ChatGPT Just Turned On Pay-Per-Click

OpenAI has done something that affiliate marketers should have seen coming — and that many are still underestimating. ChatGPT is now an advertising platform. The implications for SaaS affiliate programs run deeper than a new ad slot on a popular website. This is a structural shift in how software gets discovered, evaluated, and bought.

For years, the affiliate channel in SaaS relied on a relatively stable set of assumptions. Publishers wrote reviews. Readers clicked links. Products converted. Commissions followed. The discovery journey was search-driven, click-dependent, and — crucially — trackable.

ChatGPT is in the process of dismantling all three of those assumptions simultaneously. And the pay-per-click rollout is where the disruption gets expensive.

What Has Actually Changed

OpenAI confirmed it would begin displaying advertising inside ChatGPT, initially targeting free and lower-cost tier users in the United States. Criteo became the first major ad-tech company to formally integrate with the platform's advertising pilot in March 2026, operating across ChatGPT's Free and Go tiers.

The ads appear below ChatGPT responses and are labelled as sponsored. OpenAI has stated that paid placements do not influence the answers the platform generates — a structural separation the company will need to defend convincingly as commercial relationships deepen.

For context on how serious the infrastructure is: Criteo brings visibility into more than $1 trillion in annual commerce sales to this integration. This is not an experiment. It is the beginning of a commercial layer built directly into the world's most widely used AI platform.

The Conversion Data That Should Worry SaaS Programs

Here is the detail that affiliate program managers need to sit with. Criteo's data from February 2026, drawn from 500 retail clients, showed users referred from ChatGPT converting at approximately 1.5 times the rate of other referral channels.

That is a striking number. And it creates an immediate budget allocation problem.

If finance teams can see clear attribution data from ChatGPT paid placements — visible, measurable, demonstrably high-converting clicks — and compare that against affiliate channels that are struggling to prove influence in AI-mediated customer journeys, the answer about where to direct budget writes itself. And it is not a comfortable answer for affiliate program managers.

There is an important caveat to those conversion numbers. Criteo's figure comes from a paid advertising context, not organic referral traffic. A user clicking a sponsored placement inside a ChatGPT response has already demonstrated commercial intent at a high point in the conversation — they are not casually browsing. Earlier data across a much larger sample of nearly 1,000 e-commerce sites found ChatGPT referrals converting considerably worse than traditional affiliate links.

The discrepancy matters because it reveals the distinction between paid and organic AI traffic as a performance driver — and it signals where SaaS affiliate programs are most exposed.

Why SaaS Is Differently Exposed

The risk profile for SaaS affiliate programs is not identical to retail. Software buying decisions are research-heavy, comparison-driven, and deeply reliant on third-party credibility. Review content, feature comparisons, use-case breakdowns — this is precisely the content that affiliate publishers have produced for years, and it is exactly the content that language models are trained on and trained to synthesise.

When a prospective buyer asks ChatGPT to recommend a project management tool, an email platform, or a customer support system, the AI draws on publisher content to construct its answer. The affiliate publisher whose review informed that recommendation is invisible to the transaction. No click. No cookie. No commission.

Now layer the paid placement system on top of that. A SaaS brand that buys a sponsored placement inside a ChatGPT response about project management tools is directly competing for the same buyer attention that an affiliate publisher was influencing organically — and doing so with cleaner attribution data, faster feedback loops, and a direct relationship with the platform.

Sam Altman has also publicly discussed a potential affiliate commission model — something approximating a 2% fee on purchases completed through ChatGPT research sessions. If that model moves forward, OpenAI would effectively be positioning itself as both publisher and network, inserting a new intermediary layer between brands and the affiliate programs currently serving them.

The Measurement Problem Is Not Solved

Before SaaS affiliate managers panic and redirect budgets into ChatGPT placements, there is a critical operational reality to understand: the platform's ad product currently provides almost no usable performance data.

There is no automated route to purchase inventory. Deals are negotiated manually. Reporting infrastructure is, by multiple accounts, primitive relative to established performance marketing platforms. OpenAI has reportedly advised early advertisers that performance may improve if they supply more creative variations — a response better suited to an early-stage creative test than a performance channel operating at commercial scale.

This is precisely where the affiliate model retains its structural advantage. When a partner drives a verified sale, the data trail is the product. Advertisers pay for confirmed outcomes rather than modelled impressions or probabilistic attribution. That accountability is the channel's core proposition, and it is the reason brands have been increasing affiliate investment as paid media measurement has become less reliable elsewhere.

The pressure on affiliate programs does not come from paid AI advertising being better. It comes from the direction AI search is taking organic discovery — away from click-based traffic that generates trackable signals, and toward in-platform answer delivery that bypasses the publisher entirely.

Three Actions SaaS Affiliate Programs Should Take Now

• Audit your attribution setup for AI-influenced journeys. Analyse how your current analytics attributes conversions that arrive via direct navigation or branded search after an AI discovery session. A buyer who asked ChatGPT about your product category, heard your brand mentioned, then navigated directly to your site is almost certainly not being credited to any affiliate partner — even if a publisher's review content influenced the AI's recommendation.
• Invest in affiliate publisher visibility inside AI outputs. The publishers who influence AI recommendations are frequently the same publishers managing your most valuable affiliate relationships. Supporting those partners in building genuine content authority — through exclusive data, case studies, and first-party access — directly affects how frequently your brand is cited in AI-generated responses. Organic AI citation and affiliate performance are no longer separate questions.
• Run a limited test on ChatGPT placements before committing at scale. The platform is scaling rapidly while its measurement infrastructure remains underdeveloped. Brands in categories where AI recommendation influence is demonstrably high should monitor how their products appear in ChatGPT responses, but should not treat current paid placement as a mature performance channel. The test stance is appropriate — the full budget commitment is not.

The Bigger Picture

ChatGPT is following a trajectory that anyone who has watched Google or Meta evolve will recognise. Platforms that attract users with a clean, commercially neutral experience eventually build advertising infrastructure to monetise that attention at scale. The pattern is reliable enough that the surprise is not that OpenAI has done this — it is that some programs are still treating it as a future concern rather than a present one.

For SaaS affiliate programs, the strategic response is not to abandon the channel. The affiliate model's accountability advantage — verified outcomes, transparent data trails, commission on confirmed conversions — is exactly what brands need more of as AI-mediated attribution becomes more opaque, not less. The channel's core proposition is more defensible than it appears when you examine what paid AI advertising actually delivers today.

What changes is the work required to protect that position. Discovery is moving. Publishers need support adapting to it. Attribution frameworks need updating. And program managers who act on those three things now will be considerably better positioned when the next version of this infrastructure arrives — because it is already being built.

Want to hear insights like this and more? Affiliate Summit East 2026 is BACK in NYC July 27 – 28... Register your space here today!