Most brands launching ChatGPT Ads today will waste their first few months. The platform won’t be the problem. Their readiness will.
- The ads are real and self-serve. You can buy them today.
- They only reach Free and Go tier users, in a limited set of markets, so the reachable audience is a lot smaller than ChatGPT’s headline numbers.
- Readiness predicts results better than budget does.
- At KINAI we check six things first. Each one tells us go or wait.
- The first campaign is for learning. Scaling is a later problem.
The search bar is losing the first conversation
Your customers have not stopped buying. They have changed where the research happens. More of them now open ChatGPT, ask a real question, push back on the answer, ask a few more, and only then start looking at actual products. By the time they reach your site, the AI already shaped what they think they want.
ChatGPT Ads allow brands to appear inside that conversation, at the moment buying intent is forming. That is rare, and it is valuable. It is also easy to get wrong, which is exactly the gap we built KINAI to work in.
We have spent the last few months studying the platform, watching who moved early, and building our first campaigns on it.
“A business does not become ready because OpenAI opened the doors. It becomes ready when every dollar spent generates meaningful learning. And those are rarely the same day.”
Javier Gómez · Founder, KINAI
How ChatGPT Ads actually work
Sponsored cards that appear below ChatGPT’s answer when your question points at buying something. OpenAI labels them clearly and says they do not bend the assistant’s real response. Matching is contextual. There is no keyword auction like Google; the system reads the conversation and decides what fits.
The journey is the part people underestimate. Someone might ask five questions, compare two competitors, talk themselves out of one, and land on your site already halfway sold. Or already gone. Your creative and budget still matter, and so does everything the click lands on.
Who actually sees them (read this before you budget anything)
Here is the fact that quietly kills a lot of plans. For now this is a limited rollout. The ads only reach logged-in adults on the Free and Go tiers, and only in a short list of markets that OpenAI is still adding to, currently a handful of mostly English-speaking countries like the US, the UK and Australia. Anyone paying for Plus, Pro, Business, Enterprise or Education never sees one. Above Go, it is ad-free by design.
That list will look different in a few months, so the real question is never whether the ads exist. It is whether they are live where you sell, right now, and whether your buyer is the kind of person who sees them at all.
Because think about what the tier rule does to your audience. If you sell consumer or e-commerce products and most of your customers use the free version, the reachable audience is enormous. If you sell something premium or B2B, and your decision-makers are exactly the people who pay 20 dollars a month or more to strip the ads out, you can run a flawless campaign that your actual buyer will never encounter.
So the first question we ask at KINAI has nothing to do with the framework. It is this: do the people who pay you use ChatGPT for free, in a country where the ads run? If the answer is no, most of the rest is academic.
Why the smart money is paying attention anyway
Because a real chunk of buying research now starts here, and getting good at a channel while it is still cheap and uncrowded is how you build an edge that is hard to copy later. Google search ads looked unglamorous in 2003 too.
Early does not mean today, though. Sometimes the best move is to spend a quarter fixing the leaks first, then launch once the pipes hold.
Not every business should launch ChatGPT Ads today.
At KINAI, we believe the decision to advertise should follow a readiness assessment, not market hype. That is why we built the KINAI AI Advertising Readiness Framework.
The KINAI AI Advertising Readiness Framework
Every client conversation opens with the same question. Should we invest in ChatGPT Ads? At KINAI we answer a different one first: are you ready to invest well?
Does AI already mention you? When ChatGPT answers a question in your category, does your name come up, and is any traffic already trickling in from AI tools? If the honest answer is no, and nothing is trending up, your first money belongs in content and authority, not ads. Paid spend amplifies a signal. It cannot invent one out of silence.
Early inventory has been running near a 60 dollar CPM in the US, and you can start at roughly 25 dollars a day with no account minimum, so testing is cheap to begin. Cheap to begin is not the same as safe. What we actually look at is whether your margins can carry an acquisition cost that will sit above your normal blended CAC while the channel finds its feet. High-margin, high-AOV or long-LTV businesses have that room, which is exactly why we pressure-test the economics before spending, not after. If the numbers only work once everything goes right, the business is not ready to fund the learning phase this channel requires.
People arriving from an AI chat have done half the work already, so your page has to finish it fast. Load under 2.5 seconds, and answer the handful of questions a buyer would otherwise go and ask ChatGPT. If the assistant explained your category better than your own landing page does, the traffic bounces and you paid for the privilege.
The boring one that decides everything. Before launch you need to know where a conversion is recorded, how you will credit it, and the single KPI the pilot lives or dies on. ChatGPT Ads take UTMs, so the tracking is doable. If you cannot tie a sale back to the channel with a straight face, every decision after launch is a guess wearing a suit.
A first campaign that only answers “did it work” wasted most of its budget. A good one answers more. Which prompts pull people in? How do they behave once they land, which page converts them, and what were they still unsure about at the end? Write those questions down before you spend, because if they are not written, you will rationalise whatever the numbers happen to say.
Availability is patchy on purpose. The ads run in a limited set of markets and only on the Free and Go tiers inside them, so geography and audience are both gates. At KINAI we score each market on its own: availability, the size of the free-tier audience there, demand, the state of the local site, authority, analytics, and whether the economics survive contact with reality.
Recently, we applied this framework for a luxury jewellery brand selling across multiple international markets, some now eligible. The instinct was to switch it on everywhere at once. We did not. Instead, we selected the one market where platform availability, audience fit, website readiness and commercial viability aligned before launching the pilot.
In our first live market, average CPMs are landing around A$29 (approximately US$21), well below the CPMs being discussed during the US launch. Over a 30-day window, the pilot has run in short, deliberate bursts rather than continuous spend: A$1,379.67 in total spend, 47,310 impressions and 466 clicks, at a A$2.96 average CPC and a click-through rate just under 1%. At this stage, success isn’t measured by immediate revenue. Our focus is validating attribution, understanding user behaviour, refining landing pages and documenting insights before making any scaling decisions.
Where people go wrong
We see the same pattern repeatedly. Someone launches because the button exists, then expects Google-search performance from a channel that behaves nothing like search, then cannot explain the results because nothing was tracked, and finally scales the confusion.
Two more are specific to this platform. Brands assume ChatGPT’s headline user count is their reachable audience, when the ad-eligible slice is a fraction of it. And they read the first campaign as a verdict rather than a lesson, so they quit a week before it would have taught them the thing worth knowing.
Questions we get
Who actually sees ChatGPT Ads?
Logged-in adults on the Free and Go tiers, in the markets where OpenAI has switched ads on so far. Paid subscribers above Go do not. Check that your buyers live inside that group before anything else.
Are they available worldwide?
Not yet. It is a limited rollout that keeps expanding, so confirm your market is live before you plan around it.
Who benefits most?
Businesses with real margins, working analytics, a site that converts, and customers who use ChatGPT free. Miss that last one and the rest will not save you.
Should everyone jump in?
No. For plenty of businesses the smarter first move is to build authority and fix measurement, then revisit this a couple of quarters later.
How do we start?
One market, one clear success metric, treat month one as tuition, and expand only when the data earns it.
The honest version
Being first on a new ad channel only helps if you are ready to use it. Most of the brands sprinting into ChatGPT Ads this year will waste an entire quarter learning things they could have known before they spent a dollar, and the platform will get the blame. In most cases, the platform isn’t the problem. Readiness is. We bring the same discipline to this channel that we bring to the rest of our paid media work.
Technology changes.
Great strategy doesn’t.
Powered by AI. Led by People.
At KINAI, we help brands decide not only where to invest, but when they’re ready to invest. If you’re exploring ChatGPT Ads simply because everyone else is, that’s rarely the best reason to start. We’d rather recommend waiting than launch a campaign your audience will never see.
Book a Strategy Call →About KINAI. KINAI is an AI-powered performance marketing and strategic consulting agency helping ambitious brands grow through paid media, AI advertising, digital strategy and executive communications. Powered by AI. Led by People.
