Obviously, OpenAI will monetize with ads

Last week, OpenAI announced that it has hired Fidji Simo, the CEO of Instacart, to become the CEO of its applications division. From Reuters:

Instacart head Fidji Simo will step down to join OpenAI as its chief of applications, CEO and co-founder Sam Altman said on Thursday, adding she will report to him and that he will remain head of the Microsoft-backed ChatGPT maker … The announcement comes two days after OpenAI dialed back a significant restructuring plan, with its nonprofit parent retaining control in a move that is likely to limit Altman’s power.

The news is noteworthy for a few reasons. The first is that Instacart is a successful business: its stock is up 34% on a year-over-year basis, and the company beat analyst expectations on both revenue and earnings in Q1 2025. While OpenAI recently raised money at a $300BN valuation — compared to Instacart’s $11BN market capitalization — it is nonetheless unusual for the CEO of a publicly traded company to decamp for a non-CEO role in a company operating in a completely unrelated category.

The second is that it indicates that the business models of Instacart and OpenAI are likely to become much more similar. As I remarked when Instacart filed its S-1, Instacart is an advertising and subscription business: the company’s advertising revenue growth outpaced that of transactions in Q1 2025 (14% vs. 8%, respectively), with transaction revenue capturing the Instacart+ subscription business. Instacart operates a $1BN advertising business that likely contributed disproportionately to the company’s $106MM of net income in Q1.

Fidji Simo was recruited to lead Instacart from Meta, where she served as the general manager of the Facebook app and reported directly to Mark Zuckerberg. But more appreciable than having run the Blue app at the end of her tenure with the company, Simo is credited with having constructed the company’s revenue engine: ads in the mobile News feed. As Antonio Garcia Martinez, a former product manager on the ads team at Meta (then, Facebook) remarks in his memoir, Chaos Monkeys (emphasis mine):

For all the Sturm and Drang, Facebook’s quarter-saving gold mine, the thing that catalyzed the stock out of the post-IPO doldrums, wasn’t Custom Audiences or FBX. A third product, the only other novel ads product Facebook launched during its harried IPO period, code-named “Neko,” was that savior. The product itself, like so many, was simply the combination of two otherwise disparate domains: Facebook’s ever-addictive News Feed and ads inventory on the Facebook mobile app, instead of the desktop site. That’s it: ads in News Feed, while the user was on his or her mobile device—that’s what saved Facebook. The person most responsible for this coup was a product manager with the improbable name of Fidji Simo. She was one of two office wives of mine (yes, I was an office Mormon), who had started her career in Facebook Ads as a lowly product marketer. She had very quickly and skillfully navigated herself up the Facebook corporate ladder, landing herself the product manager job where Ads and the rest of the company overlapped, placing herself in the larger Facebook (and Zuck) spotlight.

Back in March, in an interview with Ben Thompson, Sam Altman, OpenAI’s CEO, remarked that the company wasn’t currently seriously considering advertising as a monetization strategy:

Currently, I am more excited to figure out how we can charge people a lot of money for a really great automated software engineer or other kind of agent than I am making some number of dimes with an advertising based model.

But OpenAI was almost certainly in the last stages of negotiations with Simo at the time of the interview. And it seems improbable — frankly, inconceivable — that Altman could simultaneously believe that advertising represented an opportunity sized at “some number of dimes” and that Simo, who deployed one of the most lucrative advertising products ever conceived, was the best person to run OpenAI’s applications division. Those two notions are incompatible.

Obviously, OpenAI will monetize with ads. Hiring Simo represents such an on-the-nose acknowledgment of that fact that I almost didn’t write this piece. Except that OpenAI’s admission that advertising is its path forward on monetization serves to dispel a common misconception (really, a fallacious, superstitious tech dogma) that advertising is but one of many monetization strategies that are all equally capable of achieving optimal revenue for scaled consumer technology products. This isn’t true. If maximizing revenue is an organization’s objective function, and its product can potentially reach a scale of billions of users, then advertising stands alone as its optimal monetization strategy.

This is the crux of the freemium model: a product with potentially billions of users should be given away for free in order to reach them. Attention is an asset. The product use cases capable of agglomerating the attention of hundreds of millions or billions of people are, almost as a bi-product of their accessibility and broad breadth, likely not the ones best positioned to derive maximum value from that attention.

Luckily, advertising exists. Firms can bid on that attention: not in the aggregate, for an entire product’s user base, but for individuals, based on what is known about them. The asset of attention can be atomized down to a single user and sold to the product that can best monetize it. Advertising is a matching market that allocates attention to the firm positioned to best activate it. I discuss this idea in Digital Advertising, Demand Routing, and the Millionaires’ Mall:

One challenge with optimizing for conversions is that, in an ecosystem as vast and mostly heterogeneous as the internet — even in the context of specific, scaled products — the presence at any given moment of a user that 1) has an interest in some product and 2) possesses the disposable income to purchase that product is rare. Digital advertising doesn’t create consumer demand; rather, digital advertising should seek to route existing demand to the products that best serve it. An efficient digital advertising channel matches consumer demand for a product with the most satisfying and fulfilling variant of that product. An advertising channel is not a demand factory but a demand highway: the more efficiently an ad channel can route consumer demand to products, the more economic value it produces.

OpenAI will introduce advertising. It has built a generational product, ChatGPT, that has succeeded in aggregating attention at a rarefied scale: the company revealed in April that it has 500MM weekly users (WAU). But just 4% of those users are subscribers. The remainder will be monetized with ads. This is not a design decision; it’s an optimization exercise and a commercial necessity.