The latest development in OpenAI’s advertising saga has emerged. eMarketer reports that it’s pacing well below its own ad revenue projections. Specifically, the company is reportedly 90 percent off its mark, as it experiences growing pains in bringing sponsored answers & content to ChatGPT.
Before diving into those details, let’s take a step back for context. What is OpenAI’s ad program and how did we get to this point? In short, it announced plans in early 2026 to integrate advertising into ChatGPT. With a rumored IPO on the horizon, this is meant to monetize users’ high-intent ChatGPT questions.
As we’ve examined, this mirrors the rationale and business case behind paid search. And sure enough, ChatGPT ad formats so far naturally resemble sponsored search results. There have been some ups and downs in rolling out this program (more on those in a bit), but it’s largely been based on a logical idea.
This natural fit caused OpenAI to be ambitious in its goals. It projected $2.5 billion in ad revenue for this year and $100 billion by 2030. But eMarketer pegs the entire AI ad market as less than $1 billion in 2026, growing to $5.41 billion by 2030. That makes OpenAI’s 2030 revenue projection ~20x the entire market.
The Heineken Method
That brings us back to the latest development, as OpenAI not only overshot the long-term possibilities but is reportedly tracking below its own projections. One reason could be an overly ambitious vision for the broader role of AI engines in the media mix. Established ad channels take a while to disrupt.
Putting some numbers against that claim, the entire global advertising market is about $1.14 trillion. Segmenting the digital portion of that, we’re talking about roughly $750 billion. Segmenting further, search advertising – the most comparable format to AI-engine advertising – is about $290 billion.
In that light, OpenAI’s projections aren’t terribly ambitious for 2026. But ramping up to $100 billion in five years – a 110 percent compound annual growth rate – is highly questionable. In the market-sizing world, we call this the Heineken method, which entails drinking a 6-pack of Heineken while calculating.
Of course, anything is possible. But for OpenAI to hit this target means taking a large portion of search ad dollars while expanding the overall ad revenue pie with a new format. The latter is hard to do. It ensures greater supply… but aggregate demand (volume of advertisers, budgets, etc) is relatively fixed.
Latest Chapter of the Saga
Back to the part about ups and downs in OpenAI’s ad program, all the above is just the latest chapter of the saga. After the program was announced, it quickly began to stumble with unpopular decisions like limited formats and exorbitant pricing. For example, the campaign spending floor was set at $200K
Similarly, CPMs in the program’s early days were a whopping $60. And CPM is the operative term, as the program only offered impression-based ads. In an ad world increasingly obsessed with attribution metrics – especially for an ad medium based on lower-funnel intent – this saw a swift backlash.
To OpenAI’s credit, it quickly revised some of these early limitations, including lower campaign minimums and performance-based options. And it started to partner with third-party ad-tech players to integrate ChatGPT ads into familiar ad managers, thus meeting brand marketers where they lived.
But those moves weren’t enough to attract ad dollars at the pace that OpenAI had first anticipated. This could sting as the company’s ad revenue is projected to make up $36 percent of its top line, and that will be scrutinized in any upcoming IPO road show. We’ll keep watching to see what it does next.
