ChatGPT Ads and the $200k Minimum: What Replaced It

For the first two months of the ChatGPT Ads pilot, the $200K number did more work than any other piece of the story. It was the gatekeeper. Adthena's CEO told Digiday in early February that "the only clients we've heard that have been approached are the biggest of the big." Everything else about the pilot (the $60 CPM, the conversation-context targeting, the managed service) filtered through that minimum.

Then on April 10, 2026, Digiday reported that OpenAI had quietly launched a self-serve Ads Manager and cut the minimum to $50,000. The wall came down. This post is the retrospective on what the $200k gate did to the market, who got in under it, and what replaced it. For the current state of the self-serve pilot, see our dedicated breakdown: ChatGPT Ads Are Now Self-Serve: Minimum Spend Dropped to $50k.

What Was the $200k Minimum?

OpenAI confirmed the $200,000 minimum commitment to Adweek in early 2026. That number was the floor, not a fixed price. In practice, OpenAI's sales team pitched different minimums to different advertisers based on brand size, category, and existing relationships:

• $1,000,000 for the original beta wave (January 2026 pre-launch conversations)
• $250,000 commonly pitched to agencies
• $200,000 as the officially confirmed floor (Adweek)
• $100,000 to $125,000 reported for a small number of advertisers with direct relationships

The variation told you something important. It was not a price list. It was a negotiation. OpenAI was hand-selecting advertisers for a controlled beta, not running an open marketplace. Every deal was custom, and the $200k headline was the low end of the public conversation, not the average.

At the launch $60 CPM, $200k bought roughly 3.33 million impressions across ChatGPT's US Free and Go tier users. That is a significant but not massive buy against a user base of 900 million weekly active users globally. What the spend did not buy was everything performance marketers normally expect: no self-serve dashboard, no conversion pixel, no A/B testing, no post-view attribution. Advertisers received weekly CSV reports. That is it.

Who Got In Under the $200k Gate?

Adweek's February 6 reporting named the launch cohort that cleared the minimum in the first wave: Target, Ford, Mrs. Meyer's, Adobe, plus the agency triad of WPP Media, Omnicom, and Dentsu. Those were the brands OpenAI wanted in the room for the initial pilot, anchoring the platform's price and prestige positioning.

Adthena's independent analysis of 500+ ChatGPT prompts in mid-February identified the next wave of confirmed advertisers through observed sightings: Expedia, Qualcomm, Best Buy, Enterprise Mobility, and The Knot Worldwide. By late March, CNBC was reporting that the pilot had grown to over 600 advertisers, with brands like Williams-Sonoma joining the roster.

What the launch cohort had in common was not just six-figure paid media budgets. It was the ability to absorb a $200,000 commitment with negligible measurement infrastructure. These were brands with enough existing attribution tooling, enough category-level intuition, and enough willingness to pay a premium for being first in a new channel. The $200k gate did a specific thing: it selected for advertisers who could think of the spend as a learning budget, not a performance budget.

Why OpenAI Chose $200k

The minimum was not arbitrary. Three forces converged on the $200k number.

Supply scarcity. At launch, 85% of ChatGPT users were eligible to see ads but less than 20% did on any given day. Adthena later found ads appearing in just 0.8% of 500+ analyzed prompts. That deliberate scarcity meant the available inventory was limited, and premium pricing was how OpenAI rationed it without an auction.

Managed-service overhead. Every launch-cohort campaign routed through OpenAI's direct sales team. Without self-serve infrastructure, each advertiser required handholding, custom creative review, and bespoke reporting. A $200k floor made the unit economics of that overhead work. At smaller spend levels, the service cost would have eaten the margin.

Premium positioning. The $60 launch CPM placed ChatGPT in the same zone as Netflix's 2022 ad-tier launch ($55 to $65 CPM) and above LinkedIn's B2B rate. $200k was the matching spend floor for that CPM band. You do not anchor an ad platform alongside premium streaming and then let mid-market advertisers in at $10k. The gate was load-bearing for the positioning.

What the Wall Did to the Market

The $200k minimum shaped behavior well beyond the launch cohort itself. Three secondary effects emerged in the nine weeks before the wall came down.

Agency pooling became a primary access path. Holding companies with direct OpenAI relationships, particularly Omnicom, WPP, and Dentsu, started aggregating client budgets to clear the minimum. Agencies with enough clients willing to contribute $20,000 to $30,000 each could hit $200k collectively, then allocate impressions across the portfolio. That workaround was invisible to brands but widely used inside agency relationships.

Criteo emerged as an ad tech alternative. On March 2, 2026, OpenAI named Criteo its first ad tech partner. Criteo manages roughly $4 billion in annual media spend across more than 17,000 advertisers, and the integration let mid-market brands access ChatGPT inventory through an existing programmatic relationship rather than a direct OpenAI commitment. On March 31, Criteo expanded the GO platform to SMBs in the US and UK. That was the moment the $200k filter effectively stopped being the only gate.

Mid-market brands waited. For businesses without agency relationships or Criteo integrations, the pre-April 13 posture was essentially to wait. The trade press told that story clearly: register interest at openai.com/advertisers, build organic AI visibility in the meantime, and assume the self-serve platform would eventually open access at a lower minimum. That prediction turned out to be right about the direction and dramatically right about the timeline. Every major analyst expected self-serve in late 2026 or 2027. It arrived in April.

What Replaced the $200k on April 13

Digiday reported on April 10, 2026 that OpenAI had quietly launched a self-serve Ads Manager to a subset of advertisers. Three days later, the minimum spend formally dropped to $50,000. eMarketer described the change as OpenAI "slashing" the entry threshold. The wall did not shrink. It collapsed.

Four things happened at the same time, and together they represent a structural reset of who the pilot is for:

• Self-serve Ads Manager went live. The dashboard is an early version with impressions and clicks in real time. CPC bidding shipped on April 21. The OpenAI conversion pixel and a server-side Conversions API then launched broadly on May 5, 2026 as a self-serve beta, tracking 10 events with a 30-day attribution window. View-through attribution remains in development; cost-per-action bidding reaches early access on June 5, 2026. Daily budgets and granular geo-targeting shipped on May 22, 2026.
• Minimum dropped to $50,000. That was the new floor for pilot participation between April 13 and May 5, down from $200,000 to $250,000.
• CPM began compressing. Digiday reported April 17 that advertisers in Criteo's channel were seeing $25 to $35 CPMs, with Jellyfish's Jai Amin reporting averages closer to $45 in negotiated deals.
• Geography expanded. Canada, Australia, and New Zealand were added on March 26, opening new auction pressure.

On May 5, 2026, the access reset went one step further: OpenAI opened the Ads Manager to every US advertiser and removed the $50,000 minimum entirely. The 86-day journey from invite-only enterprise pilot to open self-serve completed faster than any major digital ad platform on record. Every business with a US billing address can now launch a campaign at ads.openai.com without clearing a spend floor.

For the full news story on the original April changes, see ChatGPT Ads Are Now Self-Serve: Minimum Spend Dropped to $50k. For the May 5 minimum-removal news, see ChatGPT Ads Just Opened to Every US Advertiser. For the current ROI math and budget scenarios, see How Much Do ChatGPT Ads Cost?.

The Access Barrier Before and After April 13

The simplest way to see the shift is side by side. Every material dimension of access changed in the same window.

Two observations about this table. First, it is not just the minimum that dropped. The access model, the ad tech pipeline, the geography, the dashboard, and the CPM all changed in the same nine-week window. The $200k number is the headline change, but it is the combination of these shifts that materially rewrites the pilot's ICP.

Second, the "change" column is directionally consistent: every dimension moved toward wider access, faster measurement, and lower prices. There is no countervailing trend. OpenAI is pushing in one direction across every meaningful dimension of the platform, which is the strongest signal yet that the April 13 reset was planned, not reactive.

What It Means for Advertisers Who Were Priced Out

If you were priced out of ChatGPT Ads before April 13, you now have three decisions to make. The May 5 minimum removal pushed the access threshold to zero, so the question is no longer "can I afford the floor" but "what is the right test budget given my product economics."

Decision one: what test budget makes sense for your business? With no platform-imposed minimum as of May 5, 2026, $1K-$5K reads directional signal and $25K-$50K reads creative-level insight. If you are already running a mid-size paid social program, a $25K-$50K pilot fits naturally. The next question is whether you have the creative operations and measurement discipline to learn from a channel whose attribution infrastructure is still incomplete. We built a readiness framework in the Phase 1 news post that counts the boxes.

Decision two: is your brand already visible in AI answers organically? A brand that is invisible in ChatGPT's organic responses is buying cold impressions at paid CPMs. A brand that already appears in ChatGPT's recommendations or Google AI Overviews is buying a multiplier on an earned signal. Organic AI visibility is the foundation that makes paid work harder. Build it first, then layer ads on top.

Decision three: is your product economics suited to conversational intent? Criteo's February 2026 analysis of 500 US retailers found that LLM-referred users convert at 1.5x the rate of other channels. That lifts the ROI math for high-intent purchase categories (insurance, travel, retail, SaaS). For low-margin consumer products, the current pricing is still likely too expensive until measurement catches up.

The honest framing for May 2026: any first ChatGPT Ads test is a learning budget, not a performance budget. You are buying early signal and creative literacy in a channel whose unit economics are still being discovered. Advertisers who can absorb that for one or two quarters will know the channel before measurement matures. Advertisers who need last-click ROAS to justify every dollar should wait.

What Should You Do Right Now?

• Read the May 5 news breakdown for what changed when OpenAI removed the minimum and opened the platform to every US advertiser
• Read the news breakdown on the April 10 early-access launch and the $50k minimum for the original timeline, readiness framework, and primary sources
• See current ChatGPT Ads cost math for budget scenarios across the $25 to $60 CPM range
• Run your free AI Visibility check to see if you are showing up in ChatGPT organically before spending on ads
• Preview your creative with the free ChatGPT Ads Mockup Generator before committing to a pilot
• Read the complete guide to ChatGPT Ads for the full platform picture
• If you are deciding where to put paid budget, see how ChatGPT Ads compare to Google Ads in 2026

Frequently Asked Questions