brands outside these answers with minimal visibility.
2. Pick one fight. Don't spread thin across six channels. New brands beat incumbents in AI search precisely because the incumbents are twenty years deep in old SEO habits. Focus is the advantage.
3. Move the check upstream. Put accuracy, substantiation, and compliance review at the point of creation — not after legal sends it back. What's right at creation is what's citable at launch.

The new shelf

For decades, the fight was for a shelf the shopper could see. Now the fight is over a shelf they never look at because, similar to an Instacart shopper, the AI is standing in front of it, reading every label, and handing them one.

Have you checked whether it hands them you?

Frequently asked questions

What is AI visibility?

‍AI visibility is whether AI assistants — ChatGPT, Gemini, Claude, Perplexity — name your brand when a customer asks them to solve a problem. It's the AI-era equivalent of shelf space: the assistant reads every option and recommends only a few, so being named is the difference between being considered and being invisible.

Is GEO the same as SEO?

‍No, though they overlap heavily. SEO earns a ranked link in a list the customer still chooses from. GEO (generative engine optimization) earns a place inside the AI's answer itself. Google's 2026 guidance treats GEO as an additional layer on top of solid SEO, not a replacement — the same fundamentals of clear, trustworthy, well-structured content apply.

How do consumer brands get recommended by AI?

‍AI assistants favor brands whose claims are specific, substantiated, consistent across the web, and credible. Vague or unsupported claims get skipped. Accurate, sourced, on-brand claims — the same ones that pass regulatory review — are the ones AI repeats.

How does marketing review affect AI visibility?

‍Review is where claims are made accurate, substantiated, and compliant, before they ship. Those are the exact qualities AI systems reward when choosing which brand to name. Catching issues at creation, rather than after launch, builds the clean, trustworthy footprint AI visibility depends on.

More on AI Visibility: Puntt AI CEO Joins CPG Insiders Podcast

Below is the full episode where Puntt AI CEO Ronnie Kwesi Coleman discusses the origin story of Puntt, the future of AI for marketing teams, and how he's helping marketing teams position their products and increase AI visibility.

Mark Young: Welcome everyone to another edition of CPG Insiders. I'm your host, Mark Young, with co-host Justin Girard.

Justin, we've been talking on this show for some time about AI. Part of what I keep telling people is that the world we have today isn't going to look like this in 24 months. AI is doing some miraculous things, and one of them is democratizing and demonetizing access. As my friend Peter Diamandis would say, it's leveling the playing field — especially for entrepreneurs.

It used to be that to have the best lawyers, the best accountants, the best graphic artists, you needed a lot of money. That playing field is being leveled today. Even having a website that looks as good as Procter & Gamble's is doable now. Which means the opportunity for new brands, upstart brands, and people who've come up with great products is actually better in some regards.

When we look at product innovation in CPG — not to pick on our friends in Cincinnati, but for P&G, product innovation is now a new spearmint scent. The real innovation comes from the inventors: the people in garages, tinkering in their kitchens, working in their basements. And most of these people, you see it all the time, Justin — how many new clients come through the door because they invented something to solve a problem for their family?

Justin Girard: Almost all of them.

Mark Young: "My dad had this." "My wife needed this." "I couldn't find this." So they went out and solved a problem other people share.

I'll go back to Peter Diamandis for a second. One of his quotes is: if you want to make a billion dollars, go help a billion people. The world's biggest problems are the world's biggest opportunities.

Justin Girard: Absolutely.

Mark Young: So if you're an entrepreneur listening today, go solve a problem — just make sure it's a problem with a reasonable amount of mass appeal. Don't solve a problem that only you have.

Now, we have a great guest today: Ronnie Coleman — and I already teased him that he's not Ronnie Coleman the bodybuilder. Ronnie is an inventor and internet entrepreneur known for building future-of-work companies. He built a global creative marketplace that enabled over 100,000 jobs for Africans and became the subject of a Harvard Business School case study. He scaled content operations for Fortune 500 marketing teams using distributed talent, and co-founded a hospitality workforce platform that exited for $40 million. Now he's building AI systems to review the world's work, so people can focus on progress, not process.

This connects to something I say about AI. In the old work model, a supervisor had about six employees. The supervisor did the first 10% of the work — what are the goals, why are we doing this — then handed it off. The employee did 80%. The work came back, and the boss polished the last 10%. Ten, eighty, ten.

AI changes that. Now you do the first 10%. You need to be the boss — know the questions, know the problem to be solved. The AI does the 80% in the middle. And you do the last 10%.

This is part of why the single largest group of unemployed people in America right now are recent college graduates. The social contract broke down: I did what I was supposed to do, I got the degree, and now where's the job? We don't need people who can do the work anymore. We need people who can think. And as somebody with a PhD, I'll tell you — you don't learn how to think in school, you learn how to memorize.

So, Ronnie, tell us about yourself and about this company, because what you're doing is amazing.

Ronnie Coleman: Thanks for the kind intro. You said a few interesting things in there.

So, on thinking versus school — I came to America as an immigrant. My dad's from Ghana, my mom's Ukrainian, so I was born in Ukraine, but we lived in Ghana. My dream was to build a company. And the way to do anything successful, as my dad said, was: go to school, go to college.

Mark Young: That's the typical path.

Ronnie Coleman: That's the normal path. I came to America with scholarships and offers from several schools. But instead of going to college, I got approached by a company to join them as part of a startup and go build. The founder told me, "Look, you can always go back to college."

Mark Young: By the way, Ronnie, I only went and got my PhD a few years ago.

Ronnie Coleman: My dad did the same thing, actually, so I commend that. But it was just: come do it, come work with us, we're building cool stuff. So I did. I went there, fell in love with it, and after two years left to start my own company with some folks. That ended up exiting for $40 million — all before I'd have finished a college degree. So sometimes, just diving in and doing the work makes all the difference.

Mark Young: I agree with that. The majority of my wealth was built before the education. The real education was jumping in and being an entrepreneur from a young age. I almost think I went and got the PhD just so I'd have credibility when I tell people it doesn't matter. Now that doesn't mean school is a bad idea — if you want to be a doctor, go to school. If you want to be a lawyer, you probably need to go. But it's not the traditional path for many careers anymore.

So, tell us about this business. It's AI package review — is that right?

Ronnie Coleman: Let me start with the problem. One of my advisors, somebody I look up to, is Uri Levine, who founded Waze. He has a great book called Fall in Love with the Problem. So I'll tell you the problem I fell in love with, and how it became the company we have today.

At my last company, Meaningful Gigs, we ran a marketplace connecting creative talent with marketing teams at Fortune 500 companies. One of the problems those marketing leaders had was: how do we review all of this content we're creating? We're a large global company, we have regulations, we have quality and brand standards, and we need to make sure everything that goes out to consumers is great and legally sound.

So you've got someone paid $150,000 or $200,000 a year sitting there line by line, checking copy, making sure things are aligned. And everybody hates it.

Mark Young: And it's still a human who can make a mistake.

Ronnie Coleman: Exactly. Not only do they hate it, it's error-prone. So companies would hire teams of people from our marketplace. Imagine any package you see — there might be 40 people who had to look at it: artwork, design, brand, regulatory. And what stood out to me was that everybody hated it.

You said it: if you want to make a billion dollars, help a billion people. This was the problem. If everyone hates this, and we could automate this manual, tedious process, we'd have a customer for life.

So, around the wave of AI, we started telling people we could automate this review process — brand, quality, artwork, regulatory. And, by the way, you don't have to get rid of the people. That's a big thing we believe in. Your best people can spend their time on higher-order work. They can save real time doing something more meaningful. Let the humans do what humans are uniquely good at — the thinking.

Justin Girard: And to your point, it elevates the person's fulfillment, because now they're doing work that energizes them. In turn, their productivity skyrockets, because they're being more creative and imaginative — which is where true scalability of productivity comes from — instead of the drudgery you're describing.

Ronnie Coleman: Yeah, and it's critical now more than ever. Everything is shifting. AI is shifting how consumers buy and how teams operate. But the thing that won't change over the next 5, 10, 100 years is that you're going to need great people. That's the thing a lot of people miss. There are recent grads without jobs right now, but humanity adapts quickly. Those grads will get new types of jobs that never existed before.

Someone said something that resonated with me: you can outsource the work to AI, but you can't outsource the understanding. The understanding of what needs to get done — that's still a fundamentally human trait. So at Puntt AI, that's what we do: we build tools that automate the tedious work, like reviews, so humans can apply their understanding. They can solve big problems and help their companies win long term.

Mark Young: And think of the time saved. So when we go to Puntt AI and put our products in, is it reviewing for things like FDA and FTC compliance? Are all those features built in?

Ronnie Coleman: Absolutely. The way we talk to customers about it: if you're a marketing leader, you have a set of rules you need to follow. If you're a regulated company and the FDA is one of your regulatory bodies, you need to make sure your work follows those rules. You have a regulatory team, but for the marketing leader there's a lot of anxiety. If I send this work in, it might take two or three weeks. If we put the wrong thing on the label, it could get sent back a month later — and now I'm behind.

Mark Young: It takes the lawyers three weeks to tell me I've got two bad words. Now I have to change them, send it back, and they keep it for three more weeks.

Ronnie Coleman: Exactly. So what we're giving marketing leaders is control and visibility into how the regulators and the lawyers think. Again, it's not replacing them — it's making it right the first time. The marketing and brand teams get the right information before they send it to legal. And legal says, "Oh, the marketing team gets us now. We don't have to send this back." Things go to market much smoother.

Mark Young: And it'll do things like making sure the font is the correct size according to regulations? What about multiple languages?

Ronnie Coleman: This is one of the places AI shines best. Let me tell you a story. There's a huge multinational brand we work with. They have brand managers — young, often just out of college — who have two fears. The first is regulatory: we'll do all this work, create all these TikToks, do all these cool things for the brand, send it to regulatory, and a month later it comes back, and I look bad. That's one fear we mitigate.

The second is language. I'm in Poland and I have to translate this for six markets — Hungary, Latvia, somewhere I don't speak the language — and I have to make sure it aligns with our brand tone and regulatory requirements. That's genuinely hard to do. At this point, Puntt can review and translate across 147 languages, just like the best translator in the world would.

Mark Young: There are some famous examples. I believe it was Pepsi in Taiwan — the slogan "Pepsi brings you back to life" got translated as something close to "Pepsi brings your ancestors back from the dead."

Justin Girard: Wow.

Ronnie Coleman: What a claim.

Mark Young: And another one. Do you remember a product for upset stomachs — a powder you put in water? They decided not to translate any language at all, and did everything in pictures. It started with a picture of a person bent over with an upset stomach, then tearing the packet, putting it in water, drinking it, and standing up happy. Seems obvious. Then they shipped it to a market where people read right to left. So now the sequence reads: you feel good, you take this, and then you're sick. Would your system catch that?

Ronnie Coleman: It would catch the nuance for the brand. This is the amazing thing AI does. If my brand is meant to convey health, wellness, and value — and that's the brand tone in place — our AI is trained for that. It's purpose-built for brand and regulatory adherence. So when it sees something that's going to read differently in another market, it flags it for the brand manager before it goes out.

There's another interesting one, more recent — about a year ago. Did you hear about the Wicked-branded toys? There was a lot of packaging, and on it there was a link that was supposed to go to the movie's website. It linked to an adult website instead.

Mark Young: Oh no.

Ronnie Coleman: These were toys for kids. Somebody likely left off a letter or got the URL slightly wrong. That's an error a human could easily make. They printed it, and it became a national problem. That's the level of detail our AI would catch.

Mark Young: Can smaller brands use this?

Ronnie Coleman: They can. What I recommend for smaller brands is two things. First, if you're regulated, we still recommend everyone have some kind of legal review. A lot of people are using ChatGPT or Claude to review a document before they send it — but Puntt is purpose-built for brand, regulatory, and global review. If you're going to pay someone $10,000, this might save you a big chunk of it, because it can do the main reviews and tell you what to send over.

Second, if you have a high standard for your brand and you don't want an agency sending you rework ten times over, Puntt reviews the work with the agency or contractor and makes sure they get it right before it reaches you as the owner or executive.

Mark Young: Justin, it sounds like we need an account at Puntt AI to run all our clients through.

Justin Girard: It does.

Mark Young: What about packaging material compliance?

Ronnie Coleman: When a customer asks what we review, we tell them: any visual asset. It could be a TikTok video — and this is what differentiates us from a Claude or ChatGPT; they don't work with video. You can put a thousand videos in, and it'll review them globally. Then there are packaging materials, which are very specific — bleed lines, print lines — and have to be reviewed against labels or technical drawings. All of that is a visual asset Puntt can review and keep on brand.

Mark Young: Could I put a television script or a radio commercial through this to check if it's compliant?

Ronnie Coleman: You could, but honestly you don't need us for that — the foundation models are good at plain text. The magic for us is the complicated visual work: artwork, brand videos, packaging. If you're a big brand that wants purpose-built visual review workflows for regulated markets, and you want to scale without spending a fortune, that's where we shine.

Justin, you were going to talk about state-by-state packaging requirements.

Justin Girard: Sure. Ronnie, have you and your team been researching the new EPR — extended producer responsibility — requirements popping up at the state level for recycled materials? That's something we're seeing tied closely to packaging materials. Brands are going to be responsible for paying for the lifecycle of their packaging.

Mark Young: Basically, you sell a million dollars of product and you're responsible for that packaging for the rest of its life.

Justin Girard: Right. So I'm curious — you were describing the visual analysis, the bleed lines. Is material analysis something that could be done as well?

Ronnie Coleman: A few things. One is that we look out for customers on regulations that exist today and regulations that are coming. Sometimes you've run thousands or hundreds of thousands of assets through Puntt with a label that's no longer going to be compliant in one specific market. Either someone reports you and you've got a legal problem, or you have to proactively review everything.

Anything that goes through Puntt gets compared against new regulations too. So even something you reviewed five years ago, two years ago, Puntt can flag: this is coming, you're at risk here and here. They still have their government affairs people, their lobbyists, their internal teams working on the regulation itself — that's strategic work that stays human. Our job is the manual, tedious part: reviewing past, existing, or future regulations against the artwork and everything you have.

Justin Girard: I want to make sure it's clear to listeners — because a lot of them are challenger brands and entrepreneurs — that you're not saying Puntt AI replaces a legal review. Don't just put your work into Puntt and ship it. You're still at risk there. But what it can do is provide a first step of analysis that removes most of your risk before a legal team reviews it.

Ronnie Coleman: Right. I tell customers it's similar to a radiologist using AI for X-rays. Radiologists review thousands of X-rays to find things. AI came along, and now there are arguably more radiologists, because people are excited to do the job — they don't have to sit and review only X-rays. The AI tells them which ones are most important to review. You still need the professional. AI just speeds up the process and makes it cheaper.

Mark Young: When we look at the legal profession, a couple of things change with AI. One: the one- or two-person law firm can now do things that used to take a hundred- or thousand-person firm, because they have an army of experts they didn't have before. Two — and this matters for early-stage clients — legal fees are expensive. This gives small companies access to better legal service. They can do the review, and then afford a high-level lawyer for a one-time pass instead of paying for endless back-and-forth.

And having a real lawyer look at it matters if there's ever a mistake. Because what the FDA and FTC look at isn't just whether you made a mistake — it's the intent behind it. If you can say, "First I ran it through an AI system, then I sent it to a law firm that confirmed it," now the agency is more likely to say, "Fix this in 90 days." You took reasonable steps. You're not looking at a big fine or a lawsuit. They don't believe you set out to mislead the public.

The other thing that happens — outside our industry — is that good, hardworking, working-class people who need legal services and can't afford them will get access. AI is going to increase the amount of legal work being done, because more people can participate. So is AI going to get rid of lawyers? No. But if you're a lawyer worried about your job, you won't lose it to AI — you'll lose it to a lawyer who understands how to use AI.

Justin Girard: Absolutely. So, Ronnie, for a smaller brand really getting into this, what are the inputs they need to have to make the most of the platform you've built?

Ronnie Coleman: Let me reframe it. For a small brand, what's important is building the company AI-first — being an AI-native company, because that's the only real advantage they'll have. There's a shift in the market. We talked about P&G and the big conglomerates — some of them are moving incredibly fast. I've built multiple companies, and this is the fastest I've seen enterprises move with new technology. So if you're a small brand and you're not moving ten or a hundred times faster than an enterprise, you're messing up, because there are a lot of great tools.

So first, ask: what does AI-native mean for me? If I'm in a regulated market and I have a brand I care about, what are the purpose-built tools that help me move faster without sitting down to write a hundred prompts? As a busy leader, what can I get out of the box that helps me move faster and reduce my risk? That's where something like Puntt comes in.

The scope of it: they can ask what claims they can make in a given market, and who else is making those claims. If you want to sell into a specific market and you're doing positioning work, we have global data on claims. We work with some of the largest companies in the world, and with smaller brands, and we know what claims are being made in what regions. So before you write a claim and go through all the legal work only to find out it won't sell there, you can do claims research on our product. Then you put in your AI-generated content — thousands of pieces — and our system reviews it and tells you what's best for your brand, what language to use to stay compliant, and then helps you manage the project end to end.

Part of the magic of these agents is that they're proactive. We build them not just to tell you where the faults are, but to go solve the problem. If something needs to move, if an email needs to go to the lawyer, our agents don't just sit around telling you what's wrong — we've trained them to go solve it and tell you, "Here's what I'm going to do." All you have to do is say yes.

Mark Young: This gets back to something I say in here, Justin: no one erects statues to critics.

Justin Girard: Exactly. Ronnie, you mentioned something interesting — that the content review tells you what's best for your brand. How does the system know that? That's a very specific statement.

Ronnie Coleman: Two things. One is that you have a brand identity and you're selling a product. But the main one is positioning. This is something I teach all the time: you can have a great product, a great brand, a great idea — but if you're not positioned correctly in the market, you'll fail. This is how big brands crush small brands all the time.

It's also why a brand like Poppi did really well. They positioned against unhealthy sodas and unhealthy food. They had a specific thing. They didn't try to do a million things. Or take Himalaya popcorn with the Buddha on it — it used to be "butter and indulgence," and now it's "you're enlightened if you eat this." Positioning is important. But to position well, you need to understand your market and what you're claiming. That research is expensive and takes time. So when we say "what's good for your brand," we're looking at the positioning data — what you can claim in what market, and how you can stand out.

Mark Young: So how do people start using the platform? What's the on-ramp, especially for small brands?

Ronnie Coleman: You go to Puntt — that's P-U-N-T-T dot A-I. The name is like punting a football, because every marketing leader we used to talk to would say, "We'll punt this over to creative," or "We'll punt this over to legal." Now they can punt it to AI.

You go and set up a demo, and someone walks through it with you. The unique thing about our demos is that we invite you into the platform and you use it in real time, with someone guiding you. On-boarding typically takes about two weeks. It's like a new employee: you give it your knowledge dump — videos, transcripts, documents — and it learns from that. It'll make some mistakes, you give it feedback, and after about a month a good employee gets it. Then it stays, and it keeps getting better over time. That's how AI works.

Mark Young: We'll put links on the website, so if you're driving, don't worry about pulling over to write it down — though it's easy enough: Puntt, with two Ts, dot AI. Justin, any last thoughts before we wrap?

Justin Girard: No, this has been fantastic. Ronnie, I really appreciate you.

Ronnie Coleman: There's one thing we haven't covered that's incredibly critical right now: AI discovery — what some people call GEO, or generative engine optimization. How are brands being discovered by new consumers?

As a founder and someone who thinks about technology, I'll tell you: for us, that's the only strategy. We're not doing traditional SEO — we're 20 years behind there, and we're not going to beat an incumbent with an incumbent's technology. But every new brand can win with AI discovery. Every brand that started 20 years ago and thought about Google as the new way to search, before it was hot, won. So the one thing I'd tell every new brand: everyone's saying you need an AI discovery strategy, and there are a million people telling you what to do. There are observability tools and agencies that help you create content and watch how you rank.

The thing we're doing that others aren't is sitting with the designers and marketers directly in their workflow. Every time you create a piece of content or launch a claim, we're also reviewing how well that content would perform in an AI search. Everyone else is over on one side creating and observing after the fact — we're directly in the workflow, before it goes to market, giving you those signals. And that matters, because whether you use us or anyone else, if you want to win, you have to go where the new consumers are buying. That's not Google anymore. It's AI.

Mark Young: And this is a real fact, folks. A lot of people are using ChatGPT like a search engine. It really should be used as a collaborator, but people are using it to search. Where we used to go to Google and type "best deodorant" and get paid ads and the highest-ranked results, people now go to ChatGPT and ask, "What's the best deodorant for someone with hyperhidrosis?" And ChatGPT gives them an answer. If you're not in that consideration set, you may be invisible to that consumer — and you can't buy your way in. You can't do paid search and buy your way into ChatGPT. This has got to be compressing a large portion of Google's income.

Ronnie Coleman: Exactly. And you've got to ask, or you get left behind.

Mark Young: I'd encourage everyone to go to ChatGPT and ask questions about your category, and see if you show up anywhere. Ronnie, what else can people do to improve their chances in AI?

Ronnie Coleman: The main thing is to have multiple tools and focus on one strategy. A lot of people think, "I'll do SEO, and I'll do AI discovery, and six other digital channels." If you're fighting a battle, you have to understand where you can win. New brands can win in AI discovery, but you need focus. Get as many tools as you can that help — and they're generally not that expensive. If anyone tells you it'll cost a million dollars to rank high on ChatGPT, they're lying. That's not how it works. You just have to do multiple things over time, similar to Google SEO — multiple things, over time, using different tools and agencies.

Have a very focused strategy and do it over a long period, 12 months, because if you get left behind, the other brands win. There are trillions of dollars in buying power shifting to AI, and you want to take a shot at it.

Mark Young: Here's what's going to happen. In the not-too-distant future, someone picks up their phone and says, "Find me the best product for toenail fungus, order it on Amazon, and send it to my house." They won't even have a brand in their head. They'll just say, "Find me a product that fixes this problem and ship it to me."

Ronnie Coleman: Exactly. And you want that to be your brand.

Mark Young: This sounds like science fiction, folks, but most people have seen Iron Man — Tony Stark has a bot called Jarvis. We can all have a Jarvis right now. Not everyone does, but we can. And before long, your bot will start to anticipate what you need. "Justin has a flight tomorrow at noon — I'll order him a car at ten, and I happen to know he prefers a black car over an X." When your needs are being anticipated, you're not even involved in brand selection anymore. It's a brave new world. I wake up every morning excited about what's going to happen and a little frightened by it.

Ronnie Coleman: That's the only kind of world I want to live in. I love both. That's what courage is — courageous people aren't the ones without fear, they're the ones who move forward despite it.

Mark Young: My friend Dan Sullivan says fear is wetting your pants, and courage is doing what you need to do while wearing wet pants.

Ronnie Coleman: I love it. That's exactly right.

Mark Young: Well, Ronnie, thanks for being with us today. Folks, if you enjoyed the show, leave us a five-star review wherever you get your podcasts, and go check out Puntt — P-U-N-T-T dot A-I. If you're using it or experimenting with it, reach out and tell us how it's working for you, so we can talk about it in future episodes. Ronnie, I hope you'll stay in touch, because I'm sure you'll have all kinds of new things coming that none of us even know exist yet.

Ronnie Coleman: Happy to. This was great. Thank you both.

Mark Young: Thank you, folks. That's it for today on CPG Insiders. We'll see you on the next episode.