Your customers hate AI. The numbers are hard to ignore. A Quinnipiac University poll found that 55% of US adults believe AI will bring “more harm than good,” up from 44% just a year ago. Booking.com‘s research puts 26% of consumers in the “AI detractor” category, actively opposed to using it. NBC News found that nearly half of those surveyed hold a negative view of AI.

This isn’t a niche concern. It’s a mainstream sentiment. And if you’re still planning to use AI to grow your business, it’s your problem to solve.

What makes this more than a PR challenge is that many of the criticisms are valid. There will be job disruptions. Deep fakes and misinformation are real. Environmental and energy costs are escalating. And when corporate leaders publicly predict 20–30% unemployment in the next few years while announcing layoffs “because of AI,” they’re making your customers’ concerns worse, not better.

This episode is about what you can actually do about it. Tim offers four practical steps that let you use AI in ways that don’t put you on the wrong side of the backlash.

Key Insights for Strategic Leaders

Whether you’re in hospitality, B2B, retail, or services — and especially if you’re the leader who has to answer “what’s our AI strategy?” while customer sentiment is moving in the wrong direction — this episode gives you a practical framework for using AI in ways that build trust rather than erode it.

Research and References

Related Episodes

Tim Peter has written a new book called Digital Reset: Driving Marketing Beyond Big Tech. You can learn more about it here on the site. Or buy your copy on Amazon.com today.

Rutgers Business School MSDM Speaker: Series: a Conversation with Tim Peter, Author of “Digital Reset”

We have some free downloads for you to help you navigate the current situation, which you can find right here:

Contact information for the podcast: [email protected]

Recorded using a Shure SM7B Vocal Dynamic Microphone and a Focusrite Scarlett 4i4 (3rd Gen) USB Audio Interface.

Running time: 24:52

You can subscribe to Thinks Out Loud in iTunes, the Google Play Store, via our dedicated podcast RSS feed (or sign up for our free newsletter). You can also download/listen to the podcast here on Thinks using the player at the top of this page.

Welcome back to the show.

I’m sure you know this, but a significant group of folks hate AI. Not, “I’m not sure if I like AI,” but viscerally, passionately hate AI.

Booking.com conducted research and found that while 49% of consumers have a positive view of AI, 26% — one in four — are what they call AI detractors. They’re actively opposed to AI.

NBC News has new research that adds to the story. In their study, 22% have a “very negative” view of AI. Another 24% have a “somewhat negative” view. Again, that’s almost half of everyone surveyed has a negative view of AI.

The Wall Street Journal piled on and noted “…in a Quinnipiac University study of 1,400 adults in March, 55% said they felt AI would bring more harm than good, up from 44% in a poll last year.”

So yeah, that’s bad.

This sentiment is breaking through to the mainstream. John Oliver had a fantastic, funny, and fearlessly heartfelt segment this past week on the very real harms of AI chat. While the episode was less about the business costs of AI and more about the societal costs, it spoke to the very real fears that roughly half of America feels right now, half of the country feels right now. The episode is worth watching in its entirety, even if you don’t agree with everything Oliver said. In fact, it might be most worth watching, especially if you don’t agree with everything Oliver said. It’s important for you to hear what people think about AI.

At the very same time, lots of corporate C-suites are pushing AI everywhere, along with “crediting” AI with helping them save all kinds of money, often through job reductions. There was another Wall Street Journal article where Verizon’s CEO Dan Schulman, quote, “has predicted 20% to 30% unemployment within the next two to five years.” I love personally that he added, “I think being authentic, being realistic, telling the truth as best as you can…” is important.

Which, I mean, I applaud his sentiment. Being authentic was a fairly crucial point in last week’s episode. But, jeez, dude. Read the room. Saying that “change is necessary, but it can be difficult,” doesn’t exactly make folks feel warm and fuzzy, especially if you’re talking about depression-level unemployment. Like that’s not the positive message that I think you think it is.

I have to make a disclosure here, by the way. I’ve done work with Verizon in the past. They’re not currently a client, uh, but they could be again, unless they really hate this episode.

The point is that there’s a very real backlash that customers feel about artificial intelligence. They’re concerned about what it means for them in their day-to-day lives. 55% of people don’t think AI will do “more harm than good” because, you know, they’re excited to see what happens. Seriously.

And many of the criticisms of AI are valid. As I talked about last week, “AI slop” didn’t become a term because people love AI content and can’t get enough of it.

So if an increasingly large number of people hate AI, if an increasingly large number of your customers have real valid concerns about AI and its use, and you are still planning to use AI to help your business become more effective and efficient, then you’ve got a real problem on your hands. Not a potential problem. A genuine, real, honest to goodness problem right now.

The question then is what do you do about this problem? This is episode 493 of Digital Reset. I’m Tim Peter. Today we’re talking about the AI backlash and what you can do about it. Let’s dive in.

As I mentioned before the break, various reports claim that 22% of people surveyed have a very negative view of AI, while another 24% have a somewhat negative view. 26% of people consider themselves AI detractors — they’re actively opposed to AI. And fully 55% of US adults believe that AI will cause more harm than good. That’s bad enough, but you need to know that that number was 44% a year ago. So the story for AI is not getting better.

First, we have to consider the fact that many criticisms of AI are built on valid points. To get a big one out of the way, there will be job disruptions. It’s not likely that there will be — we are seeing this in real time. Company after company is either laying people off because AI can do those jobs, or they’re saying that they’re laying people off because AI can do those jobs. I think the latter is more common than the former. I think they’re looking for something to blame the job losses on other than, you know, the economy or trying to save money or things along those lines, or trying to grow profit. But no matter what the reality is, AI is the bad guy. That’s not going to make your customers feel all that great about AI.

And if we’re being totally honest, of course AI is going to make some jobs go away. Technology of any kind always does.

I’ve told the story many times about my grandmother, Marie. Marie was a telephone operator, which at the time was one of the most common jobs for women in the country. She and her sister, when they were 24 and 26, were the only people working in their household during the Great Depression. The Depression had put their father — my great-grandfather — out of work. My grandmother and her sister paid for their family’s rent and utility bills and groceries. Today, the job that paid for their rent and utility bills and groceries doesn’t exist. They were the only two telephone operators I’ve ever met in my entire life. Technology made that job go away, period.

I want to be fair — there is a happy ending to that story. I’m going to come back to that a little bit.

So it’s safe to say that technology and automation and artificial intelligence will affect jobs for sure. There are also very real and very serious issues related to AI being used to create deep fakes and other types of misinformation, such as things like “nudify” apps and other more broad concerns. These uses are obviously terrible. Terrible. Horrible.

I’m not going to spend much time on these other than to say that anyone using AI to harm others — either individually or at scale, either as a human or as an enterprise — should be prosecuted and punished. Full stop. Period.

I am not trying to brush the criticism aside. It’s a real problem and it’s one that we need to do more to fix. But it’s going to take politicians and prosecutors to make that problem go away. It’s also not something that any credible business is doing, and if they are, they should be prosecuted and punished for it too.

Additional criticisms about electricity usage and environmental concerns also have a number of well-founded bases. I’m doing a bunch right now to learn more about this topic. I don’t feel super qualified at the moment to speak about it at any length.

What I do know is true is that you’ve heard me quote the philosopher Paul Virilio many times, who said, “when you invent the ship, you invent the shipwreck.” Another part of that quote continues: “Every technology carries its own negativity, which is invented at the same time as technical progress.” That’s 100% true. That’s why I use the quote so often. It was true when the internet and mobile and social came along. I’ve lived through this story a number of times in my career, and some of you probably have too.

Of course, this reality is far older than just the internet. One of my family’s go-to holiday movies, “Desk Set,” stars Katherine Hepburn and Spencer Tracy, and is about office workers at Christmas time worried that computers will take their jobs away. That movie came out in 1957, so almost 70 years ago. And no, I didn’t live through that one.

You probably also know that the word “sabotage” came about when a group of French workers who wore wooden shoes — they’re called “sabot” in French — interrupted production at their factories. I’m a little sad about this, but there’s a famous story that they threw their “sabots” into the machines to slow them down or wreck them, and that story apparently appears to be false, and I’m so bummed about that because that is a great story.

At any rate, though, the first use of sabotage to mean slowing down work came about very late in the 19th century and was undoubtedly inspired by fears of automation, machines, taking away jobs. Given that that was roughly 130 years ago, this is not a new problem. I am confident you could find earlier examples of workers who feared technology would take their jobs away. You know, as Maslow told us, people wanting to be able to pay rent and buy food is a fairly fundamental concern throughout human history.

The fact that we’ve heard these concerns before, strangely, gives me hope. We have lived through this problem in the past. And generally it’s worked out okay in the longer term. Yes, we still have concerns about the internet and social and mobile, mostly related to misinformation and social connections. But losing your job because of one of those technologies generally isn’t one of them.

I told you there was a happy ending around the telephone operator story. The truth is that even though telephone operator was one of the most common jobs for women roughly a hundred years ago, and today that job no longer exists, there are more women working just in the top 10 careers today than all of the women who worked when my grandmother was a telephone operator. A hundred years of technological development have created far more jobs, even in just 10 fields, than existed back then. And that’s true for men too, of course. The fact is that there are more than three times as many people working in the US today as there were when “Desk Set” came out 70 years ago, despite the 70 years of technological advancements since then. We’ve done more than survive that shipwreck. We’ve thrived because of technology.

Of course, that only matters if AI doesn’t make your job go away. Having more jobs in the long run is great. Your customers — and you — have to pay rent and buy food in the short run too, right? As Keynes famously said, “in the long run, we’re all dead.”

I am not trying to be provocative or morose about this. Instead, it’s one of those cases where two things can be true at once. The first is that AI isn’t a problem in the longer term, at least related to jobs. Probably. And the second is that we need to come up with solutions to help our customers, our communities, and our employees in the short term too.

The fact is — the truth is — that I’m not afraid of AI. If I’m afraid of anything, it’s the people using AI and making rules around AI that scare me more. Human beings, unfortunately, are very good at finding ways to be terrible to one another. As I’ve said in the past, terrorists have committed atrocities all around the world over the last few decades, and probably before that, without using AI. Sure, AI may help them do those things more successfully in the future, but the point still holds. They didn’t need AI to come up with some of these really awful things.

The real risk isn’t that AI is going to put everyone out of work. The real risk is that some CFO will — and my bigger concern is that they do so without helping people along the way. That’s what I think your customers care about more than anything else.

So yes, AI might create and is creating new risks. Your customers’ concerns exist because of how we use the tool and what steps we take — or don’t — to address those risks. That’s the reality. That’s what you’re dealing with.

Given that, I come back to the question of what do you do about it? What can you do if half of your customers hate AI? How can you use AI in ways that make your customers not hate you too?

And it’s a big question. It’s not one that has a simple answer. There are, however, four steps I’d think about.

First, follow the law. You know I’m not a lawyer. You know I don’t pretend to be one. I have no interest in being one. There are a number of disclosure laws on the books around AI in the US and in the EU, individual states in the US. You should be talking with competent, qualified counsel to understand your risks and appropriate responses for your business. Don’t ever break the law by mistake. Know what you’re doing, assess your risks, and act accordingly.

Second, understand the message that you’re sending with your company’s AI use. Verizon’s CEO talked about the ways that they’re investing in reskilling their employees, investing to help their employees reskill and help them deal with the shifting AI landscape. That is a great idea. Super cool, and something worth celebrating. They’re investing, quote, “a $20 million career transition and retraining fund for the ‘age of AI’,” unquote. That’s roughly $1,500 per person they’ve laid off. And while that may not sound like a huge number, it’s a good start. Frankly, it’s about $1,500 more than what many companies are doing.

So, I don’t know, maybe they’d want to lead with that versus “AI is gonna take away 30% of all the jobs.” Call me crazy.

When you choose to use AI, are you helping or hurting your workforce? Are you helping or hurting your community? That’s the question you should be asking. If you’re helping, how clear is it that that’s what you’re doing? How clear do you make that to your customers and to your community? There is no one-size-fits-all approach here. I can’t tell you in a single podcast episode exactly the words you should say or exactly when you should say them. But if you want to discuss on an individual level, give me a call.

The reality is you are sending a message when you choose to use AI. You need to make sure the message is the one that you mean to send.

Third, and very much to that end, develop your own voice for your brand, your business, and yourself, even when you’re using AI. Too many people — I talked about this a lot last week, so I don’t want to belabor the point — but too many people are using AI to turn out lots of content. Remember that content quantity is not your goal. It never should be. Content quality is.

Let’s just start with your voice. What is it that matters to you? What are your company’s values? What do you do? Why do you do what you do? And how do you convey that in your content and in your customer experiences every time a customer connects with you? How does that shine through in your day to day interactions?

Then when you’ve got that settled — then and only then — can you ask, how can AI help us do this more effectively or more efficiently? Because it’s only when you know what you want the output to look like, to sound like, to feel like, that you can say whether or not it’s doing a good job of reflecting you and your brand and your values.

If your goal is lots and lots and lots of cheap content, well, sure, AI can do that. But you’re creating commodity content by definition, and nobody gives a crap about commodity content.

If your goal instead is “tell a great story about our customers, their lives, and how our business helps them lead better lives,” then you have something where AI might be able to help. My favorite hospitality clients do this naturally. They want their guests to have a great experience and a comfortable, safe space to sleep or get some work done, or reconnect with their partners and kids, day after day after day. AI can become a tool to help do that better, not “the big scary monster coming to take all the jobs away.”

By the way, I need to note that while this might sound like I’m talking only about B2C businesses, B2B is also about making your customers more effective and efficient at what they do too. You are helping the folks who purchase your products and services do their jobs better, make better products and services for their customers. The same point about your values and your voice apply.

Which leads me to my final point. It’s also the most important point. And it is this: be human.

Your job is to create a customer, to connect with people in ways that make them want to work with your business. That’s what your job is in marketing and sales and customer experience every day.

I am confident that we are already seeing customers who won’t want to work with some businesses because of their approach to AI use and the impact of AI on their employees, their customers, and their community. Again, you need to ask, why are we using AI? How does it help us create better experiences for our customers? How does it help our customers in their lives?

I said sometime back that we owe it to our customers to make their lives better. I still believe that. If you’re not using AI to make your customers’ lives better, what on earth are you using it for?

None of what I’ve talked about is me saying don’t use AI. Far from it. I use AI all the time to solve all kinds of problems for my business and for my customers’ businesses. But it starts with saying, use AI in ways that are beneficial to your customers, your communities, your teams. Think about what you can and should disclose and in what ways you should do that to help your customers understand how and why you use AI. Again, maybe legal frameworks don’t require you to disclose your use, but does it make you a better member of your community if you do that?

It is clear that some customers hate AI. And it’s clear that even more have concerns about AI. It’s also clear that those concerns are built on valid points — they start from a ground truth that we need to acknowledge. We don’t know exactly how AI will affect the job market. We don’t know its long-term costs in terms of the environment or intellectual development. We are conducting real-time experiments at scale that will change the world we live in. And we have to be honest — just because we’ve seen similar shakeups in the past from internet and mobile and social, and going all the way back to the 19th century, doesn’t mean that it will all work out all right in the end this time. I’m willing to bet it will. But we know that there will be shipwrecks for sure along the way. Nothing ever comes without some costs, especially in the short term.

Your job is to acknowledge your customers’ concerns. Your job is to listen and learn from them about what they care about. And then your job is to help your customers understand what you’re doing to address those concerns. Ignoring their concerns will not make them go away — it will only make it look like you don’t care. And then AI won’t be the thing the customers hate. They just might hate you too. So maybe let’s not do that.

If this episode gave you a better picture of how you should approach AI for your business, do me a favor — send it to a colleague who’s working on AI for your organization. You might save them a world of trouble.

You can find the show notes for this episode and the full archive of past episodes at timpeter.com/podcasts.

And if you’re ready to go deeper on making your brand the answer that AI reaches for, my book, “Digital Reset: Driving Marketing and Customer Acquisition Beyond Big Tech,” is the roadmap you need. You’ll find the link in the show notes.

Thank you so much for listening today. I genuinely appreciate you. Until next time, please be well, be safe and be excellent to each other. I’ll see you soon.

“Digital Reset with Tim Peter” helps you look beyond the “shiny objects” to build a business that lasts. How can we help you today?

The post 55% of People Hate AI: How to Use it Without Losing Your Customers (Digital Reset Episode 493) appeared first on Tim Peter & Associates.

The cost of producing a 1,500-word article has collapsed to somewhere near zero. That’s the supply shock, one my friend Mark Schaefer has talked about for years. The more interesting question today — the one most marketing leaders haven’t priced correctly yet — is what that collapse does to everything else.

When a factor of production goes free and infinite, value doesn’t disappear. It shifts. And in marketing today, it’s shifted somewhere most content strategies aren’t looking.

Academic evidence has seen this coming. A National Bureau of Economic Research paper using Pixiv data shows that generative AI is crowding out human creators. Ahrefs data shows that 86.5% of top-ranking pages now contain some amount of AI-generated content. And Graphite.io found that the total quantity of AI-generated articles probably surpassed the quantity of human-written articles published on the web within the last couple of of years.

Additionally, research published in Nature on "model collapse" — the degraded outputs that occur when AI trains on AI output — singal a related, and more problematic reality for marketers: a "collapse to uniformity.” That’s the steady, unrelenting increase in “textual similarity” across the web since AI started publishing in the 2010s. Those similar, AI-generated outputs have accelerated ever since ChatGPT’s earliest models, and are projected to reach 90% saturation around 2035.

If that forecast holds, we’re adding roughly 10 points of uniformity — bland, boring, blah content — every year. And that means the window for you to differentiate is not some theory. It’s real. And it’s closing on you right now.

The result is two things happening simultaneously:

That second reality isn’t happening just when content was generated by a machine. It’s also happening when it doesn’t sound human enough, when it’s too corporate, too polished… too fake.

It’s fairly likely that at least 25% to 30% of your customers will actively demand authenticity within the next two to three years… if they aren’t demeaning it already. My two-to-three year forecast isn’t pulled out of thin air — it’s the sheer math of 10 points of AI detection per year building off a base of roughly 10% today.

This episode of Digital Reset with Tim Peter identifies the three specific assets AI cannot reproduce:

We also share a three-question framework for auditing whether your content strategy is actually building any of them.

The closing argument is my own practice of writing our podcast scripts by hand, despite having AI tools running in the background: not because I can write faster than a machine. We all know that’s not true.

Instead, it’s my experience, my beliefs, my humanity are the scarcest resource our content can put to work. That the distinction that matters — separating what is abundant from what is actually scarce. It’s the core claim I’m making this time. And it’s the one that will determine which brands still get seen in three years time… and which will have blended into the background.

Key Insights for Marketing and Business Leaders Navigating AI Content in 2026

In this episode, we break down:

Whether you’re a CMO deciding where to concentrate your content budget in 2026, a marketing leader who keeps being asked about AI, or a brand that’s already noticed its content working less well than it did a year ago, this episode gives you the framework to understand why… and what you can do about it.

Tim Peter has written a new book called Digital Reset: Driving Marketing Beyond Big Tech. You can learn more about it here on the site. Or buy your copy on Amazon.com today.

Rutgers Business School MSDM Speaker Series: a Conversation with Tim Peter, Author of "Digital Reset"

We have some free downloads for you to help you navigate the current situation, which you can find right here:

Contact information for the podcast: [email protected]

Recorded using a Shure SM7B Vocal Dynamic Microphone and a Focusrite Scarlett 4i4 (3rd Gen) USB Audio Interface.

Running time: 21m 52s

You can subscribe to Thinks Out Loud in iTunes, the Google Play Store, via our dedicated podcast RSS feed (or sign up for our free newsletter). You can also download/listen to the podcast here on Thinks using the player at the top of this page.

Welcome back to the show.

A few years ago, I made the claim that content wasn’t expensive. Instead, I said that content that doesn’t convert was expensive.

Today, that’s not really true either. Not because you want content that doesn’t convert. I mean, really.

But AI has essentially driven the cost of producing content to zero. You need a 1,500-word article or a 15-second video? A quick prompt, and you’ll have what you need within minutes, if not seconds.

Sure, there’s some nominal cost to you for the AI platform you’re using. And crafting a solid prompt might take you a few minutes, assuming you don’t already have an effective process or a library of prompts to work from. Both of those are still orders of magnitude cheaper than what content creation used to cost.

Hell, according to Ahrefs, 74-plus percent of all newly created web pages contained AI-generated content.

I mean, I know I use AI all the time as part of my workflow, and I’m pretty sure if you’re watching or listening to this episode, you do too. That’s part of the point.

When everyone has access to the same tools, when the tools become part of our ambient reality, the outputs of those tools become ambient reality too. We are all contributing to the flood every day, all by ourselves.

Now, before we get started, I want to be clear. Today’s episode is not anti-AI. I use AI. Clearly I’m in favor of it. Instead, it’s about the costs of marketing and where we want to and need to invest our budget and time.

For example, I had Gemini and Claude and ChatGPT doing work in the background while I was writing this script. And they took less time to complete their tasks than even writing this introduction did.

So why did I bother writing the intro at all? Why not just let AI do it?

And that gets to the core of what we need to talk about today. Because the volume of effectively free content that’s flooding the internet has driven up costs somewhere else. As content has become free and abundant, something else — something more critical — has become expensive and scarce.

And that something else, when done well, is what will differentiate your brand and it will help you stand apart in the marketplace.

That something else is what AI can’t reproduce. It can’t generate. It can’t hallucinate. It can’t fake.

Your lived experiences, your real-world relationships, your personal touches with your customers, and a genuine voice all your own are what actually are worth cultivating. When content becomes a commodity, your human connection to your customers becomes the most scarce asset of all.

Building that asset, growing it, and nurturing it is what is going to set you apart in a flood of free, fake content. In fact, I’m going to argue in this episode that you’ve got no more than two to three years to get this right. And getting it right is what will make your brand one customers will ask for by name today and in the future.

This is episode 492 of Digital Reset. I’m Tim Peter. Let’s dive in.

It’s abundantly clear that AI-generated content is here and that it’s absolutely flooding digital spaces every single day. You do not need me to tell you that. You see it every single day.

If you’re interested in some of the economic underpinnings though, there’s a National Bureau of Economic Research paper from Kim, Jin, and Lee, published earlier this year, that used Pixiv data and shows that generative AI crowded out human creators. I’ll link to it in the show notes.

Similarly, research last year from Ahrefs said that "86.5% of top-ranking pages contained some amount of AI-generated content." That’s a quote. While I have a few quibbles with their methodology, I’m relatively convinced that a fairly healthy chunk of content on the web involves AI in its production somewhere along the way, just as their research found. And again, I’ll link to their research in the show notes.

Finally, some research from Graphite.io claims that, and this is a quote, "The quantity of AI-generated articles has surpassed the quantity of human-written articles being published on the web." They claim, contra Ahrefs’ findings, that these articles "largely do not appear in Google and ChatGPT." They also ignore "AI-generated/human-edited articles," while noting that "they may be even more prevalent." Ya think?

And yes, I’ll link to their research in the show notes too.

The point is that AI-generated content is everywhere online. It’s a core part of how content comes into being today. Most of that content is, at minimum, pretty okay, and some of it is actually quite good. I’m willing to bet that you are seeing that same reality.

At the same time, AI-generated materials are so common now that we’re starting to see a bit of a backlash lately, with lots of folks talking about "AI slop" showing up in search and on their social media timelines.

That slop critique hints at the problem. It kind of gets at the core of the problem. It’s a flashing red light that says, "we need to start doing something different." And I’m going to come back to that in just a minute.

Before I do though, there’s one other signal that I have to address that shows why we need to think beyond AI content. It’s called model collapse, and it points to a likely reality your "free" content faces longer term. In fact, it’s something you might even be dealing with today.

The basic principle behind model collapse came from a paper published in Nature by Shumailov and his team — all the way back, way, way back in July of 2024 — so less than two years ago. It found that, and this is a quote, "indiscriminate use of model-generated content in training causes irreversible defects in the resulting models."

In simple English, what their research showed was that training artificial intelligence on AI-generated content generally leads to increasingly bad outputs over time, much the same way that making a photocopy of a photocopy of a photocopy will get blurrier and blurrier until eventually the output is unusable. Shumailov’s team rightly worries that as large language models get trained on the increasing amounts of AI-generated content on the web, those LLMs will produce worse results over time. That concern seems 100% valid if we agree about the photocopy example, right?

It’s also apparently not likely to happen in practice. Because follow-up research from Gerstgrasser, Schaeffer, Day, and Rafailov shows that as long as there’s at least some original human data in the model — you know, like all the web content produced before ChatGPT emerged — we’re probably fine.

But — and this is a key point — there is a related term we do need to think about. It’s called "collapse to uniformity." A group of researchers, while testing the idea of model collapse, discovered that, and again, this is a quote, "textual similarity has been steadily increasing since 2013, with a significant acceleration from 2018 to 2022, coinciding with the public release of…" ChatGPT’s earliest models. So that would be ChatGPT 2.0 and ChatGPT 3.0.

They go on to say that "the ecosystem may reach 90% saturation around 2035," and then just get worse from there.

While 2035 may sound like a long way off, that’s less than nine years from now. If you are younger than say 50, that will occur in your career. Also, if we’re going to hit 90% in less than nine years time, it suggests that we’re hitting some percentage of it today. And I suspect that we’re going to see an increase of roughly 10 points every year from now until then. I mean, if you think about it, that’s just logical. If we’re going to be at 90% roughly nine years from now, we’ll be at 80% eight years from now, and 70% seven years from now, and 50% five years from now, and 30% three years from now.

Which is where the AI slop criticism becomes so critical.

Human beings generally are pretty good — and I’m gonna use a technical term here — they’re pretty good bullshit detectors. Sure, not everyone’s good at it to an equal degree. And sure, we all have our blind spots where we can get fooled. I know I do. But human beings are generally pretty good at sniffing out when something sounds fake.

If you’ve ever heard of "the uncanny valley" in video games or CGI, it’s where a human face looks very close to real, but not close enough, and people find it disturbing. Our brains just won’t let us accept it as real. Think about a movie like The Polar Express. It just weirded people out because it was in that uncanny valley — close to real, but not quite close enough.

AI slop, to me, is the uncanny valley of AI-generated content. AI-generated content sometimes doesn’t sound human enough and sets off your customers’ bullshit detectors, and they flag it as AI slop. They’re the first 10% this year noting that something seems off in what companies and creators they’re listening to are saying.

The funny thing about this is that sometimes people online accuse creators of AI slop even when a human being actually created the content. And it’s not because it was AI-generated. It was because it didn’t sound human, or at least human enough. It was too corporate. It was too polished. It was too fake.

I’d argue as more people get wary about the text patterns and speech patterns of AI, they’re going to also get more wary of corporations and creators relying on heavily polished statements. Marketing thought leaders — including people like me — have claimed for years that it’s important to be authentic when you speak to your customers.

If we’re picking up 10 points of AI detection every year, it’s likely that a healthy chunk of customers — 25 to 30% — will demand that you be real, that you be clearly real, in the next couple of years, if they’re not already demanding it of you today. That’s where I got that two-to-three year number in the intro.

By the way, as a quick aside, I cannot talk about this topic any further without referencing my very good friend, Mark Schaefer. Mark has written extensively on this topic and has influenced my thinking significantly. His mantra about how the most human company wins in the age of AI echoes in my brain almost every single day. I cite him in my book. He’s just a key, key speaker on this topic that you want to know about. I’m going to link to some of his thoughts in the show notes, and I would encourage you to check those out and follow him regularly.

At any rate, that’s why focusing on human connection to your customer isn’t just scarce. It’s the most precious asset you can build. Your voice — the true you — is the most precious asset you can build.

That’s why your content can no longer just be a generic explainer or a how-to video. They’re dead. They don’t work anymore. They don’t matter.

Instead, you have true, scarce, valuable assets that your content must put to work today. What are they? Well, they include several things.

The first is proprietary data. Data that you have, that you can see, and that no one else can about your customers, your market, your industry, and the world at large is something that no one else can talk about with the genuine expertise and insight that you bring to the table. It’s your story to tell, and you need to tell it if you’re going to stand apart from generic AI-generated drivel.

The second are original examples. Talk about what you’ve lived in real life — your customers, your projects, your outcomes, your experiences. There’s a reason I talk about my hotel clients and other businesses I’ve worked with on this podcast, and that’s because no one else has those specific experiences. They back up my thinking and research with real-world examples that no one else can claim. It works for my business. It’ll work for yours too.

The third, of course, is expert voice. Your expert voice. Did you notice my two-to-three year timeline that I mentioned earlier? That’s my opinion. I don’t have anything other than my expertise, my experience, and my assessment of the data that exists to back that up. I could be wrong. I could absolutely be wrong. Maybe it’s four years. Maybe it’s one and a half years. I don’t know, but I’m willing to put my name on it and make a prediction to stand out from the crowd. I’m willing to take the risk of being wrong because I believe it will help you now and it will help you in the future. And candidly, because it will help others know who I am too. If I’m right, everybody wins. If I’m wrong, I will either learn and improve — or no one should listen to me anyway. Right? Period.

Together, those three attributes help me build a connection with my clients and listeners to the podcast and audiences on social media. They all work together to build the Digital Reset brand and to foster a community of like-minded individuals who are committed to learning and growing together. That same approach can work for you too.

Here’s how I would do it.

Start by asking yourself three questions this week.

First, what is our ratio of proprietary to commodity content over the last 90 days? Are we building named assets all our own, or just generic outputs that any AI could produce?

Two, when a potential customer asks ChatGPT, Gemini, Claude, or Perplexity about our category, what specifically about us shows up? Is it a reference that could have been written only by us? Or could it have been written by anyone? One of those is better for your brand long-term than the other, as you might expect.

Three, who on our team is our named voice? Are we amplifying their signal? Are we boosting their signal? Or do they get buried in the content calendar of just generic stuff we put out there? Think about which of those has a future too.

One big takeaway these questions should point you to is to focus on what makes you… you. What makes your brand stand apart? What makes you distinct? What makes you different from anyone else in the space — and different enough that people will listen to you, and different enough that eventually people will give you money? Not for your content, but for the solutions and products and services that you offer. That’s where you want to live. That’s where you need to live — because in two to three years, if you don’t already live there, you are in big, big trouble.

Also, I don’t want you to listen to this episode and necessarily think, "we need to publish less content." I mean, sure, this is about quality over quantity. Absolutely true. But it’s about publishing more of what’s scarce and less about publishing what’s already abundant.

That’s why I wrote the introduction to this episode — and the whole episode, for that matter. Not because I can write faster or cheaper than artificial intelligence. I can’t.

But because only I can produce the most scarce resource of all: me. For better or worse, when you’re hearing these episodes — when you’re watching them on YouTube or listening to them on Spotify — you’re hearing my experiences, you’re hearing my data, you’re hearing my voice. And that will be true today, two to three years from now, and for as long as you’re willing to keep listening.

AI did not kill content marketing. AI will not kill content marketing. What it’s done is make the cheap parts worthless and the expensive parts crucial. Your job is to know which is which. Your job is to put your focus on the parts that are crucial. And it’s to do that now and in the longer term too. And if you get that right, you don’t have to worry about the next two to three years or the next nine. You’ll be in great shape no matter what happens.

If this episode gave you a clearer picture of how you should approach content marketing in the age of AI, do me a favor. Send it to a colleague who’s thinking about that problem for your business. You might just save them from going down the wrong path.

You can find the show notes from this episode as well as the full archive of all past episodes by going to timpeter.com/podcasts.

And if you’re ready to go deeper on making your brand the answer that AI reaches for, my book, Digital Reset: Driving Marketing and Customer Acquisition Beyond Big Tech, is the roadmap you need. You’ll find the link for that in the show notes too.

Thank you so much for listening today. I genuinely appreciate you. Until next time, please be well. Be safe. And be excellent to each other. I’ll see you soon.

“Digital Reset with Tim Peter” helps you look beyond the "shiny objects" to build a business that lasts. How can we help you today?

The post AI Made Content Free. Here’s What It Made Priceless (Digital Reset Episode 492) appeared first on Tim Peter & Associates.

Your marketing goal should be to win no matter what happens in the digital space. Yes, the channels your customers choose first change all the time. We’ve gone through search and social and now AI. Just remember that tomorrow could be something else altogether. You don’t want to chase channels. You want a strategy that works no matter what. And that’s what the “Hub and Spoke” and “CORE” methodologies are all about.

The CORE Methodology is a framework for choosing the right channels to market your company. Your “Hub and Spoke” is about how you use those channels — the spokes — to grow your hub (website, CRM, community).

Together, both work to ensure you win no matter what. This Foundation episode of Digital Reset with Tim Peter breaks down what you need to know so you can win no matter what happens with AI… or whatever comes next.

Key Insights for Marketing Strategy Leaders Navigating the Shift to AI

Want to learn more? Here are the show notes for you.

Tim Peter has written a new book called Digital Reset: Driving Marketing Beyond Big Tech. You can learn more about it here on the site. Or buy your copy on Amazon.com today.

Rutgers Business School MSDM Speaker: Series: a Conversation with Tim Peter, Author of "Digital Reset"

We have some free downloads for you to help you navigate the current situation, which you can find right here:

Contact information for the podcast: [email protected]

Recorded using a Shure SM7B Vocal Dynamic Microphone and a Focusrite Scarlett 4i4 (3rd Gen) USB Audio Interface into Logic Pro X for the Mac.

Running time: 25m 35s

You can subscribe to Thinks Out Loud in iTunes, the Google Play Store, via our dedicated podcast RSS feed (or sign up for our free newsletter). You can also download/listen to the podcast here on Thinks using the player at the top of this page.

Gatekeepers always offer you a shortcut, at least at first. That’s what I talked about in episode 489, and it’s why ChatGPT started charging brands $200,000 to advertise alongside answers that used to appear organically for free. That shortcut eventually becomes a toll road, always.

I glossed over one key detail in that episode.

The way out of the shortcut trap isn’t a better tactic. It’s a strategy. That strategy is what the episode you’re about to hear is all about.

A quick note before we dive in. If you’ve been listening for a while, you may know this show as Thinks Out Loud. We’ve rebranded to Digital Reset. The same show. Same host. Sharper focus on what you actually need to navigate AI and digital change right now.

The episode you’re about to hear was originally recorded in June, 2024 when everyone was worried about what Google might do next. I think you’ll find it’s aged remarkably well.

The framework at the center of this episode is the "Hub and Spoke." The hub is the audience you own: your website, your email list, your direct relationships.

The spokes are everyone else’s audience: search and social and AI and creators and employees and customers and press. The job of the spokes has always been the same: Grow the hub, not replace it, grow it.

In 2024, that was a good strategy. In 2026, it’s survival.

Your hub is your only durable defense against platform risk, including AI. So sit back and enjoy our foundation episode, “Win No Matter What: The Hub and Spoke Strategy.”

Let’s do this.

There are four ways to succeed:

If, though, you are worried about where you find these folks and how you reach these folks and you get them to do this, I always like to start with where you are first. Where does your traffic and revenue come from today? I assume that you’re using Google Analytics or Adobe Analytics or something along those lines to understand your traffic and revenue sources.

Whenever I start with a new client, whenever we start with a new client, we tend to look at the last 13 to 18 months of traffic and revenue. Then, we might look at a shorter period like the last two or three months, and then we might look at the last few weeks, a week at a time, to get a sense of what’s going on.

Looking at the longer periods helps you to understand if there’s any seasonality. You know, does your company get more business some times of the year than others? That can help you understand if there are different demand patterns or if there’s something that you can do differently to get more traffic and business at some points in the year.

If you’re not accustomed to seasonality, if your business doesn’t have it, it’s pretty common in many businesses. I talk about hospitality a lot. Hospitality tends to have very well defined peak seasons, off seasons, and what are known as “shoulder seasons,” between the peak and off season. We also see it in B2B a lot more than you think, too, based on budget cycles, you know, things like quarter end, year end, start of the new year, that sort of thing.

Getting a sense of that picture is really important because it can point you to traffic sources that you can build on, or revenue sources that you can build on.

It is sometimes easier to increase the traffic you get from existing sources, than it is to find or create entirely new sources of traffic and revenue. If you’re already getting traffic and revenue from a specific source, that’s a pretty good sign it’s something that works for your customers. And then it becomes more a question of how do we grow that source rather than “Oh my gosh, we don’t get any traffic outside of Google. Where do we go?” So it’s always a good place to start to look at the reality and understand are there any seasonal patterns that play into that. Are there times where you can expect more traffic or less traffic just as a natural consequence of where that comes from and of your customer’s natural behaviors?

The next thing that you want to do is ensure that you have a content calendar. Now, I talked in the last episode quite a bit about understanding your customers as well as you possibly can and what your customers pain points are.

We love doing this because it allows us to line up specific messages with specific seasonality where it exists. Again, if you don’t have seasonality, don’t worry about it, but there are natural things that occur during the course of your business year or occur in the natural course of your customers lives that might lead them to look for your product and service more often.

That’s really important then that you have a calendar that aligns with that. And we always think in terms of a “rock/pebbles/sand” approach.

Your rocks are the big items, the things that you know will happen, and you can plan those out weeks or months or maybe even a year in advance. You know, again, if I’m going to use the hospitality example, summer is a peak period for many, many companies, for many, many resorts, for many, many hotels. If you’re a ski resort, we know ski season is a big peak period. Well, we can put a rock saying, we need to start talking about this X weeks before, months in advance, even a year in advance.

If you’re a B2B company, maybe it’s a big trade show that occurs every year. We worked with a software company for years that went to Dreamforce every year, and that was a rock, because they knew when it occurred. If you’re, I don’t know, a restaurant, and you know the Super Bowl is going to be in your town next year, you can start planning your content for that.

Now, because we already know when that’s going to occur, we already know when there’s going to be more demand in your market. You then start to fill the calendar with pebbles.

And the pebbles are little things, smaller things, more tactical, but messages that you want to dribble out over the course of the year, that fill in some of the spaces between the rocks, right? If you think about we’re putting rocks, pebbles, and sand in a jar, you put the rocks in first, and then there’s spaces between them, and that’s where the pebbles go.

And you can plan those weeks, or even a couple of months in advance. Usually about 90 days is pretty typical.

And then the sand is small little tactical stuff. Maybe they’re social media posts. Maybe they are little blog posts or quick little videos that you’re going to put up on your website. And those get sprinkled in where there’s space.

It’s okay if you have a period where you don’t have anything to say. Just make sure that you’re doing that intentionally, you’re choosing not to speak there, you’re choosing not to create content and put it out there because it doesn’t work for your customer at that time of year. It’s only a mistake if you completely ignore your customer’s behaviors and you’re not taking advantage of it.

Now, one of the reasons that search has worked for so long is because everyone searches and they’re literally telling you what’s most important to them. Google didn’t get to an $80 billion per quarter revenue company, a $300-plus billion annual company in revenue, by ignoring what people typed into the search box every day.

Instead, they listened to those, they interpreted those searches, and they gave customers what they were looking for both in the search results and in the way they’ve modified the search results over the years to put the information in front of customers quickly and easily. That’s how we’ve ended up in this place in the first place.

Look at your site search, if you have it, as well as the terms that are driving traffic to you because those are the things that customers are saying are important. You can use tools like Google Search Console, you can use tools like Google Analytics to look at your site search. You can use tools like SEMrush or Ahrefs to look at the types of terms that your company and your competitors rank for to get a sense of the kinds of messages and the kinds of information that your customers are looking for.

Now you’ll note, in all of this, we haven’t talked about channels yet. We’ve always been talking about how do we understand our customer, how do we understand our objectives, how do we understand our content. And the reason is because the channel and the format come last.

First, you have to understand who are you talking to and what do they need to hear you say? Then you can think about “Where do I say this and in what form do I say this?”

We have always used a framework called “Hub and Spoke,” where the hub are things you control, and the spokes are things other people control.

One way to think of it is, your audience is the hub, and the spokes are other people’s audiences, you’re leveraging other people to help you reach those folks.

The hub is your audience. It’s the people who are actively seeking you out. It’s your website. It’s your app. It’s your email list or your CRM. And I’m going to come back to that in more detail in a minute.

But I want to start in detail talking about the spokes. The spokes are other people’s audience. And the job of the spokes is to grow the hub. You don’t just go out on the spokes for no reason. Ideally, you turn those spokes into customers. But when you turn them into customers, even if they buy on a spoke, you should have some contact information for them that goes into your CRM, which allows them to come back to your website or your app the next time directly without being on a spoke.

So the spoke is where we’re working with other people’s audiences. Google, obviously, has long been the biggest. It’s the whole reason we’re having this conversation.

Obviously, social is huge too. So that could be Instagram, it could be Facebook, it could be TikTok if you’re B2C, if you’re selling directly to consumers. It might be LinkedIn if you’re a B2B company, selling to businesses. You know, LinkedIn is my go to social network, but they are places where you connect with other people through other people.

It also doesn’t mean you’re creating your own content. You may be partnering with creators who have a big audience, or have, I shouldn’t even say “a big audience;” they have “the right audience.” They’re talking to your customers.

So you might partner with creators. You might work with them to say, “Hey, can you help us craft a good message for our customers who you also happen to be talking with?” That could be a partnership. It could be something where you’re paying them. It could be where you’re commenting and engaging with others content, but to be a valuable member of the community first. You’re not just going in there and saying, “Hey, let me jump into your conversation and promote me.” It could be you’re engaging with them in a really effective way by being a member of their community and then letting that organically grow into people knowing about you from that over time.

I wrote a piece of research for meta platforms and HSMAI a few years ago during the pandemic on how you can research your other creators and learn about other creators and work with other creators over time. I definitely recommend you checking that out. I’ll have the link in the show notes. But the point is that you’re letting the creators help tell a positive story on your brand’s behalf.

Another type of creator you might want to think about are your employees. I’ve worked with many companies who’ve had great success with an employee engagement program where you’re leveraging your employees social networks to talk about your products and services. You help your team understand what they should be talking about, you provide them some guidelines, you provide them some content assets, and you provide them some recognition for the work that they do. Could be a thank you, could be they win an internal contest, it could be that they get a, you know, a bonus or something along those lines. But it’s all about engaging the employees in telling a positive story.

I will be very clear, this works better for some companies than others because sometimes your employees are very representative of your company and sometimes they’re not, so you have to be clear about that. But when it works, it can work tremendously well and you’re increasing your network by the size of your total employees times the size of their network.

The other great form of engagement that you can work with as creators are your customers. Your secret sales forces, I call them, when they talk on ratings and review sites. You’ve heard me say a billion times on this show that customer experience is queen. And the reason that’s so is you want to create an experience that your customers want to talk about, that they want to talk about positively to their network.

The average person on social, whether it’s your employee, whether it’s your customer, has about 200 friends and family and fans and followers on social. So, how are you engaging them to tell a really positive story about your company and about your business? Well, you’re doing that through the experience that they get.

You also can look to PR. Are there, you know, and when I say PR, we can talk about traditional journalists or we can talk about people who are, you know, creators and what they’re doing. How are you creating a great experience that they want to share, that they want to talk about? Your goal is to build a community of people who want to tell a positive story on your behalf, whether that’s journalists, whether that’s creators, whether that’s employees, whether that’s customers.

All of these are people who have an audience that you can build off of. And of course, there’s all sorts of other things you can do. Direct mail, loyalty programs, conferences and trade shows, trade organizations, traditional advertising, paid media, etc. But the point is, you’re looking for the ones that work for your business and help you connect to your customers.

Another thing you might want to look at is what types of content do your customers engage with? Are they engaging with written articles either on your own site like a blog or on using other people’s audiences like a spoke? Are you writing in a trade journal if you’re a B2B company? Are you writing it, you know, are you getting coverage in a An online magazine if you are a B2C company.

You can do things like polls. You can do things like video and images and audio, like, I don’t know, podcasts. Now, I want to be very clear. You don’t necessarily need to do the video. You don’t necessarily need to do the audio. You don’t necessarily need to do the poll. Those can be great places to partner with creators who do that well.

The reality, though, is you want to give the people who create those something worth talking about. So, for instance, if you’re doing a podcast, why are you a good podcast guest? You find somebody who is a podcaster who speaks to your audience. What makes you a good guest?

I get pitches every day from people who say, Oh, we’ve got somebody who’d be a terrific guest for your podcast. And it’s clear they’ve never listened to the show once. Not at all. They don’t understand the show. They don’t get the show. And they’re saying, we’ve got somebody who wants to come on and talk about, you know I don’t know, women’s fashion. Nothing against women’s fashion, but it’s clearly not something we have anything to do with. They’re not being a good part of the community.

Another thing you can do is think about images. Think about things on Pinterest or Instagram. Now, the term infographics? Nobody uses any longer, but the idea of beautiful data visualizations still always works as a content mechanism. I have clients who have tremendous success with this. If you have research that’s important to your customer and you can package it in a way that’s beautiful, folks will share it on your behalf. You could do this in B2B with “Here’s something you want to know about the market today.” You can do it in B2C. You know, “Here’s a hot trend in—I don’t know—women’s fashion that you really should know about.” Right?

But the point is you’re trying to engage other people’s audience. And the way you’re going to do that is engage those people to want to share your story. You have to give them something. You have to give them a hook that’s going to be interesting to them.

The thing that you want to remember is the point is to get it to grow your audience, your hub. That’s the job of any of these spokes. And there’s a model we use for picking these. You’re not going to do every single one of these. I’m just giving you, you know, ten different high level ideas. The reality is you don’t have the resources to do all ten. So keep in mind what we call “Core and Explore.”

Remember when I talked about where your traffic and business comes from today, where’s 80% of that? What’s the 80% that’s most important to you? That’s where you should be putting most of your efforts. That’s your “Core.”

Then spend maybe 20% of your time on “Explore,” finding new channels, finding new opportunities. You’re not exploring for its own sake, though. You want to find things that can become part of your core over time.

And we use a framework that I call CORE to choose those:

So, Customers. Are your customers on that channel? Do they use or interact with that channel regularly?

Outcomes. Is this an effective platform for delivering your key message? Can it help you reach more of the people you want to reach and drive more business? And most importantly, does it actually drive more business? As you test it, are you seeing positive business results from it?

Resources. Do you have the knowledge and the skills needed to use this platform effectively? If not, how will you close those gaps? That’s a great place where working with a creator can work for you because they already know how to close those gaps.

And of course, Execution. What is your plan? What are your next steps to get started?

So as you evaluate potential channels, look at “Customers, Outcomes, Resources, Execution (CORE),” and determine which ones work best for you.

Once you’ve answered those questions, take your plan out and test it. And as you learn, see if you can make it part of your core activities. Ultimately, that’s where we’re going to go.

Now you’ll note in all this I haven’t talked about search because my concern is how Google is going to change search over time. I also want to be clear, search isn’t completely going away.

What’s also true is that the best way to beat Google at its own game is to have customers search for you by name. Every single one of these activities that I’ve mentioned could actually increase your search traffic. The difference is, customers won’t be looking for, you know, “generic service near me,” “hotels near…,” or, “restaurants near…,” or “businesses who solve X problem.” They’ll be searching for you by name. And that’s the long game.

Because then no matter what Google does, you win:

So to recap everything we’ve talked about the last two weeks

And that’s the whole point of this, is to make sure that no matter what anybody does, you’re still in a position to succeed. I can’t wait to see what you do.

 I want to close this episode with something I said at the end of the original recording. I think it’s very much worth repeating. The best way to beat any gatekeeper at their own game is when customers ask for you by name. It is not about relying on a platform. It’s not about buying ads, just your brand by name every time.

That was true of Google in June of 2024. It’s true of ChatGPT in April, 2026, and it’ll be true of whatever comes next. The long game doesn’t change.

If you want a practical playbook for playing that long game in an AI first world, check out my book, Digital Reset, Driving Marketing and Customer Acquisition Beyond Big Tech. The link is in the show notes at timpeter.com/podcasts.

Also subscribe so you don’t miss what’s coming and connect with me on LinkedIn. Just search Tim Peter. Or email me at [email protected]. I always love hearing from you.

This has been Digital Reset. I’m Tim Peter. Thanks for being here. I’ll talk with you next time. Take care everybody.

“Digital Reset with Tim Peter” helps you look beyond the "shiny objects" to build a business that lasts. How can we help you today?

The post Win No Matter What: The Hub and Spoke Strategy (Digital Reset Foundations 491) appeared first on Tim Peter & Associates.

ChatGPT launched ads in its responses earlier this year. But they weren’t for everybody. They couldn’t be. Their ads came saddled with a $60 CPM and a $200,000 minimum spend. That pricing is roughly in line with prime-time NFL inventory. Naturally, the tightly-managed pilot started with only around 600 advertisers.

OpenAI might be new to the gatekeeper game, but they sure understand how to collect a tax on the traffic they offer. According to CNBC, OpenAI’s ad pilot crossed $100 million in annualized revenue in under two months.

Now, Search Engine Land is reporting that OpenAI is bringing self-serve access — and getting rid of the $200,000 minimum — this month.

Early performance data around ChatGPT’s ads isn’t as simple as OpenAI’s robust revenue headline suggests. One trade publication put it bluntly: "ChatGPT’s first advertisers can’t prove their ads worked."

The big picture is more complicated though. Yes, click-through rates are low and reporting tools have had challenges. But Criteo — the first ad-tech partner integrated with ChatGPT on the pilot — says that LLM-referred users convert at roughly 1.5 times the rate of other referral channels.

Why? Because it looks like ChatGPT’s ads are a brand awareness channel, not a performance marketing one… at least for now.

This episode of the podcast serves as the paid-media companion to Episode 489’s "The Long Game." The shortcut trap that Tim described in the last episode — where every new gatekeeper offers cheap access early, then raises the toll — is coming for AI.

The question for you isn’t whether you should advertise on AI platforms. Instead, it’s what are you you’re building while the rates are still relatively low… and whether your brand has organic signal worth amplifying in the first place.

This episode of the podcast delivers a three-question framework to help you make the right decision around ChatGPT’s ads. Tim also explains why OpenAI almost certainly will have to change how their ad model works — and why that might make now potentially the cheapest moment to learn how this channel will work for your business.

Key Insights for Marketing and Business Leaders Navigating AI Advertising

Whether you’re a CMO deciding how to allocate a test budget, a marketing manager preparing for the question from your CEO, or a small business owner trying to understand what’s happening in AI advertising, this episode of the show gives you the framework to answer the right questions before you commit.

Tim Peter has written a new book called Digital Reset: Driving Marketing Beyond Big Tech. You can learn more about it here on the site. Or buy your copy on Amazon.com today.

Rutgers Business School MSDM Speaker: Series: a Conversation with Tim Peter, Author of "Digital Reset"

We have some free downloads for you to help you navigate the current situation, which you can find right here:

Contact information for the podcast: [email protected]

Recorded using a Shure SM7B Vocal Dynamic Microphone and a Focusrite Scarlett 4i4 (3rd Gen) USB Audio Interface.

Running time: 19m 59s

You can subscribe to Thinks Out Loud in iTunes, the Google Play Store, via our dedicated podcast RSS feed (or sign up for our free newsletter). You can also download/listen to the podcast here on Thinks using the player at the top of this page.

Welcome back to the show. Last week I laid out the shortcut trap, the pattern that every new gatekeeper follows, where early access looks cheap, the shortcut looks smart, and by the time the toll arrives, too many brands have built too much of their strategy depending on the thing that’s about to change the terms on them.

In particular, I focused on how the shortcut trap applies when you’re working to show up in AI answer engines. This week I want to talk about the paid media version of that story because that’s where the tolls that the gatekeepers want you to pay show up pretty much every single time.

This isn’t some theoretical future scenario. OpenAI has had ads live on ChatGPT for roughly 600 advertisers since February. And those ads were not cheap. OpenAI required a $200,000 minimum spend and roughly a $60 CPM, about three times what you’d pay on Meta. Hell, that’s roughly comparable to what you’d pay for primetime NFL inventory. And it’s worked out well… at least for OpenAI. CNBC confirmed last week that this pilot they’ve been running has already crossed $100 million in annualized revenue in under two months.

Search Engine Land is reporting that self-serve access and the elimination of the $200,000 minimum spend requirement is coming in April. I’m putting air quotes on that because we don’t have a firm date yet. But basically they should be live this month.

And of course, Google has had ads in AI Mode and AI Overviews, at least in English, in roughly a dozen countries for some time.

I’m going to limit today’s conversation to ChatGPT only. Google is Google after all. And we’ll get a much clearer picture of their situation in a few weeks at their upcoming earnings call.

The fact remains that up until now, ads on ChatGPT was an enterprise level brand decision. With the introduction of self-serve, though, it’s about to become a decision every marketing leader is going to be asked about, doesn’t matter the size of your company.

So what’s going on with ChatGPT’s ads? How are they doing?

Well, the early performance data is mixed. One trade publication’s headline was "ChatGPT’s first advertisers can’t prove their ads worked." Woof. I want to be fair. That doesn’t mean that that takes them off the table. But it is a pretty fair if harsh description of where they stand today.

The channel isn’t proven yet. That’s a fact.

What I intend to do though, is give you a framework to decide whether ChatGPT ads belong in your budget — and what you should be doing, whether you use them or not.

This is episode 490 of Digital Reset. I’m Tim Peter. Let’s dive in.

Okay, so let’s get some big questions out of the way. As I mentioned before the break, OpenAI says they’re launching self-serve ads this month. They’re also saying that the $200,000 entry fee is going away. So for most businesses, that’s not really something you need to worry about. You don’t need to worry about those.

You don’t have $200,000 to spend. Don’t worry about it. You worried when it’ll be ready? Well, they’re coming within weeks. In short, you’ll be able to test ChatGPT ads if you choose, and I promise the rest of this episode will help you answer the question of whether or not you should choose to do so.

Before I do that though, let’s be clear about what’s happening here. I’m not going to spend too much time on setting up what the ads are and how they work because frankly, OpenAI could change the way this works at any time. In fact, as I’ll argue in a bit, they’re almost certainly going to have to change.

At any rate, the way OpenAI is handling ads on ChatGPT right now is to show them only at the bottom of ChatGPT answers. They’re clearly labeled as ads. They’re only shown to users in ChatGPT’s free and "Go" tiers. In other words, customers in the Plus, Pro, Business, Enterprise, and Education tiers won’t see the ads.

While that might sound like a lot of people in practice, something like 85% of ChatGPT’s users fall into the two tiers that are eligible for ads. At the same time, according to CNBC, this is a quote, "less than 20% are shown [ads] on a daily basis."

It’s not clear to me if that’s 20% of all users or only 20% of the 85% who could in theory see them. Since that’s still 17%, I’m not gonna consider that a material difference. And according to OpenAI, its "ads do not influence the answers ChatGPT gives you."

As I also mentioned before the break, that early performance data on the ads is mixed, which I’d think you’d expect, given that at best, only about 17% to 20% of their total audience sees these ads regularly. I’ll link to the various reports in the show notes.

The early data suggests though, that they’re seeing very low click-through rates, strikingly higher conversion rates, and some real reporting challenges in the early days. We can ignore the reporting challenges. They’ll fix that.

Criteo said that users referred from LLM platforms like ChatGPT convert at approximately one and a half times the rate of other referral channels. That’s actually pretty good.

And if we think about the low click-through rates and the high conversion rates, those make some sense.

Low click-through rates seem pretty likely given that the ads are down at the bottom of the page. And given that most conversations with large language models like ChatGPT are about getting answers, why would anyone need to click an ad once they got their answer? Assuming they ever scroll down far enough to see them.

Again on the conversion rate side, that also follows logically that it would be higher than you might expect. Because the ads shown really function more as a brand awareness tool than a direct response tool. Again, the user got an answer and the ad shows a product, service, or brand that’s related to their answer.

I suspect, and I want to be fair, I have no evidence for this other than my intuition and experience, that the ads getting actioned accompany responses containing a brand.

Well, what makes me think this? So we have seen over the last couple of years a number of clients getting increased branded search traffic because AIs are citing them in their answers.

We know that branded search converts at a much higher rate than unbranded search. And I think something similar is happening here. If you’re using these ads, your brand is showing up in the answer, it’s showing up in the ad, or it’s showing up in both.

That’s driving increased awareness of your brand and attracting people who are interested in your offering. The folks seeing those ads become more likely to buy because of the trust signals in the answer and the ad together.

And because they’ve got some affinity for your brand from seeing this, those folks likely convert at a higher rate. They’re like branded searchers. Probably.

Obviously I’d love to see more data, but my gut and my experience in digital marketing make me feel slightly confident about this. I want to be fair, this is not guaranteed. I’d peg my confidence level at maybe 60% to 70%. But it feels kind of likely that that’s what’s happening here.

In either case, I think these ads today are much more of an awareness play than a performance marketing play. Some of the data does seem to support that, and again, you’ll see those in the show notes.

Another point that you want to consider as you think about this tool, this platform, is that you don’t want to get too excited or pulled in by OpenAI’s reported hundred million dollars in revenue.

Just because they’re making money is no guarantee that you will make money.

We are in very early days of these tools and there’s very little data to go on at the moment, so you’re going to want to proceed with caution.

I’m going to give you a framework to think about how you can proceed, but I do want to touch on one point before I get there.

Earlier I said that ChatGPT almost certainly will have to change how their ads work. Their $100 million in annualized run rate kind of demonstrates why. While we’re a few weeks away from Q1 numbers, we know that in Q4 Google made almost $63 billion of its Q4 revenues — fully 55% of its Q4 revenues — from search ads.

If AI ads don’t work, advertisers aren’t going to use them. And the AI companies’ ad platforms won’t produce revenue for them.

OpenAI’s $100 million in annualized revenue is a great number. Seriously. And the fact that they managed to pull that off since February is extraordinary.

What’s also true is that Google made just short of $63 billion from search ads… in a quarter. That’s an annualized run rate just short of $251 billion from search ads alone.

For OpenAI to equal that, they’d need to grow that $100 million number over 2,508 times larger.

Again, I want to be really fair to OpenAI. If they manage to only achieve 1% of that, that’s still $25 billion in annual revenue. That would be an extraordinary, remarkable accomplishment.

It’s also only 1% of Google’s revenues. Put it another way, today they’re pulling in roughly 4% of 1% of Google’s search ad revenues. Not total revenues. Search ad revenues. Period. That is not what their investors are looking for. No way, no how.

As Matthew McConaughey famously said, "you gotta pump those numbers up. Those are rookie numbers." And that’s not going to happen if they only show ads at the bottom of the page to 20% or less of their users. Nope, not gonna happen.

I can’t tell you how they’re going to change the model. I can’t tell you when they’re going to change that model. But they’re going to pump those numbers up. I can almost guarantee it.

If you look at it that way, now might be the cheapest time you could ever choose to test their ads.

So the question comes back around, should you look at ChatGPT ads? Should you test ChatGPT ads for your business?

To answer that question, I’d like you to consider a framework built on three key questions.

The first is, what does AI know about your brand right now? You’ve heard me say this in past episodes, I’ve proposed a test that you can and should conduct so you can learn what AIs think of your brand.

Before you think about running ads in ChatGPT, ask what it knows about your brand. Ask ChatGPT, ask Gemini, ask Perplexity to describe your company. I would prompt it first to act as a novice and then to act as an expert. For example, if I were looking to see what it knows about my company, I’d ask, "tell me what you know about Tim Peter and Associates."

Afterwards, I might provide a more sophisticated prompt, such as, imagine you’re an expert in B2B consulting services with a 180 IQ, and a specific expertise in digital strategy for hotels. Give me your honest assessment of Tim Peter and Associates. Assume that I’m watching you closely and have this expertise myself. Get this right.

Obviously substitute your brand in place of mine. But see what the AI comes back with.

Whatever an AI hedges on or gets wrong is the gap you need to close. Advertising on top of a confused AI signal doesn’t fix the confusion. It amplifies it.

If ChatGPT and other LLMs cannot describe your brand accurately in an organic response, a paid placement at the bottom of that same answer isn’t going to help you.

Your number one task then has to be to help the AI learn the truth about you. And I’ll provide links in the show notes for past episodes where we’ve discussed how you do that.

The second question you need to ask is something I talked about last week. Are you building a foundation for your business or are you renting visibility? AI advertising that helps you drive email capture and direct relationships and app downloads or enrollment in your loyalty program, builds something that persists even after the campaign ends. That’s a foundation you can build on over time.

AI advertising that just drives traffic to your website, traffic that returns to zero when you stop spending, is rent. And rent isn’t necessarily wrong. But you’ve gotta know which one you’re paying for before you commit.

The third question is also from last week’s episode. Would this investment you make still matter if the AI platform changed their algorithm tomorrow?

Most AI advertising currently fails this test. Which means what you’re buying is brand awareness, not infrastructure. Don’t get me wrong, brand awareness is a legitimate goal. But only if brand awareness is what you actually need right now.

Those three questions: "what does the AI know about your brand? Are you building a foundation? And would the investment still matter if the algorithm changed tomorrow?" are where I’d start.

AI ads that help you drive direct relationships are worthwhile. Again, email list signups, first-party data collection, loyalty program enrollment, app downloads. The ad spend might be transient, but the relationship it creates when done well is permanent. That’s always something worth investing in.

This is the same lesson that independent hotels that have survived in the era of OTAs have learned. Hoteliers who used early OTA access to build direct booking programs and build up their CRM had options when costs went up.

The ones who only used OTAs to fill rooms were hostage to the commission structure.

You know that I’m a fan of playing the long game. I think there’s a long game to be played here too.

Again, ChatGPT’s revenues are 4% of 1% of Google’s search ad revenues. Search ads are the biggest driver of ad spending in the world, and AI search ads are projected to grow 10 times over in the next three years. That’s tremendous upside. Not just for ChatGPT, but for you.

And as we’ve seen before, the brands and businesses that use new platforms as a distribution channel to accelerate their direct relationship building and not as a substitute for it are the ones that will win.

If you can test this channel while it’s cheap, drive email capture, drive loyalty signups, build first-party data assets that compound — you are going to be in a really, really good place no matter what happens.

The brands that won’t fare as well are the ones who either wait entirely, and then get hit with higher costs with no first mover advantage, or the ones who treat AI advertising as a performance channel substitute, expecting direct response ROIs from what is, at least today, primarily a brand awareness play. Those folks are in trouble.

Playing the long game here doesn’t mean don’t use AI ads. It means advertise in a way that builds something you keep. Do that right, and you won’t find yourself caught behind the gatekeeper’s gate. You won’t find yourself caught in a shortcut trap. No matter what happens with AI ads in the longer term.

If this episode gave you a clearer picture of how you should approach AI ads for your business, do me a favor. Send it to a colleague who’s getting asked whether it’s time to consider AI ads. It might save them from going down the wrong path.

You can find the show notes for this episode and the full archive of past episodes at timpeter.com/podcasts.

And if you are ready to go deeper on making your brand the answer that AI reaches for, my book, "Digital Reset: Driving Marketing and Customer Acquisition Beyond Big Tech," is the roadmap. You’ll find the link in the show notes.

Thank you so much for listening today. I genuinely appreciate you. Until next time, please be well, be safe, and be excellent to each other. I’ll see you soon.

“Digital Reset with Tim Peter” helps you look beyond the "shiny objects" to build a business that lasts. How can we help you today?

The post The Gatekeeper’s New Tax: What ChatGPT Ads Mean for Your Marketing Budget (Digital Reset Episode 490) appeared first on Tim Peter & Associates.

I’ve got a big secret for you today: Brands winning in AI didn’t pivot to an AI-first strategy six months ago. Almost universally, they’ve been building direct customer relationships, earning independent reviews, and publishing content credible enough to be cited, usually for years.

City of Hope probably didn’t start with a GEO strategy or an AI optimization consultant. But they still appear in 97% of AI queries for their category. Why? Because they made decisions 10, 20, and 30 years ago that continue to pay off today.

AI inclusion, it turns out, is an inheritance. It’s something you build over time — and if you’ve been building the right things, the AI will find you.

That raises two obvious questions:

We’re going to look at the second question in detail in next week’s episode. Today, we’re diving deep into the first one.

And the answer to that first question is that companies often fall into a “shortcut trap.” Every new gatekeeper’s entry into the market comes with a period where taking the shortcut looks like the smart play.

The challenge for many businesses is that the shortcut isn’t a scam — it works… at least for a while. And that’s what makes it dangerous. By the time its true costs becomes visible, too many businesses have built far too much of their strategy around it. They own visibility but not the customer relationship.

This episode traces 15 years of how that pattern has repeated across a variety of platform shifts — Google, social, OTAs, and now AI. It also outlines two clear tests you can use to separate a genuine foundation investment from a shortcut dressing up as strategy.

If you’re the one who has to explain your AI strategy at your next budget meeting, this episode highlights the pattern and the language you need to make the case.

Key Insights for Strategic Leaders to Close the Gap

Whether you’re in hospitality, retail, or B2B — and especially if you’re the person who has to answer "what’s our AI strategy?" while watching the platform landscape shift under your feet — this episode gives you 15 years of pattern recognition to work with.

Tim Peter has written a new book called Digital Reset: Driving Marketing Beyond Big Tech. You can learn more about it here on the site. Or buy your copy on Amazon.com today.

Rutgers Business School MSDM Speaker: Series: a Conversation with Tim Peter, Author of "Digital Reset"

We have some free downloads for you to help you navigate the current situation, which you can find right here:

Contact information for the podcast: [email protected]

Recorded using a Shure SM7B Vocal Dynamic Microphone and a Focusrite Scarlett 4i4 (3rd Gen) USB Audio Interface.

Running time: 24:25

You can subscribe to Thinks Out Loud in iTunes, the Google Play Store, via our dedicated podcast RSS feed (or sign up for our free newsletter). You can also download/listen to the podcast here on Thinks using the player at the top of this page.

Welcome back to the show. The brands winning in AI right now didn’t pivot to an AI first strategy six months ago. Almost universally, they’ve been building direct customer relationships, earning independent reviews, and publishing content credible enough to be cited for years.

The AI didn’t teach them anything new. It just made visible who has done the work and who has been renting their results from the nearest gatekeeper. I’ve been watching this pattern play out for 15 years through Google algorithm updates, organic reach collapse on social media like Facebook and LinkedIn, increased commissions and take rates from intermediaries like online travel agencies and Amazon, and now the first wave of AI driven discovery.

Each time the businesses that navigated the shift with the least damage had the same things in common: a direct relationship with their customers, an earned reputation that didn’t depend on any single platform, and data that they owned, literally owned.

Here’s the thing. That framework isn’t complicated. Build credible content. Collect independent reviews. Earn the right to show up in your customer’s inbox. Make your brand signal clear.

Most people listening to this show already know that. The question this episode answers is, if the formula is this well understood, why are 82% of companies still stuck in the AI value gap? Why don’t they just do this?

The answer has everything to do with a trap that every new gatekeeper sets and that smart, experienced marketing leaders fall into anyway. I’ve watched it happen through four different platform shifts. I’d like to make sure it doesn’t happen to you on this one.

This is episode 489 of Digital Reset. I’m Tim Peter. Let’s dive in.

A couple of weeks ago, you heard me talk about City of Hope and the fact that they show up in 97% of AI queries for their specific category. Well, City of Hope did not have a GEO strategy. They didn’t hire an AI optimization consultant.

They showed up in 97% of AI queries for their category because of decisions they made 10 years ago, 20 years ago, 30 years ago or more. That should tell us something about what AI actually rewards. AI inclusion is an inheritance. It’s not an acquisition.

City of Hope inherited their position from years of peer reviewed research, independent media, and an earned reputation with their patients. The AI was trained on all of that.

Most GEO strategy, quote unquote, is sold as something you can acquire this quarter. But if AI inclusion is primarily inherited from prior fundamentals, that changes and should change your budget discussion, your budget conversation.

AIs see a weak signal, a contradictory signal, or no signal, and it loses confidence in your brand. When an AI sees a strong signal, a clear signal, a coherent signal, that’s when you win. The brands who are showing up consistently today are showing up because they’ve built brands worth people asking for by name. This is what I mean when I say the brand is the prompt.

But also brands that are worth answering by name, that the AI can confidently say, "As a concierge, I know the answer to your question. I know who you should be talking to."

Winning in the long run isn’t just about what you do this quarter. It’s what you do for a long time. If you go to your AI of choice — it doesn’t matter if it’s Google Gemini, it doesn’t matter if it’s ChatGPT, it doesn’t matter if it’s Perplexity, it doesn’t matter if it’s Claude — go to the AI you like the most and ask it to describe your brand.

Everything it gets right is a sign that you have a strong signal. Everything that it gets wrong or hedges on or isn’t quite clear on, that’s where you have a gap. And that’s your roadmap. You don’t need a vendor to do an audit. The AI itself is going to tell you this is what it knows to be true about you. That’s really, really key.

Now the most common gap is when you say one thing about your brand and your customers are saying something else about your brand. It sees a difference between your statements and your customer’s reviews. That’s a huge contradiction, and that means the AI will lose confidence in you. It cannot confidently recommend you to a potential customer.

Note that this isn’t just about the discussions happening on platforms that the AI trusts. It’s that you have never done the work to build review velocity for your business, that you haven’t worked to gain the earned media presence that gives the AI some corroborating evidence beyond just what you say on your site or beyond just what it sees in reviews.

Neither of those gaps is going to get solved by a GEO vendor — immediately. Both are solved by doing the same things that improve your direct business: giving a better customer experience, gaining better reviews, and building clearer signal.

I’ve worked with businesses to reduce their dependency on big tech companies over the last 15 years — through Google updates, through the emergence of OTAs, through social reach, and now AI. And what’s interesting is how this era looks similar to what I’ve seen before.

In the book Digital Reset, I talk about a pattern that happens: a new discovery channel emerges. Something comes around and we go, "oh, this is cool, we should check this out." We get good results from it early. We test it and we see that this is really working, and usually at a pretty low cost.

Over time, the platforms with legs grow more dominant. They build a bigger base of customers and often send more customers your way, usually at a pretty low cost. That’s super attractive. So you double down on that. You dive in even further until suddenly that becomes a major source of your business.

But at that point, that puts them in a position of gatekeeping power. And as you’ve heard me say many times before, gatekeepers gonna gate. They have to. They are required to, because they owe it to their shareholders to monetize the traffic and the connection that they have with customers to grow their revenues and grow their profits and grow their shareholder value. And so what happens is the gatekeeper then raises the toll to you.

This is a vicious cycle that occurs again and again and again. We’ve seen this repeatedly with search, with social, with mobile, with OTAs. It happens consistently.

Every time there’s a new platform shift, there’s a window. It could be two years, it could be three years, it could be five years, where the new gatekeepers are still building their position and they haven’t yet started collecting the highest tolls they can.

That’s huge because marketing leaders look at that and say, this is a great opportunity for our business. And that’s good. That actually is a good idea.

We saw this, I’ll give you a real world example, with independent hotels and hotel brands. They had opportunities to build direct booking capabilities and direct booking connections with customers before OTAs became kind of non-negotiable. First before September of 2001. Then in the later 2000s.

The ones that took advantage of those windows built direct websites. They built email lists. They built loyalty programs — either recognition programs or reward programs — to connect directly with customers and gain data. And they built direct revenue.

The ones that didn’t do those things are still paying 20% commissions as table stakes for 35, 40, 50% of their business. That’s not a great place to be. The boutique hotels that appear in ChatGPT answers today for questions like "the best independent hotel in Charleston," or "the best independent hotel in Orlando," are there because of the reviews they’ve been gaining for the last 10 years, for the content that they’ve been publishing for the last eight to 10 years, for press coverage that they earned five or six or seven years ago, and every year in between. They’re not there because of some AI strategy. They inherited that position because they had an overarching brand strategy and an overarching strategy of how to connect with their customers directly every single time.

The hotels that went all in on OTA distribution, the retailers that outsourced their audience to Facebook, the publishers who handed their traffic to Google — each of those folks made, I want to be fair, a rational decision in that moment. It wasn’t a mistake in the small terms. The problem was that they didn’t own anything when the platform changed the terms of the relationship.

So the question that I would ask is whether we are seeing a different game or whether it’s the same game with slightly different rules. Spoiler alert, I think it’s the same game with slightly different rules.

There are definitely new things happening here. Artificial intelligence, AI answer engines, and AI assistants and AI agents as they arrive, are weighing things like corroboration quality — is your story being backed up in other places — more than just "did you get a bunch of links." It’s much harder to game that with volume or technical tricks.

In that sense, AI is doing what search has always supposed to be doing before they had to spend so much time fighting black hat SEO types and people trying to game the system. Candidly, one reason why I’m so bullish on Google is that they know what getting gamed looks like. I’ve argued that many of the AI answer engines right now are doing a speed run through Google’s search spam learnings, and that they’re going to have to make changes that Google’s been making for years.

What hasn’t changed is expert authored content. You’re familiar with EEAT probably, from Google — Expertise, Experience, Authority, Trust. Companies and businesses that have built that expertise and have built that authority and have gained that experience and have gained that trust are the ones doing well.

What also hasn’t changed is independent validation and trusted platform reviews and a strong, well-regarded direct brand that follows from those. Those drove organic authority 10 years ago, and they’re driving AI inclusion today.

Maybe you’ve heard me say before that content is king, customer experience is queen, and data is the crown jewels. That’s what we’re still seeing even with artificial intelligence.

One thing to keep in mind here is a "do no harm" principle, a "first, do no harm" principle. If a GEO tactic is going to hurt your search performance, it’s probably not going to work for artificial intelligence either — and certainly not in the long term. The tactics that work for both are the same: quality content, earned external references, active review management.

The brands that win in AI today are the ones that have inherited that position. The question you should be asking right now is not "how do we acquire AI visibility?" It’s "how do we build the brand that produces AI visibility as a byproduct of being a brand that AI values?"

Now, if the fundamentals are this clear and the pattern is this consistent, why are 82% of companies still stuck in the value gap? Why don’t they just do the thing?

Every new gatekeeper’s entry into the market included — and includes — a period where taking the shortcut looks like the smart play. In Google’s case, it was things like link building programs even before you had to pay for them. In social media’s case, it was building organic reach and building your follower count. In OTAs, it was things like low cost early commissions. And with AI it’s things like GEO vendors and AI content farms and churning out high volumes of low quality, low cost content so that you show up.

But those all stop working at a certain point. They realize people are spamming this. So Google shuts that down. The social media channels say we need to actually earn money off of these folks, so we’re going to pull back the algorithm and change what your organic reach is. The OTAs are saying, we’re contributing a lot of business to your hotel, so we need to raise commission rates.

What makes you think that the AIs are going to do anything different? We don’t know exactly when that will occur, but I’m really confident it’s coming because we’ve seen this happen again and again and again.

Now, I want to be very fair to people who have gone down this path before and chosen that path. It’s not a scam. It’s a trap.

It’s not a scam because it works — at least temporarily. That’s what makes it so dangerous. A scam would be easier to resist. You’re savvy enough — the people who’ve done this are savvy enough — to know that if you’re not getting value out of something, you would never put your efforts there, you would never put your money there.

So the people who have taken these approaches are not fools. What they are doing is saying, "Hey, this is producing results for me. I should double down on this."

That’s what makes it dangerous — the fact that it actually does work. The challenge is that you end up owning visibility, but not the relationship with the customer.

The shortcut is always attractive exactly at the moment when you need it the most, because they’re helping you reach customers you haven’t reached before, usually at a relatively low cost. All that you see at that point is the upside. By the time the cost becomes visible, too many people find themselves in a position where they’ve built far too much of their strategy depending upon that thing. That’s not a great place to be.

Marketing leaders need to think in terms of: when is this a genuine investment in our foundations, and when is this a shortcut dressed up as a strategy?

I would think there are a couple of ways you would test this.

The first is to ask whether this investment would matter if AI changed tomorrow. If we look at things like expert authored content or review velocity programs or first party data infrastructure or earned media from credible sources, those are going to improve your business regardless of which AI model is dominant six months from now, 12 months from now, 18 months from now. If the investment only works because of how ChatGPT works in Q2 of 2026, that’s probably a warning sign.

The other test is: does this investment compound its value over time, or does it require a reset every couple of months?

Shortcuts work. Foundations compound. A review earned today is in the training data for the next model update. Content that gets cited once tends to get cited again. First party data gets more valuable as you collect more of it from your customers.

If an investment’s value has to be reinvested every time the platform updates, that’s a shortcut. If it compounds regardless of the platform updates, then you’re building a foundation for success long term.

When you walk into your monthly review or your quarterly review and you’re making the case for the budget you need going forward, you should not be thinking in terms of "we shouldn’t invest in AI." That’s not what I’m saying here.

You should be saying: we should invest in AI the way businesses that survive every platform shift invest. We should be investing in the things that improve our business and compound across every platform, not only in things that work for this particular algorithm or this particular artificial intelligence. One of those has a long-term opportunity for you. One of those means you’re going to keep throwing money after money after money every time the algorithm changes.

When we think about budget categories: expert authored content that earns citations — a hundred percent. Review velocity programs — a hundred percent. First-party data infrastructure — a hundred percent. If we’re talking about people selling you, "you’re going to appear in the AI top every single time" — you might want to take a really close look at that. You might want to start small and test, because maybe they do know something. But you want to make sure you own the result, not just the visibility, before you double down there, before you try to scale this up.

The businesses that I have watched navigate every platform shift without getting captured have all had one thing in common. It’s not that they saw the future first. It’s not that they were smarter than everybody else. It’s just that they never fully gave up the direct relationship with the customer.

They built a long lasting brand platform, a long lasting customer relationship that survived and thrived every time the platforms shifted. They didn’t chase any short term wins at the expense of the long term opportunity.

I have had this exact conversation through the Google updates, through the collapse of reach on social media, with hotels and OTA commissions, and now with AI. Some of this isn’t that I’m predicting the future. I’ve seen this and seen folks get burned by it plenty in the past, including me from time to time.

This is hard won experience. You eventually learn, “Hey, maybe we shouldn’t chase the ‘ooh, shiny object,’ but we should build something of lasting value." The folks I’ve worked with who’ve acted on these conversations and learned from them and applied them — and the folks who aren’t even clients who figured it out on their own — they’re the ones who are doing great and they’re the ones who have options to continue to improve over the long run.

This is not some sophisticated, brand new AI strategy. It’s a 15 year pattern that keeps working no matter who the gatekeeper is next. And ultimately, that’s the place where you want to be.

If this episode helped you think more clearly about where to put your budget this year, or gave you language for the conversation you’re about to have with your CMO or CFO, do me a favor — send this to a colleague who’s wrestling with the same question. That’s how these conversations get into the rooms where the decisions are made.

You can find the show notes for this episode, including the research we discussed and the full archive of past episodes at timpeter.com/podcasts.

And if you’re ready to go deeper on building a brand for the next platform, whatever it turns out to be, my book, "Digital Reset: Driving Marketing and Customer Acquisition Beyond Big Tech," offers the full framework. You’ll find the link in the show notes.

Thank you so much for listening today and for all of the episodes you’ve listened to. I genuinely appreciate you. Until next time, please be well, be safe and be excellent to each other. I’ll see you soon.

“Digital Reset with Tim Peter” helps you look beyond the "shiny objects" to build a business that lasts. How can we help you today?

The post The Long Game: What 15 Years of Digital Marketing Teaches Us About AI (Digital Reset Episode 489) appeared first on Tim Peter & Associates.

Over the last three weeks on this show, our fearless leader, Tim Peter, covered three big ideas. These are:

Those are not three separate problems. They’re three symptoms of the same underlying condition: a weak brand signal. The brands that show up consistently — City of Hope appearing in 69 of 71 AI responses, not 2 of 71 — have built something the machine can’t easily ignore.

The question this episode answers is how they’ve successfully done this.

When this episode first aired, Tim called it the "first, do no harm" framework: the bridge between traditional SEO and the emerging world of generative engine optimization. He introduced the shift from a world of card catalogs to a world of concierges. He laid out why content is king, customer experience is queen, and data is the crown jewels also works as an operating model to drive prompt brand equity for your business. And it’s a framework that has been validated in every episode that followed.

One thing has changed since the original recording: the SparkToro research Tim mentioned at the time has since been published in full, covering 2,961 prompts by 600 volunteers across nearly two months of runs using ChatGPT, Claude, and Google. The numbers confirmed everything the original episode predicted… and then some. If you are staring at your 2026 budget wondering where to place your bets, this is the blueprint.

Key Insights for Strategic Marketing Leaders

Whether you’re in hospitality, retail, or B2B &mdash and especially if the last three episodes left you with a framework but not the foundation — this episode makes everything click.

Want to learn more? Here are the show note for you.

Tim Peter has written a new book called Digital Reset: Driving Marketing Beyond Big Tech. You can learn more about it here on the site. Or buy your copy on Amazon.com today.

Rutgers Business School MSDM Speaker: Series: a Conversation with Tim Peter, Author of "Digital Reset"

We have some free downloads for you to help you navigate the current situation, which you can find right here:

Contact information for the podcast: [email protected]

Recorded using a Shure SM7B Vocal Dynamic Microphone and a Focusrite Scarlett 4i4 (3rd Gen) USB Audio Interface.

Running time: 18m 08s

You can subscribe to Thinks Out Loud in iTunes, the Google Play Store, via our dedicated podcast RSS feed (or sign up for our free newsletter). You can also download/listen to the podcast here on Thinks using the player at the top of this page.

Welcome back to the show. Over the last few weeks, we’ve taken a deep dive into some of the uncomfortable realities about the state of AI. I’ve looked at why 88% of companies are playing with these tools while only 6% are actually seeing significant value. I call that an AI tax, the time and money spent fixing machine mistakes and chasing efficiency at the cost of customer trust.

Last week, we saw how deep that rabbit hole goes. And I discussed how AI recommendations change more than 99% of the time, proving that if you’re chasing a specific ranking in ChatGPT or Claude, you’re basically bidding your strategy on a coin flip… on a series of coin flips. And before that, we saw OpenAI pull back on agentic commerce features because operational complexity and customer behavior don’t always align with the tech press hype.

These are not three separate problems. They’re symptoms of a weak brand signal.

As we’ve seen, the brands that show up consistently, like City of Hope showing up 97% of the time, do so because they’ve built a foundation of corroborating stories, of digital witnesses, that the machines just can’t ignore.

So how do you actually build that foundation for your business? How do you move from chasing the shiny object to investing for optionality? Today, I want to revisit a foundational episode that maps exactly how you show up when the world moves from card catalogs to concierges. This is the "first, do no harm" framework that bridges the gap between traditional SEO and the new world of GEO.

If you are staring at your 2026 budget, wondering where you can place your bets most successfully, this episode is the blueprint for you.

This is Digital Reset with Tim Peter. I’m Tim Peter. Let’s dive in.

So I think we all agree it’s unbelievably important that you show up in AI answer engines. What I’m not sure we completely agree upon is the fact that you need to show up in SEO still too, in traditional search.

I want to give full credit to SEO expert Lily Ray for this idea, but we want to think about the fact that first, “do no harm.” Forget for a moment how you show up in AI answer engines. Of course you want to do that.

I also suspect, and see this with my own clients, that for most businesses, even in a world of zero-click searches, Google still represents a significant, and probably your largest, single source of traffic.

So the first thing you want is don’t do anything that screws up what Google gives you. This isn’t an SEO versus GEO question. It’s SEO plus GEO. Remember, you don’t get the plus if you mess up one side of the equation or other.

As you read GEO advice, as you listen to GEO experts, as you listen to people talk about this, you need to ask yourself, “does this have the potential to hurt our current organic situation?” If the answer is no, go ahead and do it. Do it immediately even.

If the answer is yes, well, don’t ignore it outright.

Instead, then ask, “okay, how do we gain the benefits of this recommendation while also keeping our existing SEO position — or better yet, improving that existing SEO position?”

Again, it’s not either SEO or GEO. It’s SEO and GEO, SEO plus GEO.

Second, and you’ve heard me say this many times that customer experience is queen. I’m kind of bundling the first two portions of our royal court of brand building here. You’ve heard me say many times in the book, Digital Reset, that content is king and customer experience is queen.

Those are still true. What’s also true, is that your content has always included what others say about you. You have a secret sales force working for you — or for your competitors — based on what your customers’ experiences are and what they say about those experiences in social and in ratings and reviews. They’re telling their friends and family and fans and followers all about what they think about your products and services and your brand more broadly.

If you’re in a category like hospitality or travel or any kind of B2C, what others say about you in public ratings and reviews plays a massive role here. You’re undoubtedly familiar.

In B2B, the same is still true. It just tends to happen in private. It could be text or LinkedIn messages. It could be email. It could be on Slack or Teams channels or phone calls. But folks ask other folks, “Hey, what do you know about this company? What do you know about their products and services? Do they work well for you?”

Your content has always included what others say about your brand and business. Well, the data seems pretty clear that those discussions influence whether or not you show up in AI responses. That’s something we just have to keep top of mind.

You also want to keep do no harm in mind, too, because what people say about you, whether it’s public or private, drives awareness, interest and action among your potential customers.

Ideally, that’s to your business’s benefit. But even a poor customer experience creates a narrative around your brand and one that you’ll fight against for a long time to come, especially in GEO. If “the concierge” is reading all kinds of bad things about your brand, guess what it’s going to know about your brand?

We want to move to an era where your brand is the prompt. And if somebody types in your brand and the answer engine says, “Man, I don’t know about that brand…” that’s really, really, really bad for your business. Not just now, but over the long term.

Making sure that when your brand is the prompt that the concierge has great things to say about you, about your brand, about your business, is a clear component of why customer experience is queen now more than ever. So seriously, take care of the queen. This really matters.

Finally, if you had have some exposure to our Royal Court framework that I’ve talked about many times, you’ll know that I say data is the crown jewels. And I want to talk for a minute about what that means in this case.

I’m seeing many businesses experience a big lift in branded traffic or conversions that don’t have an obvious source. Folks are just coming to your website or opening your app or reaching out to your sales team or sending you an email without starting on your site. They just kind of appear. And if they’re not doing that, well, you want them to, regardless of whether they’re doing that today or not. Listen to my prior points from a moment ago if you need more help on how you can do that.

The question we want to be thinking about then is if people just show up, what metrics matter to your business? We can’t rely on the metrics we’ve used for a long time. Now is the time to get your KPIs and your metrics in order so that you can actually determine what is our current state and are we moving forward in a direction that we want to.

Again, data is the crown jewels. They help you make better business decisions and take better care of your customers so that they actually have a better experience, so that you actually have a better shot of showing up more often.

Clearly, we’re going to lose some visibility into what’s happening compared to what we’ve had for years. You might not get the same level of tracking or the same level of insights you’ve had in the past. That’s okay. We just need to keep track of the metrics that genuinely matter and help you make better business decisions.

Now the first one should be obvious, it’s revenue, right? Are you seeing more revenue or less revenue? Ultimately, that is the final determinant of whether or not any of this is working for your business. If your revenue is going up, that’s probably a good sign that you’re doing the right thing. If your revenue is declining, that’s probably a clear sign that you need to do something different than what you’ve been doing up till now. So that’s a huge one you want to pay attention to.

Of course revenue is a trailing indicator. It only tells you what happened in the past. It’s not necessarily all that predictive of what’s going to happen in the future.

So another key performance indicator you want to watch is lead volume. That could include actual leads in a B2B context. It could also include the number of people who put items in their shopping carts in B2C or enter your booking engine in a hospitality context. Are people showing interest in engaging with your brand and your business in a commercial way?

Do they want to talk to you? Do they want to hear from you? Pay super close attention to whether they are signing up for your email lists. Are they giving you their information for SMS? Are they saying, “I want to have a conversation with you, not just now, but in the future and want to hear from you directly?” Again, that’s a super, super huge sign about where your business is going to go in the future.

We do a bunch of work with hospitality clients and we’ve determined that for your typical set of hotels, every new email address you collect is worth somewhere between five and thirty-five dollars in future revenue. So if that data isn’t showing up, if you weren’t collecting that data from your customers, it’s a pretty good sign that your future revenues are going to have some real problems.

Again, data is the crown jewels here. It points to real future revenue.

Another metric that matters a ton are brand searches. Are the number of searches for your brand in organic traffic shrinking or growing?

If it’s shrinking, you’re likely not showing up in AI or traditional search near as much as you’d like to or need to. People aren’t finding you. They then aren’t then going to Google and ask for you. Again, review my prior points for what you can do about that.

If you see those numbers rising though, it’s a pretty good sign that AI is likely surfacing you more often. AI is talking about you and people are learning about you to then come over to your website by searching you by name. That’s a really clear use case that we’ve seen customers do where they learn about you and then they go ask traditional search to help them find you and find your website. Hugely important.

A more recent and more sophisticated way to watch this is something called prompt brand equity. This is a term that we’ve talked about a bunch; it’s something that I think matters a lot as we go forward. And of course, I’ll link to in the show notes other times we’ve talked about this.

Now, prompt brand equity is not about where you rank in AI search results. It’s how often you show up at all.

There’s some great research from Rand Fishkin that shows where you appear in AI answers is not at all predictable. You could be number one in one chat, number two in another chat, and number three in a third chat, even though you’re an equally good answer for all three chats, even if the prompt was precisely the same. It’s not predictable. And it’s not predictive of what’s going to happen with your business.

The thing you want to keep in mind here is it’s not about where you rank. It’s does your brand appear at all? Do you ever show up? Don’t worry about tracking where you rank, but absolutely pay attention to whether or not you appear.

There are a bunch of pretty good tools for measuring this metric right at the moment. Tools like Peec AI, seoClarity, FinSEO, SE Ranking, Profound, and Otterly all seem to get good reviews and good feedback for how well they measure the frequency of how often your brand appears. It’s probably worth checking out each of those to see what might work best for you.

My point here isn’t to recommend one tool or another. Your specific needs may lean you towards one versus the other. My point is to strongly recommend that you keep track of whether or not your brand appears at all. Otherwise, you have no way of knowing whether in fact your brand is the prompt… or isn’t.

Tracking how frequently your brand appears is a great way to learn whether or not your content marketing and GEO efforts work. If you’re showing up, that’s a good signpost that your content is being used by AI answer engines. And if not, well, then your content pretty likely isn’t doing its job.

It’s critical to understand what’s working here and what isn’t. As I mentioned last week, we’ve moved from a world where you just needed to show up in a card catalog to a world where you need to convince a concierge that you’re the right choice. That’s why taking care of both traditional search and GEO matters.

Your successful digital reset isn’t about fighting with machines. It’s about ensuring your brand signal is clear enough for those machines to find you. Because in this new era, your brand isn’t just a logo or a label… not that it ever was. It’s that your brand is the prompt — or it needs to be if customers will consistently find you no matter where they go looking.

If this episode helped you make sense of some of the noise around SEO and GEO and about digital more broadly, do me a favor, send it to one of your colleagues who is currently staring at their 2026 budget and wondering what they should do next.

You can find the links to the GEO tracking tools we discussed, the show notes for this episode, and the full archive of all past episodes at TimPeter.com/podcast.

And if you’re ready to move beyond big tech for your business, my book, Digital Reset, Driving Marketing and Customer Acquisition Beyond Big Tech, provides the roadmap to reclaim your demand. You’ll find a link in the show notes.

Thank you so much for listening today. I genuinely appreciate you. Until next time, please be well, be safe, and be excellent to each other. I’ll see you soon.

“Digital Reset with Tim Peter” helps you look beyond the "shiny objects" to build a business that lasts. How can we help you today?

The post The Foundation: From Card Catalogs to Concierges — Your SEO + GEO Blueprint (Digital Reset Podcast) appeared first on Tim Peter & Associates.

Rand Fishkin’s team ran 2,961 prompts across ChatGPT, Claude, and Google AI. 600 volunteers, 12 different prompts, two months of runs. They wanted to answer one question: how often do you see the same list of brand recommendations twice, even with the exact same prompt?

The answer? Less than 1% of the time. The odds of seeing the same list in the same order are closer to one in a thousand.

Most conversations about AI inconsistency treat it as a measurement problem: how do I know if my brand is showing up? That’s a legitimate question. But it’s not the only question. And it might not even be the most important one.

If AI systems give different recommendations essentially every time, the same inconsistency is already baked into every AI chatbot you’ve deployed — your hotel chat widget, your B2B sales assistant, your customer service tool. Most teams have never measured it. And some of those inconsistent answers are already driving negative reviews for your brand and business.

This episode connects three stories Tim has covered over the last three weeks — the AI value gap, the uncertain timeline of agentic commerce, and now AI inconsistency — showing that they all stem from the same underlying condition. It also explains what City of Hope, appearing in 69 of 71 AI responses for "West Coast cancer care hospitals," tells us about how you can fix this problem for your business.

Key Insights for Strategic Leaders to Close the Gap

Whether you’re in hospitality, retail, or B2B, this episode is for anyone who’s deploying AI in a customer-facing role… or who’s who’s being asked to report on AI visibility and wants a better sense of what they’re actually measuring.

Tim Peter has written a new book called Digital Reset: Driving Marketing Beyond Big Tech. You can learn more about it here on the site. Or buy your copy on Amazon.com today.

Rutgers Business School MSDM Speaker: Series: a Conversation with Tim Peter, Author of "Digital Reset"

We have some free downloads for you to help you navigate the current situation, which you can find right here:

Contact information for the podcast: [email protected]

Recorded using a Shure SM7B Vocal Dynamic Microphone and a Focusrite Scarlett 4i4 (3rd Gen) USB Audio Interface.

Running time: 22m 01s

You can subscribe to Thinks Out Loud in iTunes, the Google Play Store, via our dedicated podcast RSS feed (or sign up for our free newsletter). You can also download/listen to the podcast here on Thinks using the player at the top of this page.

Welcome back to the show. I’m Tim Peter.

I’ve talked about a concept called "Prompt Brand Equity" for a while now, the idea that what matters in AI search isn’t where your brand ranks. It’s whether you show up at all, whether your brand shows up at all. And I mentioned some early research from Rand Fishkin at SparkToro, shows that AI recommendation lists were unpredictable. You could be number one in one chat and number three in the next, even with the exact same prompt, by the exact same person.

Well, the full research is out now, and the numbers are much more striking than I expected. Rand’s team ran 2,961 prompts through ChatGPT, Claude and Google AI with 600 volunteers over two months. The question they were trying to answer, how often do you see the same list of brand recommendations twice, even if you run the exact same prompt over and over?

The answer? Less than 1% of the time. I want to say that again. Less than 1% of the time do you see the same list twice. In other words, practically never. That has real consequences for how you measure your business’s AI visibility and for how you think about the AI tools that you’ve already deployed in your own business.

And for the ROI gap that I covered on episode 486, which if you missed it, was about why 88% of companies are using AI, but only 6% are seeing significant value from it. It turns out that these trends, these traits, these facts are connected. Today I want to get into how.

This is episode 488 of Digital Reset with Tim Peter. I’m Tim Peter. Let’s dive in.

Okay. Let me start with what the research actually found, because the headline number undersells it a little. Rand Fishkin partnered with Patrick O’Donnell at a company called Gumshoe.ai. They recruited 600 volunteers to run 12 different prompts, things like "recommend headphones under $300," or "what are the best project management tools."

They ran these through ChatGPT, Claude, and Google Gemini, Google AI, over and over for two months, nearly 3,000 runs in total. And what they found is this: the list of brands recommended changes more than 99% of the time. The odds of seeing the same list in the same order twice are closer to one in a thousand.

That’s nuts, right? So I wanna be fair about what this means and what it doesn’t mean. It doesn’t mean that AI is useless. It doesn’t mean that brand mentions in AI are random and it definitely doesn’t mean you should give up on showing up in AI answers. Far from it.

What it means is that where you appear in any given AI response, whether you’re number one or number three, tells you essentially nothing. That position is random. It’s not predictive of anything to you or to your business.

The useful metric isn’t rank. It isn’t where you show up. It’s frequency. How often does your brand appear at all, across a large sample of runs on the questions that matter to your customers. That number tells you something real.

That number is, of course, prompt brand equity, not position, frequency.

I mentioned Rand’s early work on this in episode 485 when I talked about how we’ve moved from a world of card catalogs to a world of concierges. The new data just puts specific numbers into what we already expected. The picture is much clearer now, and if I’m being really honest, a little more dramatic than I expected.

Now, here’s what I think is the most under-reported part of the story, and it matters a lot if you are in hospitality or honestly, if you’re in any business that has deployed AI in a customer facing role.

When people talk about AI inconsistency, they almost always frame it as a marketing measurement problem. You know, how do I know if my brand is showing up? And it’s a legitimate question. I’ll get back to that in just a moment, but it’s not the only problem here.

The second problem is operational, and it’s happening right now in your business.

If AI systems give different recommendations essentially every time to customers asking the same question, what do you think is happening when your AI chatbot answers the same question from two different customers? Think about that for a moment.

Let me give you a hospitality example. Let’s say you’ve got a guest who opens your hotel’s chat widget and asks, what’s the cancellation policy for a reservation that’s made between this date and this date? And your AI gives them an answer. An hour later, a different customer asks the exact same question because they’re staying during a high demand period.

Your AI might give a slightly different answer. Maybe the window is 48 hours in one response and 24 in another. Maybe it says that there’s a charge for one and not for the other. Maybe amenities that are included get described differently. Maybe the rate quote shifts. Is that happening? Uh, almost certainly.

Has anyone on your team actually measured it with any consistency? Almost certainly not.

By the way, this is not just a hospitality problem, it’s a B2B problem too. It’s a retail problem too. If you are using an AI tool to help your sales team respond to prospect questions or using AI to handle initial customer service inquiries, the same inconsistency is baked into those responses. Your AI isn’t giving a consistent answer. It’s giving a distribution of answers, and your customers are sampling from that distribution. That’s an operational liability, not just a measurement inconvenience.

In hospitality, especially where the expectation of guest forms before arrival is a huge driver of satisfaction. Or dissatisfaction and where dissatisfied guests write reviews that the AI then reads. And I don’t mean the AI on your website. I mean ChatGPT or Claude or Google Gemini or Google AI Overviews or AI Mode. This is a problem that you need to think carefully about.

So if AI recommendations are this inconsistent, why is it that some brands do really well and others don’t? Why do some brands appear in almost every other response while others don’t appear in very many at all? And there’s a phenomenal example, just an amazing example from the research that I want to share.

Rand’s team asked Google, specifically to recommend "West Coast cancer care hospitals" 71 times, and City of Hope appeared in 69 of those responses. That’s 97% of the responses.

That’s not luck, that’s not some glitch in the algorithm. City of Hope appears in 97% of those responses because the AI’s confidence weight for City of Hope on the topic of "West Coast cancer care" is so high that it essentially has to include them. There’s too much validating information from too many credible independent sources pointing in the same direction for the AI to leave them out with any reasonable probability.

I want to be fair here for a moment. There is a reality in Google searches where they refer to "your money or your life searches," things where it can cost somebody significant money or it can interfere with their health outcomes.

Google is very careful in YMYL searches, "your money or your life" searches, not to include any business that could reasonably harm your money or your life.

So it is possible, I want to be very clear, it’s very possible that this specific query ends up being a YMYL query, or because it’s a YMYL query, Google’s going to be particularly careful not to recommend a brand that it’s not highly confident about.

It still underscores the point that the concierge wants to give the best possible answer, not just what gets the most links, not just what ranks number one, but what it knows is the correct answer. And what City of Hope is seeing is what high prompt brand equity looks like when it’s working.

The inverse is also true. Brands that show up 5% or 10% of the time aren’t being penalized. They’re simply not validated enough. They’re simply not enough corroborating information about them for the AI to be willing to commit to the brand.

Now, when I say validated or corroborated, I want to be specific about what that means. Because "post more content" isn’t the answer here. That’s not how you’re going to win.

What AI systems are doing is triangulating. They’re looking at a variety of sources and they’re asking "Do multiple independent sources, sources that have no obvious reason to be aligned, point to this brand and only this brand—or mostly this brand—as a credible answer to this question?"

Sure your own website says you’re doing great. That’s a data point, but it’s not necessarily a very high weight one. You are a biased source. What your customers say in reviews, that’s a higher weight signal. What industry publications and third parties say about you? Also a higher weight signal. What other expert sources cite you for, again, higher weight.

They’re witnesses to your competence and your capabilities and your credibility. Digital witnesses backing up and verifying what the concierge, the large language model, knows about you.

This. This is the reason that I say content is king, customer experience is queen, and data is the crown jewels. It’s not just a branding concept, it’s an operating model that drives prompt brand equity.

The content you publish establishes what you are about. It says, "this is who we believe we are." The experience you deliver generates the reviews and the word of mouth to validate that, that corroborates the other witness’s story. And the data, the first party signals that you own and the external signals that others generate about you is what the AI draws from to decide whether or not you are the brand it should include.

When those three areas, the content, your customer experience, and your data, are aligned and consistent, the AI’s confidence weight for your brand goes up. The concierge can recommend you confidently.

When those data points, when those three elements are contradictory or thin or absent, you’re the brand that shows up one time in 10.

Your brand is the prompt, but for your brand to actually be the prompt for the AI to reach for you by name the way it reaches for City of Hope, it needs enough signal that it has no other choice of who to select.

So practically, what do you do about this? I want to give you four things to think about this time around.

First shift your measurement from rank to frequency. Stop worrying about whether you’re number one or number two in an AI response. That number will change constantly and tells you nothing at all reliable.

What you want to track is if you run the 10 questions your customers most commonly ask across ChatGPT, Google, and Perplexity, let’s say for 10 runs each, how often do you appear? That’s your baseline frequency. And you can use tools like Peec.AI or seoClarity or others to actually measure that for you. But run it again 30 days from now, 60 days, 90 days from now. If your frequency is going up, then your signals are strengthening. If it’s declining, then something needs your attention.

I talked about many of these tools in episode 485, and I’ll link to them again in the show notes. But you can start measuring today for free with nothing but an hour and a few open browser tabs. If you’ve got people on your team, great. Set them to the same task. Take a few people, take 30 minutes and ask them to do it, and record those in a Google Sheet or an Excel spreadsheet so you’ve got something to refer back to.

Second audit your internal AI tools for consistency. While you’re worrying about showing up in external AI answers, go check what your own deployed AI is telling your customers.

In fact, if you already have an AI chatbot, I would do this first.

Take a look at the most frequently asked questions. Could be five, could be 10, but the ones that get asked regularly and the ones that your team already know the answer to cold and ask your chatbot each one 10 times. If the answers differ materially on anything that would affect a customer’s expectations, you’ve got an operational problem that’s creating review risk right now. And that’s actually more urgent than any GEO tactic because you’re creating a negative impression of your brand.

Third, focus your content investment on increasing witnesses, credible witnesses, not just content volume. One piece of original expert authored content that gets cited and shared and referenced by independent sources does more for your prompt brand equity than 50 AI generated blog posts.

The AI isn’t counting the number of pieces of content on your site. It’s weighting the content by the degree to which independent verifiable sources, credible sources, confirm what your content says. Write less. Make it more citable. Make it more original. Make it more authentic. Make it genuinely worth citing. That’s how you get cited more often.

Fourth, treat review velocity as a strategic input, not just a reputation metric. Review velocity was always an important metric. This was always important. But AI makes it more true.

Reviews are one of the highest weight external signals that AI systems draw upon, particularly if you’re in hospitality or a local business, or, uh, a customer service oriented business.

Recent, detailed, specific reviews on Google, TripAdvisor, Yelp, your OTA listings are among the most direct inputs to prompt brand equity available to small, independent businesses. Larger businesses, the same is gonna be true for something like G2 because that’s where your customers go to talk about you. Getting 10 substantive high quality reviews this quarter does more for your AI visibility than almost any technical GEO tactic. And responding to those reviews where the systems allow you, all of them, not just the good ones, tells the AI that you’re a business that takes the feedback seriously. That’s a signal too.

Okay, so those four tactics out of the way, here’s the bigger picture I want to leave you with.

Over the last three weeks on this show, I’ve talked about three different stories that all come from the same root cause. I talked about the AI value gap, why 88% of companies use AI, but only 6% see significant results. I talked about agentic commerce and why even OpenAI wasn’t reasonably able to predict how soon AI driven transactions would happen. And now I’m talking about AI inconsistency, why the same prompt produces a different answer almost every single time.

I’m going to say this as clearly as I can. Those are not three separate problems. They’re three symptoms of the same underlying condition. AI systems work by constructing a best guess answer from the available signals. The concierge wants to know. When the signals are weak or inconsistent or contradictory, its output is going to be weak and inconsistent and contradictory.

When the signals are strong, deep, and consistent, the AI finds you. It commits to you. It recommends you 69 times out of 71 as opposed to one out of a thousand. The companies in the 6% who are closing the value gap, the businesses that appear consistently in AI answers, they’re not doing something exotic. They’re doing the fundamentals very, very, very well.

They’ve got clear objectives. They’ve got strong content. They’ve got data they own and trust. And they have customer experiences that earn reviews that confirm what their content says.

That’s the digital reset in practice. It’s not fighting the AI. It’s not chasing every new platform or tool or protocol. It’s making your signal clear enough that the machine finds you, no matter what form that machine takes, whether it’s Big Tech or AI or something else that comes down the road. It all works together. Do that well, and you’re gonna find yourself in a very good position for a very long time.

Now, if this episode gave you a clearer picture of what prompt brand equity actually means and how you can build it, do me a favor. Send it to one colleague who is currently staring at your 2026 budget and wondering where they put their money, what we do next. It might send them from spending time or money on the wrong thing. Even better, send this to the person on your team who is currently reporting on AI rankings. Ask them "if this list changes 99% of the time, what are we actually measuring?"

I want to remind you that you can find the show notes for this episode, including links to the SparkToro research, the brand visibility tools that I mentioned, and the full archive of past episodes at timpeter.com/podcasts.

And if you’re ready to go deeper on making your brand the answer that AI reaches for my book, "Digital Reset: Driving Marketing and Customer Acquisition Beyond Big Tech," is the roadmap that you need. You’ll find the link in the show notes.

Thank you again so much for listening. I genuinely appreciate you. Until next time, please be well, be safe and be excellent to each other. I’ll see you soon.

“Digital Reset with Tim Peter” helps you look beyond the "shiny objects" to build a business that lasts. How can we help you today?

The post The AI Coin Flip: Why AI Gives Every Customer a Different Answer (Digital Reset Episode 488) appeared first on Tim Peter & Associates.

OpenAI launched in-chat checkout in September 2025, promising that users could shop and buy directly inside ChatGPT. Analysts forecast trillions in sales.

Then, just last week, the company pulled back: near-zero sales, a dozen merchants integrated out of millions, and a consumer base that was happy to research in ChatGPT but wanted to buy somewhere familiar. Expedia and Booking Holdings stocks rose on the news.

So what does this tell us about agentic commerce — the idea that AI agents will act as buyers, completing purchases on customers’ behalf?

It tells us that the direction of agentic commerce is real… and the timeline is not. The question then is what do you do when you can see where we’re headed, but not when we’ll get there. That’s what this episode of the podcast is all about.

Key Insights for Strategic Leaders to Close the Gap

Whether you’re in hospitality, retail, or enterprise marketing—this episode is for anyone being asked "what’s your agentic commerce strategy?" and isn’t sure what to say.

Tim Peter has written a new book called Digital Reset: Driving Marketing Beyond Big Tech. You can learn more about it here on the site. Or buy your copy on Amazon.com today.

Rutgers Business School MSDM Speaker: Series: a Conversation with Tim Peter, Author of "Digital Reset"

We have some free downloads for you to help you navigate the current situation, which you can find right here:

Contact information for the podcast: [email protected]

Recorded using a Shure SM7B Vocal Dynamic Microphone and a Focusrite Scarlett 4i4 (3rd Gen) USB Audio Interface.

Running time: 23m 55s

You can subscribe to Thinks Out Loud in iTunes, the Google Play Store, via our dedicated podcast RSS feed (or sign up for our free newsletter). You can also download/listen to the podcast here on Thinks using the player at the top of this page.

Welcome back to the show. I’m Tim Peter.

OpenAI launched a feature in September of 2025 called Instant Checkout. The idea behind it was really simple. You could browse for a product inside ChatGPT, and buy it right there without ever leaving the chat interface. The press loved it. The tech press just went nuts. Everybody on LinkedIn said, "Oh my gosh, this is the best thing ever." Analysts said that it would disrupt Google Shopping.  Gartner forecast that by 2028, 90% of B2B purchases will be handled by AI.

Last week… OpenAI pulled the plug. They’ve gotten near-zero sales from it. A dozen merchants integrated with it out of millions.

OpenAI hadn’t built the state sales tax infrastructure and it turned out that users, consumers, customers are happy to research inside ChatGPT. They just don’t wanna buy there yet. In the travel space, Expedia and Booking Holdings stocks went up on the news. The OTAs, the online travel agents that were supposed to be disrupted by this got a lifeline.

Here’s what I want to do today, though. I’m not going to tell you that agentic commerce isn’t coming. It is. The direction is clear. But instead, I wanna talk about what you do when you genuinely cannot predict the timeline. Because we can’t. OpenAI just proved that.

And the right answer to that uncertainty isn’t to freeze. It’s also not to build your strategy around the next big press release. It’s to do the things that make your business better today, and also happen to prepare you for any version of the agentic future when it actually arrives.

This is episode 487 of Digital Reset with Tim Peter. I’m Tim Peter. Let’s dive in.

Here’s what happened. In September of 2025, OpenAI announced that they were going to make it possible for people to be able to book reservations right in ChatGPT.

They said that people would be able to buy things from folks like Wayfair right in ChatGPT. They were working on a big integration with Shopify so that people could buy right within ChatGPT. And here we are six months later—this is early March, 2026—and they’re pulling back from this. They’ve announced "we are not gonna do this."

And there are three reasons why this happened. And let’s start with the most important one and always the most important one, which is customer behavior. Customers were browsing, they weren’t buying. People today use ChatGPT, the way they use Google. They research and they narrow down their options, and then they complete their purchase somewhere that they trust.

It’s not an implementation issue that they’re working through here. This is a behavioral reality.

The second big reason is operational complexity. Payments require tax infrastructure. They require fraud prevention, refund handling, and consumer protection and compliance. OpenAI had built, well, none of it yet. For travel specifically, there’s also all the liability issues, cancellations, disrupted trips, customer service complaints, et cetera. PhocusWire called it too complex for quick fix distribution. These aren’t bugs that OpenAI is going to patch right away. They’re fundamental hard problems of commerce that Amazon and Expedia and Booking and lots of other folks have spent the last 30 years solving.

In travel specifically, Amazon has attempted more than once to build a commerce platform only to back off because, well, it’s hard and it doesn’t really align with what they do well. Right? They’ve got free shipping, which doesn’t matter in travel. They’ve got, uh, warehouses, which don’t matter in travel, right? All of these things don’t play to their strengths, and at the moment they don’t play to ChatGPT’s.

The third problem, the third reason, is about merchant adoption. At least according to some research, maybe a couple of dozen Shopify merchants actually integrated. That’s out of millions of Shopify merchants. Real world e-commerce doesn’t plug in the way that the developer demo suggests today.

And I wanna be really clear, I’m talking about today. I do not want you to listen to this podcast, I don’t want you to listen to anything I’m saying here and think I’m saying that AI is overhyped or it, you know, this is never going to work. That is such an easy thing to say. And it’s wrong. It is wrong. It is not a thing I’m saying.

What’s really happening, if I can borrow the classic podcast phrase, is that two things are true at the same time. What happened here, what has happened here is that ChatGPT checkout failed in its first attempt. I mean, we’ve all been there, right?

But things like Google’s Universal Commerce Protocol and ChatGPTs Agentic Commerce Protocol are still moving forward.

Etsy and Wayfair are already transacting within Google AI Mode and in some cases in ChatGPT. Sabre, PayPal and MindTrip have an end to end agentic travel booking system that they anticipate is gonna happen in Q2 of 2026. By the way, Q2 of 2026—I’m recording this on March 11th—that’s, you know, weeks from now. So the direction is real. A company that controls payments, infrastructure. Ahem, Google with Google Pay, and Google Wallet, can do things that a company that doesn’t, OpenAI as of February, 2026/March, 2026, can’t do right now. Google’s integration with Siri and, as a potential consequence of that, their potential integration with Apple Wallet just underscores the opportunity that’s available to Google right now and others who do the work to make that happen.

The timeline is what’s uncertain here, not the destination. It’s pretty clear where we’re headed.

Last week I talked about the fact that Klarna made a very similar misstep. There’s a pattern we’ve seen before here. This is simply the Klarna moment for agentic commerce.

You may recall in last week’s podcast, I talked about the fact that Klarna replaced 700 customer service agents with AI in February of 2024. The press again was all over this that, "Oh my gosh, this is the moment that AI is gonna replace all the people." Well, fast forward to spring of 2025, and the CEO of Klarna said "we went too far."

The technology was real. The timeline for how long it took to deploy was wrong. The gap between the press release and the operational reality was enormous.

I mean, that sounds familiar, right? You’ve seen this before. Agentic commerce is just following that same path. You know, the tech press that it was here, people on LinkedIn went nuts. OpenAI launched the feature. The feature didn’t work the way it was supposed to, and so it’s time to recalibrate a bit.

Again, the lesson isn’t AI doesn’t work. The lesson as always is build your strategy around principles that hold, no matter which version of the technology wins.

Lois McMaster Bojo is a sci-fi author. She wrote a book years ago called "The Vor Game." And the hero of the book, Miles Vorkosigan, learns that, quote, "the key of strategy is not to choose a path to victory, but to choose so that all paths lead to victory."

The point is the same as what I’m talking about. You want to win no matter what happens, no matter which technology folks eventually adopt.

Right now, executives across hospitality, retail, B2B are all being asked some version of "what is our agentic commerce strategy? Do we have one?"

That is a reasonable question. That’s a really important question to think about and work on for your business. It’s also impossible to answer today because the infrastructure, who controls the checkout moment, which protocol wins—whether it’s Google’s UCP, OpenAI’s ACP, or something that has not yet been proposed—and finally, what your customers will actually do simply is unsettled. We don’t know yet.

Here’s the thing. The fact that you can’t answer that question does not mean you can’t act. It means that you need to act in a way that doesn’t bet on any one specific answer. You have to bet so that all paths lead to victory.

There’s a useful concept in strategy. You’ve heard this before. It’s called investing for optionality. You wanna give yourself options.

Instead of betting on one specific outcome, you make moves that pay off across a range of outcomes. The right agentic commerce strategy right now is an optionality strategy. Do the things that make your business better today, and that also happen to put you in a position so that you do well no matter which version of agentic commerce actually arrives.

So there are four things I think are valuable right now—or GEO, for AI search, for traditional search, and for direct business/direct bookings in hospitality and travel—that also will make you ready for agentic commerce when the infrastructure, you know, shows up.

If your content has schema markup, if you’ve got schema markup on your site, AI systems in recommendation mode right now and in purchase mode, you know, sometime not too far down the road, can read about your system, about your inventory. They can read about your prices or your rates. They can understand your amenities and the benefits that you offer to customers.

There’s data that suggests that sites with proper schema markup are cited in AI responses three times more often than those that aren’t. This isn’t something that’s speculating about the future. This is improving your GEO visibility this week. It also doesn’t hurt your traditional search either.

Remember, a few weeks ago I talked about this idea of "first do no harm." This is a perfect "first, do no harm action." Help yourself today and help yourself in the future.

The second thing you want to do, and we’ve talked about this as well, is review velocity on trusted platforms. Review how often people are talking about you.

AI systems give a lot of credit right now to review recency and review volume on sources they’ve been trained to trust, whether that’s Google, whether it’s TripAdvisor, whether it’s Yelp, whether it’s industry publications, whether it’s social media. These aren’t just agentic commerce signals, by the way. They’re the same signals that drive business and AI search inclusion right now. Encouraging customers and guests to review you on the right platforms has always mattered. it now matters more than ever. It’s also another reason why customer service is queen now and always.

The third thing you wanna think about, and this one is more about hospitality, though it has applications in other, um, spheres as well, other domains as well, and that is rate parity. Any tool, whether it’s an AI agent, a search result, or human browsing that finds a lower price on an OTA than on your direct channel sends the customer to another hotel or to an OTA.

This isn’t a new problem. It’s just something that has become a much bigger problem.

If Google for instance, doesn’t see that rate parity, you’re not gonna show up in AI overviews in AI mode. Pretty much all the time you’re gonna lose in meta search pretty much all the time. This is a prerequisite for any distribution strategy today, tomorrow, traditional, or agentic… period.

The fourth thing you want to do is build direct relationships with your customers and collect first party data. You’ve heard me say that data is the crown jewels many, many, many times. This is probably the single deepest lesson from OpenAI’s walk back.

The checkout moment is still resolved when customers trust you. And trust flows through relationships that don’t depend on a third party intermediary. The hotels that have direct relationships with past guests, clean CRM data and loyalty programs, that aren’t OTA dependent, have options when the distribution landscape shifts. That’s true, by the way, in retail, it’s true with B2B. When customers buy directly from you, when customers have a direct relationship with you and trust you, they’re more likely to come back to you again and again and again. The brand becomes the prompt. That’s where you want to live.

The businesses and brands that don’t do this are at the mercy of whomever controls the next layer, right? And as we all know, they become gatekeepers and gatekeepers gonna gate.

Everything I’ve just talked about is the entire argument, the whole argument of "Digital Reset." You cannot predict which gatekeeper will control the connection with your customer next. You can’t. It could be, uh Google. It could be Meta. It could be Amazon. It could be Perplexity. It could be ChatGPT. It could be Booking.com or Expedia. It could be AI agents.

Each one has looked like the definitive answer at some point or other, or like they could be the definitive answer. Right now we don’t know which ones will. We don’t. We can’t.

The companies that build direct relationships that own their data and invest in their hub, their own channels rather than their spokes, you know, platform mediated traffic, always have options when the landscape shifts.

The ones that outsource their sales and marketing to somebody else don’t.

You don’t need an agentic commerce strategy. You need a "direct relationship with your customers" strategy, one that makes you resilient to the next gatekeeper and the next channels that they support, regardless of when that happens or who that turns out to be.

Fundamentally, that’s how you ensure all roads lead to victory.

Now I’m gonna give you a real world example of this and how we’ve seen this happen in the past. Independent hoteliers have seen this story before.

Early in the 2000s, the conventional wisdom was that OTAs were a great distribution channel. They gave you high visibility, no upfront cost, and you paid only on conversion. Properties leaned in, and a decade later, and found themselves in a situation where OTA commission rates had risen to 15 to 25%.

Parity clauses had had grown so onerous that they really lost control of their own distribution. And the properties that had built their entire distribution strategy around an Expedia or a Booking.com found that they had no leverage to negotiate… and frankly, not a lot of direct channel to fall back on.

By the way, the same thing happened again with Google and metasearch.

The agentic commerce question is structurally the same thing. The question isn’t, "will AI agents book hotels?" They will, in some form. Sometimes today. And definitely eventually.

The question is, when AI agents do book hotels, are they going to route through Google’s UCP, Booking.com‘s agentic layer, Sabre’s system… or something that doesn’t exist yet?

Nobody knows! And that includes Google. And that includes OpenAI.

But hotels that have built direct relationships with past guest data and loyalty programs and recognition programs, and direct booking engines with, realtime rates and availability, and a great guest experience, they have a seat at any of those tables.

A hotel that is dependent on OTAs or on intermediaries for 35%, 40%, 50% of their revenue? They’re waiting to find out which intermediary is gonna control the world as they go forward. Mews called 2026 a make or break year for hotel transformation. But if you look at what they said, they didn’t talk about "Get your agentic booking infrastructure ready."

Instead, they talked about, "Hey, get your systems, your data, and your teams AI ready before conversational search, and before AI agents and assistants move from experiments to everyday guest expectations."

They’re telling you how to prepare. That’s what we’re talking about too, right?

So there are three moves you can take right now that you absolutely should be doing.

The first is to "run the agentic test" today. I talked about this a couple of weeks ago. "What is AI saying about you?" Go to ChatGPT. Go to Google, Gemini. Go to Perplexity. Ask each one to recommend a business like yours or a property like yours in your market. Are you in the results? Do you even appear?

We’ve talked about prompt brand equity; do you show up? Retail, B2B, hospitality. You all face the same question.

Importantly, this is today’s question, not next week’s or next month’s or next year’s. This matters right now. What those AI assistants and agents surface in their recommendation reality today is what any future a agentic booking system or commerce system is going to start from. If you don’t appear, you’re not even in the game

Second, audit your schema. Use Google’s free Rich Results Test, which I’ll link to in the show notes, and check whether your business gets it right. Do your inventory or your rates, or your prices, or your amenities or your room types, are they all correctly exposed via schema? This is free. It takes you maybe 15 minutes and it will improve your current AI search visibility, whether or not agentic bookings arrive today, tomorrow, next week, next month, next year.

Third, always treat your direct channel as the asset that it is. You know, rate parity. A good, clean, very responsive booking engine, and a reason to book direct, a reason why guests should choose you, why customers should choose you, are prerequisites for any go-to market strategy, including agentic ones. If you don’t have this fixed today, no amount of schema markup is gonna matter when the agent does the price comparison.

The honest answer to what’s your agentic commerce strategy is "we’re building a foundation that works regardless of how the technology unfolds." Because the last company that had announced a definitive answer changed course, they reversed themselves six months later.

That is not ducking the question, that is the strategically correct answer. It’s the answer that your digital reset is built around.

Build direct relationships. Own your data. Create great, customer experiences. And make your brand the prompt.

Not because you’re predicting the future correctly, but because those investments pay off in every version of the future, including the ones that nobody predicted yet.

If you want the full framework for how we think about this and how you can think about this, not just agentic commerce, but the whole gatekeeper problem that keeps shifting, I wrote "Digital Reset" specifically for you. The king, the queen, the crown jewels framework; the hub and spoke model; Your brand is the prompt. All of it’s in there, all of it developed in full. The link is in the show notes.

Thank you so much for tuning in once again. I genuinely appreciate you being here. As you know, this show would not happen without you. You can find the show notes for this episode at timpeter.com/podcasts. And if you know somebody who you think this episode would help, please share it with them. I’d genuinely appreciate it if you would do that.

Until next time, I hope you have a spectacular rest of your week and a spectacular week ahead. Until the next time we talk, please be well, be safe, and be excellent to each other. I’ll see you soon.

The post Agentic Commerce: ChatGPT Bails on Its Shopping Plans (Digital Reset Episode 487) appeared first on Tim Peter & Associates.

8% of companies have adopted AI. Only 6% are seeing "significant" value. That leaves a staggering 82% value gap most businesses face.

If your team feels like they are running faster just to stay in place, you aren’t suffering from a tech problem — you’re paying an "AI Tax."

In this episode, Tim Peter breaks down the data from McKinsey, Section, and Workday to reveal why the C-Suite thinks AI is a miracle while the front line sees it as a burden. More importantly, he talks about what you can do to close the gap for your business.

Key Insights for Strategic Leaders to Close the Gap

Tim Peter has written a new book called Digital Reset: Driving Marketing Beyond Big Tech. You can learn more about it here on the site. Or buy your copy on Amazon.com today.

Rutgers Business School MSDM Speaker: Series: a Conversation with Tim Peter, Author of "Digital Reset"

We have some free downloads for you to help you navigate the current situation, which you can find right here:

Contact information for the podcast: [email protected]

Recorded using a Shure SM7B Vocal Dynamic Microphone and a Focusrite Scarlett 4i4 (3rd Gen) USB Audio Interface.

Running time: 16m 24s

You can subscribe to Thinks Out Loud in iTunes, the Google Play Store, via our dedicated podcast RSS feed (or sign up for our free newsletter). You can also download/listen to the podcast here on Thinks using the player at the top of this page.

Welcome back to the show. 88% of companies use AI. 88% of companies use AI. Only 6% see significant value from AI. 6%. That is a massive 82% gap between those who use AI and those that actually get value from its use.

82%, for those of you who don’t love math, is a lot of percent. It’s hard to have a whole lot more percent than that.

That gap was first reported by McKinsey towards the end of last year. And in the three months since the report dropped, I’ve had conversations with C-suite and digital leaders at companies ranging from the Fortune 100 to individual hotel owner/operators that usually tell me something strikingly similar. Lots of companies are "playing with AI" or "testing AI" in some portion of their business or other. They’re mostly looking to drive efficiencies, to do more with less.

Despite these efforts, they’re mostly not seeing the value. They’re falling into that 82% value gap too.

What we’re going to talk about today is what lives in that gap and how you can get to the other side of it. This is Digital Reset with Tim Peter. I’m your host, Tim Peter. Let’s dive in.

Okay, so there’s an 82% AI value gap. That is 100% real. There’s also a big gap between the people who see real gains from AI and those that don’t.

The Wall Street Journal reported a few weeks back on a research study from a group called Section that found 76% of C-suite execs thought AI was saving them “at least four to eight hours each week.” Essentially one in five claimed that they saved more than 12 hours per week using AI.

By contrast, 40% of non-management workers said that AI didn’t save them any time at all. Another 27% said that it saved them less than two hours per week. Only 14% said that it saved them greater than four hours a week.

So that’s 76% of execs versus 14% of line employees saving more than four hours. That’s a 62% productivity gap. As the article notes,

“A new report from the business software company Workday goes so far as to call frustrations with the technology an AI tax on productivity. Though 85% of the roughly 1,600 employees it surveyed reported saving one to seven hours a week by using AI, much of the time was offset by having to correct errors and rework AI-generated content.”

Think about those numbers I’ve just talked about. Those are huge differences. An 82% value gap. A 62% productivity gap. An AI tax that employees are paying. What in the world is happening here?

First, let’s unpack McKinsey’s 82% value gap. Note that AI does produce value every day. And admittedly, a small number, 6%, see “significant” value. There’s also a larger share who see some value, just not at “significant” levels.

Similarly, think about the AI tax on productivity that 62% of workers see here.

It’s not a question of does AI work. It’s far more about how and where the folks that get the value put AI to work. I am seeing too many companies, you know, “play around with AI” or “test AI” without clear plans, clear objectives, and a clear strategy for what they really want AI to do. The McKinsey data says it. I see it every single day.

I know of companies whose execs delivered mandates to their team, things like, “Everyone will use AI for X percent of their tasks,” and similar such messages.

That’s not a strategy. It’s not even a goal, really. It’s more of a hope of what could be. What it lacks is a strategic underpinning.

You might remember Klarna famously announced a couple of years ago that, “AI did the work of 700 human agents and does it much faster.” As their CEO said in December 2024, "I am already of the opinion that AI can do all of the jobs that we as humans do."

Less than five months later, they reversed course and started hiring again in serious numbers. And the reason is simple. They moved too far, too fast, and with too little thought about what they really needed.

I didn’t say that, by the way. The company CEO, Sebastian Semetkowski, said it himself. MLQ.ai had a write-up where they wrote, "We went too far, he said, noting that the focus on efficiency and cost ultimately reduced the quality of the company’s offerings and eroded trust with customers."

Let’s start with the fact that companies that get the most out of AI, by contrast, have a clear strategy in place for what they expect of their AI efforts. They’ve got well-defined objectives they want to reach.

And those objectives aren’t, “We want to use AI 50% of the time.” Their objectives are specific. They’re relevant. They’re time bound.

Consider this hotel client of mine. They want to drive more occupancy, put more heads in beds. They want to increase the revenue they achieve from their guests. They want to increase guest satisfaction and retention. They want to lower their costs for achieving these. And they want to accomplish this within a particular time frame.

I want you to notice something about those objectives. Did you notice how none said, “by using AI?”

You know why?

Because it doesn’t matter if they achieve those goals with AI or without it. It matters if they get more customers, keep those customers happy, encourage those customers to provide more positive ratings and write more positive reviews, and reduce their operational costs while they do that. That’s how their hotel succeeds. That’s how their business succeeds.

Of course, they then looked at AI for ways to meet those objectives. And they were super willing to rethink internal processes where it made sense to do so. But the conversation started with strategic objectives and a strategy for how to get there that included AI, not “use AI and hope for the best!”

They were also realistic about where AI might not work — usually because they didn’t have the right data or the right skills available — so that they didn’t waste time and resources along the way.

Companies achieving value from their AI initiatives target growth and innovation, not just efficiency. They put in place well-defined success metrics that allow them to evaluate if they’re moving in the right direction or, you know, not. They create these metrics because they’re focused on business results, not just technology, not just AI, not just whatever the new shiny thing is.

You’ve heard me talk many times before about how data is the crown jewels. Well, your metrics are one type of data. They keep you on course. They tell you when you’re succeeding, and they tell you when you’re not.

Another type of data that matters is the data that your AI tools need to learn from. Companies succeeding right now make their data available to the tools that they’re working with. That data is a massive strategic asset. It’s what makes AI useful. They’re using AI to better understand their data and improve their products and services. And when they don’t have the data, they know that that’s not a place where AI is going to help them.

Now another thing you’ve heard me say is that content is king. That’s still true. The funny thing about AI is we can use it to create content and that’s good. We also can use AI to help our content be better and that’s better.

But weirdly, it’s not just about creating high quality content, though that’s important. It’s that you need to be careful when you use AI to create large volumes of content.

I’ve long said that content isn’t expensive, but content that doesn’t convert is. Well, that needs a little bit of modification in the era of AI in which we live, because you can use AI to crank out a ton of content.

What we’re finding though is that quality now matters even more than quantity, much, much more. Because it turns out that low quality content has a massive, massive cost to your business. Google hates low quality content. They think it’s spam. And I expect most of the other AI extra engines are working their way to that too.

Lower ranking traditional search engines is a big deal — something I talked about during last week’s show when I was referring to “first do no harm” in your search engine optimization efforts. If you have low quality content, guess what happens? You don’t show up there and you hurt your overall appearance.

In a very related way, you’ll also get less visibility in AI answer engines because they will not see your website as a trusted source.

Now, does that mean that every piece of content needs to be created by human beings? No, no, definitely not. What it does mean is that someone, a person who can exercise judgment, must review your content to make sure it works well for its intended audience, to make sure it actually speaks in your voice and is quality. You can compare your content with Google’s Search Quality Rater Guidelines; I’ll link to those in the show notes.

Those aren’t just what Google tells its quality raters. It’s also how those quality raters train Google’s various AI tools to recognize high quality content. We internally have actually built a series of instructions around using Google’s Search Quality Rater guidelines in our prompts when assessing client websites and content. Drop me a line if you want to learn more about that.

But that’s how it all works together. It’s the right specific relevant time-bound goals to measure against. It’s content that answers your customers questions and shows up in both search and AI answer engines. It’s AI initiatives that are focused on improving customer experience and greater revenue. And it’s customer data, your company’s crown jewels, that power those insights and experience.

So there are three questions I’d like you for you to think about this week that I think are really important:

As I mentioned in an episode a couple of weeks ago, Google saw its best quarter and year ever by applying AI to its ads product. And that’s cool, you would expect that from a Big Tech giant.

Individual independent hotels can see the same results, though. Businesses of any size can. Size does not matter. Strategy does. A clear goal does. Content, customer experience, data, and execution do, too.

AI is not a magic wand. It’s a mirror. It reflects the quality of your strategy and the depth of your data. If you’re just using it to crank out more content or check a box on mandated tasks by the C-suite, all you’re really doing is paying an AI tax on your brand’s future.

The 6% of companies who win, the 6% who have closed that gap don’t have an AI strategy. They have a business strategy that AI makes possible. Focus on effectiveness, protect your data, and don’t let the pursuit of efficiency kill your customer’s trust. Because ultimately, your customer’s trust is what will make your strategy pay off in the long run. And that’s where we always want to be.

If this episode helped you bridge that value gap for your business, if it helped you think about how you can bridge that value gap for your business, please do me a favor. Send it to one of your colleagues who’s currently staring at their 2026 budget and wondering how to make the numbers work.

You can find the links to the McKinsey and Workday reports and everything else we’ve talked about in the show notes, as well as a full archive of all of our past episodes at timpeter.com/podcasts.

And if you’re ready to move beyond big tech, my book, Digital Reset, is your roadmap to reclaiming your demand.

Thank you so much for listening. I genuinely appreciate you. Until next time, please be well, be safe, and be excellent to each other. I’ll see you soon.

The post The AI Value Gap: Why 82% of Companies are Failing to Gain from AI (Digital Reset Episode 486) appeared first on Tim Peter & Associates.

There is a lot of mixed information out there regarding how to rank in the age of AI. In this episode of Digital Reset with Tim Peter, Tim cuts through the noise to explain the transition from traditional Search Engine Optimization to Generative Engine Optimization (GEO). While the tools are changing rapidly, the core mission remains the same: ensuring that your customers’ “digital concierge” knows exactly who you are and why you’re the right choice for your customer.

Key Insights for Strategic Leaders

Tim Peter has written a new book called Digital Reset: Driving Marketing Beyond Big Tech. You can learn more about it here on the site. Or buy your copy on Amazon.com today.

Rutgers Business School MSDM Speaker: Series: a Conversation with Tim Peter, Author of "Digital Reset"

We have some free downloads for you to help you navigate the current situation, which you can find right here:

Contact information for the podcast: [email protected]

Recorded using a Shure SM7B Vocal Dynamic Microphone and a Focusrite Scarlett 4i4 (3rd Gen) USB Audio Interface and edited in Ableton Live 12 Suite.

Running time: 20m 29s

You can subscribe to Thinks Out Loud in iTunes, the Google Play Store, via our dedicated podcast RSS feed (or sign up for our free newsletter). You can also download/listen to the podcast here on Thinks using the player at the top of this page.

Welcome back to Digital Reset with Tim Peter. There is so much mixed information about what’s going on with AI tools and most importantly, what you need to do to show up in AI answer engines.

That’s "mixed information," by the way, not "mis-information". I don’t think people are lying. OK, mostly nobody’s lying. I bet somebody is though.

The truth, though is that we’re still in the very early days for so much of what is going on with artificial intelligence. And so it’s hard to authoritatively state that this is the absolute one and only possible truth about the way that AI works and that AI answer engines work.

As you might imagine, I’m somewhat skeptical about that framing that there’s only one possible answer here.

First, up until recently, most digital marketers couldn’t even agree on what to call the practice of showing up in AI answer engines. Some called it GEO, some called it AEO, some called it AIO. Oy!

If you don’t know what your practice is called, if you don’t know what you’re actually working on, that’s a damned fine sign that the practice itself isn’t very mature. For now, I’m calling it GEO. At least judging by Google Trends data, GEO appears to be the winning term. It looks like a pretty clear winner.

If I’m being honest, I don’t really love the term GEO. I fully admit to thinking that doing right by search engines generally benefits AI and that it’s all SEO in the end. I appear to be losing that argument though. Such is life.

More importantly, these tools are evolving and fast. What’s true today may be less true weeks or even days from now. You can’t go a day or more without some new change being made or some new change being announced by the big players around these tools.

What is most important is making sure that AI answer engines and AI assistants and AI agents know about your brand and your business. It’s why I’m so keen on this concept that your brand is the prompt. It’s easiest to ensure you show up if your customers and their AI assistants and agents ask for you by name.

Regardless, you want, and need, these tools to surface your brand and recommend you to their users. That’s even more critical as we move towards agentic interfaces, which are coming. Your brand has to appear as an option or else you simply don’t exist. We’ve moved past an age of card catalogs into an age of concierges.

As I talked about last week, I think that Google’s AI Overviews are the biggest single AI that most customers interact with regularly. And I expect that growth to continue. So there’s a ton of overlap, not only between your SEO and GEO goals and efforts, but also in terms of how your customers might find and experience your brand in the immediate term to midterm.

My goal today for this episode isn’t to tell you how you show up. Instead, I want to talk about some foundational elements that matter as you work to show up, that ensure your brand and your business benefit, no matter how these tools evolve and change over time.

I want to make sure you can tell where it’s important to focus your energies and where it’s not as you help your customers find you. This is episode 485 of The Big Show. Today we’re talking about SEO and GEO in the age of AI. Let’s dive in.

So I think we all agree it’s unbelievably important that you show up in AI answer engines. What I’m not sure we completely agree upon is the fact that you need to show up in SEO still too, in traditional search.

I want to give full credit to SEO expert Lily Ray for this idea, but we want to think about the fact that first, “do no harm.” Forget for a moment how you show up in AI answer engines. Of course you want to do that.

I also suspect, and see this with my own clients, that for most businesses, even in a world of zero-click searches, Google still represents a significant, and probably your largest, single source of traffic.

So the first thing you want is don’t do anything that screws up what Google gives you. This isn’t an SEO versus GEO question. It’s SEO plus GEO. Remember, you don’t get the plus if you mess up one side of the equation or other.

As you read GEO advice, as you listen to GEO experts, as you listen to people talk about this, you need to ask yourself, “does this have the potential to hurt our current organic situation?” If the answer is no, go ahead and do it. Do it immediately even.

If the answer is yes, well, don’t ignore it outright.

Instead, then ask, “okay, how do we gain the benefits of this recommendation while also keeping our existing SEO position — or better yet, improving that existing SEO position?”

Again, it’s not either SEO or GEO. It’s SEO and GEO, SEO plus GEO.

Second, and you’ve heard me say this many times that customer experience is queen. I’m kind of bundling the first two portions of our royal court of brand building here. You’ve heard me say many times in the book, Digital Reset, that content is king and customer experience is queen.

Those are still true. What’s also true, is that your content has always included what others say about you. You have a secret sales force working for you — or for your competitors — based on what your customers’ experiences are and what they say about those experiences in social and in ratings and reviews. They’re telling their friends and family and fans and followers all about what they think about your products and services and your brand more broadly.

If you’re in a category like hospitality or travel or any kind of B2C, what others say about you in public ratings and reviews plays a massive role here. You’re undoubtedly familiar.

In B2B, the same is still true. It just tends to happen in private. It could be text or LinkedIn messages. It could be email. It could be on Slack or Teams channels or phone calls. But folks ask other folks, “Hey, what do you know about this company? What do you know about their products and services? Do they work well for you?”

Your content has always included what others say about your brand and business. Well, the data seems pretty clear that those discussions influence whether or not you show up in AI responses. That’s something we just have to keep top of mind.

You also want to keep do no harm in mind, too, because what people say about you, whether it’s public or private, drives awareness, interest and action among your potential customers.

Ideally, that’s to your business’s benefit. But even a poor customer experience creates a narrative around your brand and one that you’ll fight against for a long time to come, especially in GEO. If “the concierge” is reading all kinds of bad things about your brand, guess what it’s going to know about your brand?

We want to move to an era where your brand is the prompt. And if somebody types in your brand and the answer engine says, “Man, I don’t know about that brand…” that’s really, really, really bad for your business. Not just now, but over the long term.

Making sure that when your brand is the prompt that the concierge has great things to say about you, about your brand, about your business, is a clear component of why customer experience is queen now more than ever. So seriously, take care of the queen. This really matters.

Finally, if you had have some exposure to our Royal Court framework that I’ve talked about many times, you’ll know that I say data is the crown jewels. And I want to talk for a minute about what that means in this case.

I’m seeing many businesses experience a big lift in branded traffic or conversions that don’t have an obvious source. Folks are just coming to your website or opening your app or reaching out to your sales team or sending you an email without starting on your site. They just kind of appear. And if they’re not doing that, well, you want them to, regardless of whether they’re doing that today or not. Listen to my prior points from a moment ago if you need more help on how you can do that.

The question we want to be thinking about then is if people just show up, what metrics matter to your business? We can’t rely on the metrics we’ve used for a long time. Now is the time to get your KPIs and your metrics in order so that you can actually determine what is our current state and are we moving forward in a direction that we want to.

Again, data is the crown jewels. They help you make better business decisions and take better care of your customers so that they actually have a better experience, so that you actually have a better shot of showing up more often.

Clearly, we’re going to lose some visibility into what’s happening compared to what we’ve had for years. You might not get the same level of tracking or the same level of insights you’ve had in the past. That’s okay. We just need to keep track of the metrics that genuinely matter and help you make better business decisions.

Now the first one should be obvious, it’s revenue, right? Are you seeing more revenue or less revenue? Ultimately, that is the final determinant of whether or not any of this is working for your business. If your revenue is going up, that’s probably a good sign that you’re doing the right thing. If your revenue is declining, that’s probably a clear sign that you need to do something different than what you’ve been doing up till now. So that’s a huge one you want to pay attention to.

Of course revenue is a trailing indicator. It only tells you what happened in the past. It’s not necessarily all that predictive of what’s going to happen in the future.

So another key performance indicator you want to watch is lead volume. That could include actual leads in a B2B context. It could also include the number of people who put items in their shopping carts in B2C or enter your booking engine in a hospitality context. Are people showing interest in engaging with your brand and your business in a commercial way?

Do they want to talk to you? Do they want to hear from you? Pay super close attention to whether they are signing up for your email lists. Are they giving you their information for SMS? Are they saying, “I want to have a conversation with you, not just now, but in the future and want to hear from you directly?” Again, that’s a super, super huge sign about where your business is going to go in the future.

We do a bunch of work with hospitality clients and we’ve determined that for your typical set of hotels, every new email address you collect is worth somewhere between five and thirty-five dollars in future revenue. So if that data isn’t showing up, if you weren’t collecting that data from your customers, it’s a pretty good sign that your future revenues are going to have some real problems.

Again, data is the crown jewels here. It points to real future revenue.

Another metric that matters a ton are brand searches. Are the number of searches for your brand in organic traffic shrinking or growing?

If it’s shrinking, you’re likely not showing up in AI or traditional search near as much as you’d like to or need to. People aren’t finding you. They then aren’t then going to Google and ask for you. Again, review my prior points for what you can do about that.

If you see those numbers rising though, it’s a pretty good sign that AI is likely surfacing you more often. AI is talking about you and people are learning about you to then come over to your website by searching you by name. That’s a really clear use case that we’ve seen customers do where they learn about you and then they go ask traditional search to help them find you and find your website. Hugely important.

A more recent and more sophisticated way to watch this is something called prompt brand equity. This is a term that we’ve talked about a bunch; it’s something that I think matters a lot as we go forward. And of course, I’ll link to in the show notes other times we’ve talked about this.

Now, prompt brand equity is not about where you rank in AI search results. It’s how often you show up at all.

There’s some great research from Rand Fishkin that shows where you appear in AI answers is not at all predictable. You could be number one in one chat, number two in another chat, and number three in a third chat, even though you’re an equally good answer for all three chats, even if the prompt was precisely the same. It’s not predictable. And it’s not predictive of what’s going to happen with your business.

The thing you want to keep in mind here is it’s not about where you rank. It’s does your brand appear at all? Do you ever show up? Don’t worry about tracking where you rank, but absolutely pay attention to whether or not you appear.

There are a bunch of pretty good tools for measuring this metric right at the moment. Tools like Peec AI, seoClarity, FinSEO, SE Ranking, Profound, and Otterly all seem to get good reviews and good feedback for how well they measure the frequency of how often your brand appears. It’s probably worth checking out each of those to see what might work best for you.

My point here isn’t to recommend one tool or another. Your specific needs may lean you towards one versus the other. My point is to strongly recommend that you keep track of whether or not your brand appears at all. Otherwise, you have no way of knowing whether in fact your brand is the prompt… or isn’t.

Tracking how frequently your brand appears is a great way to learn whether or not your content marketing and GEO efforts work. If you’re showing up, that’s a good signpost that your content is being used by AI answer engines. And if not, well, then your content pretty likely isn’t doing its job.

It’s critical to understand what’s working here and what isn’t. As I mentioned last week, we’ve moved from a world where you just needed to show up in a card catalog to a world where you need to convince a concierge that you’re the right choice. That’s why taking care of both traditional search and GEO matters.

Your successful digital reset isn’t about fighting with machines. It’s about ensuring your brand signal is clear enough for those machines to find you. Because in this new era, your brand isn’t just a logo or a label… not that it ever was. It’s that your brand is the prompt — or it needs to be if customers will consistently find you no matter where they go looking.

If this episode helped you make sense of some of the noise around SEO and GEO and about digital more broadly, do me a favor, send it to one of your colleagues who is currently staring at their 2026 budget and wondering what they should do next.

You can find the links to the GEO tracking tools we discussed, the show notes for this episode, and the full archive of all past episodes at TimPeter.com/podcast.

And if you’re ready to move beyond big tech for your business, my book, Digital Reset, Driving Marketing and Customer Acquisition Beyond Big Tech, provides the roadmap to reclaim your demand. You’ll find a link in the show notes.

Thank you so much for listening today. I genuinely appreciate you. Until next time, please be well, be safe, and be excellent to each other. I’ll see you soon.

“Digital Reset with Tim Peter” helps you look beyond the "shiny objects" to build a business that lasts. How can we help you today?

The post SEO vs. GEO: How to Show Up When AI is the Concierge (Episode 485) appeared first on Tim Peter & Associates.