On May 5, 2026, OpenAI quietly did something that should have made bigger headlines: they removed the $50,000 minimum spend on the ChatGPT ads platform and opened a self-serve Ads Manager to any US business with a credit card. For the first time, an SMB running a $2,000/month ad budget can buy inventory in the same conversational AI surface where 800 million weekly users are asking for product recommendations.
I've been waiting on this since the pilot launched in 2025. Here's the actual landscape, who should test it now, who should wait, and how to spend the first $500 without burning it.
What just changed
For context — when ChatGPT ads launched in late 2025, the minimum spend was $200,000. By February 2026 that dropped to $50,000. As of May 5, the minimum is zero. You can spin up an Ads Manager account, set a $20/day budget, and start running cost-per-click campaigns inside ChatGPT conversations.
Three things that matter about the platform's current state:
- CPC bidding is live. You're not paying per impression. You pay when someone clicks. Same model SMBs have used on Google for 20 years.
- Conversions API + pixel tracking are in beta. You can attribute purchases, leads, or signups back to ad clicks. This is the bare minimum for performance accountability, and it works now.
- The ad format is text-with-favicon for service businesses, structured cards for products. Native to the conversation, clearly labeled, but visually less disruptive than a Google sidebar ad.
OpenAI is targeting $2.5 billion in ad revenue this year — meaning there's massive volume of advertiser learning happening right now, but the auction is still relatively uncrowded compared to mature platforms like Google Ads.
Why this is different from every other ad platform you've tested
The fundamental difference is the moment of interception. Google Ads catches someone searching for an answer. Meta ads catch someone scrolling past content. ChatGPT ads catch someone actively asking for a recommendation — and the surrounding context is the AI's genuine attempt to answer them.
When a user asks ChatGPT "what's a good project management tool for a small marketing team?", and your sponsored placement appears alongside the AI's organic recommendations, you're not interrupting attention. You're inside it. The user isn't bouncing past your ad on the way to something else; they asked the question your ad is answering.
This is a higher-intent ad surface than anything currently in the SMB stack. Probably the highest-intent ad surface that has ever existed, when you set aside outbound sales calls.
Who should test it now
The honest filter:
Test now if:
- You sell a product or service that's frequently the answer to "what should I use for [X]?" — i.e., your category is something people ask AI about
- You already have conversion tracking properly wired (Conversions API capability matters here)
- You have $500-2,000 in budget you can dedicate to platform learning, separate from your core media buys
- You're operating in the US (international rollout to UK, Mexico, Brazil, Japan, South Korea is in progress)
Wait if:
- Your category is rarely AI-queried. "Local plumber" or "Italian restaurant near me" still skews toward Google Maps queries, not ChatGPT.
- Your conversion tracking is broken or partial. The platform's audit data is your only feedback loop; if you can't see what's working, you can't optimize.
- Your existing Google + Meta is leaving meaningful pipeline on the table. Fix the proven channels before adding an experimental one.
- You can't afford to lose the first $500-2,000. Every new platform has a learning tax. Don't pay it with money you need.
How to spend the first $500 without burning it
Week 1 — Setup ($0 spend)
Sign up for Ads Manager. Wire the Conversions API to your CRM or e-commerce backend. Decide on 2-3 high-intent queries you want to show up for. Write 4 versions of your ad copy (we keep them short — text-based units don't reward verbosity).
Weeks 2-3 — Learn ($300 spend)
Set a $20/day budget across both campaigns. Don't optimize aggressively in this window — the algorithm needs ~30 conversion events before it learns who converts. Resist the urge to fiddle. Watch which queries trigger your placements and what the resulting click quality looks like.
Week 4 — Decision ($200 spend)
By the end of week 4 you should have enough data to know: is the CPL competitive with your Google Ads CPL, or is it 3-5x higher? Is the lead quality similar to Google high-intent search traffic, or markedly worse? Cut what's not working. Scale what is.
The point of the first $500 is not to acquire customers. It's to learn whether the channel converts for your specific category, at your specific price point. Most categories will work; some won't. The only way to know is to ship the test.
What I'm watching for the rest of 2026
Three things that will determine whether ChatGPT ads become a permanent fixture or a temporary novelty:
Auction competitiveness. CPCs are reportedly lower than mature Google search terms right now. That will change as more advertisers enter. The window where this is a meaningful arbitrage opportunity probably closes in 6-12 months.
OpenAI's commitment to organic recommendations. OpenAI says ads will not influence ChatGPT's organic outputs. If that holds — if the AI's genuine recommendation engine remains separate from its monetization engine — the platform stays trustworthy and the ads stay valuable. If it doesn't, both crash together.
Attribution quality. Conversations are private. OpenAI says they only share aggregated data, not the content of individual chats. That means the data flowing back to advertisers will always be looser than Google's click-level data. Performance marketers used to granular reporting will need to adjust.
The honest part
I'm running a test budget on three of our client accounts as of this week. Two are in B2B SaaS (high-consideration, high-LTV — exactly the profile that should work), one is in financial services (compliance-heavy, longer sell cycle — harder fit). I'll have meaningful data in 30 days and I'll write about the results, both the good and the bad.
If you want help thinking through whether your category is a good fit, or whether your tracking is wired well enough to actually measure the test, our free Marketing Score can speak to both. A senior strategist will assess your conversion tracking integrity and whether your business is the kind that comes up in AI-queried decisions. Hand-audited. Emailed within 1 business day.