ChatGPT Shopping Hype vs Reality: Why We Tell Brands Under $10M to Ignore It (For Now)

Here's a confession that probably costs me some cool points at conferences: we tell most of our clients to ignore ChatGPT shopping for now.

Not forever. For now. And the smaller you are, the more I mean it.

I know how that sounds in 2026. Half the LinkedIn posts I scroll past are some version of "the chat window is the new storefront" and "if you're not optimising for AI search you're already behind." There are people charging real money for generative engine optimisation retainers. And every week a founder asks me what our AI search strategy is, like I'm meant to look sheepish when I say we don't really have one yet.

So let me explain the actual reasoning, because it's not laziness and it's not luddite energy. It's the same discipline that keeps a small brand alive.

What the hype gets right

I want to be fair to the bull case first, because parts of it are genuinely correct.

The smart version of the argument goes like this: AI chat isn't another app sitting on top of the internet. It's closer to the internet itself. People are starting to discover, research and decide inside the chat window, and over time a chunk of buying just happens there. The chat is the store, not a billboard pointing back to your store.

I think that's directionally true. I really do. There's a type of buyer forming right now who'll find products inside a chat the same way an older cohort defaulted to typing things into a search bar. You'll want to be present for that buyer. Reports of LLM-referred traffic jumping several hundred percent year on year aren't made up, either.

So this isn't a "fad that'll blow over" take. I'm a fan of where it's heading.

The disagreement is entirely about timing and size. When does it matter, and for whom.

What the reality actually looks like right now

Here's the part the hype skips: go and pull the numbers.

I've now compared notes with a few operators running genuinely large brands, the kind doing tens of millions a year. The ChatGPT-referred traffic they're seeing is in the low hundreds of visits a month. Not thousands. Hundreds. One was sitting around 380 visitors in a 30-day window. Another a touch over a thousand. The conversions off that were single digits to low double digits.

To put that in perspective: that's a rounding error on a brand of that size. A good Meta campaign can move more new revenue before lunch than that channel moves in a month.

And these are the brands with every advantage. They've got a decade of reviews, content and mentions seeded across the web, so when someone asks a chat assistant for the best product in their category, they surface. A newer brand has none of that head start and even less to show for the effort.

Then there's the visibility problem, which is the bit that really bothers me as someone who likes to optimise things.

With ChatGPT there's no Search Console. No Trends. No keyword report. You can't see where you're ranking, what you're ranking for, or whether last week's work did anything. It's a black box. You're effectively flying blind and hoping. Tools are starting to appear that crack the lid open a little, and we're watching those closely, but right now you're optimising against a wall you can't see through.

So you've got a channel that, for most brands, delivers a tiny trickle of traffic, that you can't measure, that you can't really steer. That's not where a 6 or 7-figure brand should be pointing its best hours.

The real risk isn't doing nothing

Here's my actual worry, and it's got nothing to do with ChatGPT specifically.

A founder running a brand under, say, $10M a year has one genuinely scarce resource, and it isn't money. It's focus. The list of channels that can reliably absorb real budget and pay you back is short. Properly short. Meta, Google, and a small handful of others on a good day. Almost any decent brand can show up on Meta and Google and make the maths work. That's rare and it's precious.

The way small brands stall isn't by missing the next shiny channel. It's by chasing it. You read a thread, you get a Slack from someone, you spend three weeks down a rabbit hole, and the thing quietly turns into nothing while your core engine drifts because nobody was watching it.

I'd put it bluntly: I'm not interested in a channel we'd spend a couple of grand a day on while there's a channel right there that can take ten times that and grow. If fully optimising ChatGPT shopping today would, best case, double your trickle and earn you a few thousand dollars this quarter, that's not a strategy. That's a distraction wearing a strategy's clothes.

Pulling attention off the thing that actually scales your revenue, to go chase a channel with a low ceiling, is one of the most common own-goals I see. It feels productive. It feels current. It's neither.

The three cheap seeds worth doing now

That said, "ignore it" doesn't mean "do literally nothing." It means don't hire an agency for it or build your quarter around it. There are a few seeds that cost almost nothing and quietly compound, so that when this channel does matter, you're already in the answer set.

Here's what I'd actually do if you're a 6 or 7-figure brand:

• Get mentioned where the assistants are reading. Chat assistants pull heavily from places like Reddit, forums and review sites. So show up there honestly. Real reviews, real threads, real presence in the communities your customers already hang out in. This is just good brand hygiene that happens to feed the machine.
• Win the articles the answers cite. When a chat assistant answers a buying question, you can usually see which articles it leaned on, and they're often not the ones ranking on page one of Google. Find the pieces that get referenced in your category, and work to be included or improved in them. It's a much smaller, more targeted list than chasing classic SEO.
• Seed your reviews and content now, not later. The big brands' advantage is just time, years of positive signal stacked up across the web. You can't buy the ten-year head start, but you can start stacking today so future-you inherits it. Every honest review and useful piece of content you put out now is a deposit.

Notice what those three have in common. They're cheap. They're mostly things a good brand should be doing anyway. And not one of them asks you to take your eyes off Meta. That's the test. If a "ChatGPT strategy" pulls a person or a budget line off your scaling channel, it's failed the test.

There's a different conversation for brands selling something genuinely niche, where buyers ask very specific long-tail questions a chat can answer cleanly. If that's you, the maths can tip earlier and it's worth a harder look. But that's the exception, not the rule, and you'll know if it's you.

Where I'd leave it

So no, I don't think you're behind. I think most of the people telling you you're behind are selling something, or are nine-figure brands whose situation looks nothing like yours.

Be present in the cheap ways. Stay alert as the measurement tools mature, because the day you can actually see and steer this channel is the day my advice changes. But protect your focus like the scarce thing it is, and keep your best hours on the channel that's paying the bills today.

If you genuinely can't tell whether a new channel deserves your attention or is just quietly bleeding focus from the one that scales, that's usually a sign the core engine hasn't been looked at closely in a while. A Signal/Noise Audit is built to settle exactly that question: where your next dollar of effort actually pays you back, and which shiny thing is safe to ignore for now.