The day OpenAI announced ads in ChatGPT, my inbox filled up with the same question: "Should we be advertising on ChatGPT?"
My honest answer to every single one: probably not. Not yet.
But the reason why is more interesting than a simple yes or no. Because the real story here isn't the ads. It's what the data behind them tells us about where search advertising is heading, and why most of the coverage out there is missing the point completely.
What Actually Happened
In February 2026, OpenAI started showing ads to free-tier and Go-tier ChatGPT users in the US. If you're paying for Plus, Pro, or Enterprise, you won't see them. This is a freemium monetisation play.
The format is restrained. Sponsored results show up alongside ChatGPT's conversational answers. No banner ads. No autoplay video. It feels more like a native recommendation than a traditional ad. OpenAI has been clear that ads don't influence ChatGPT's answers, and conversations stay private from advertisers.
Here's where the numbers get uncomfortable for early adopters.
The Pricing Puts This Out of Reach for Most Businesses
Current pricing is $60 CPM with a $200,000 minimum commitment.
Let me put that in perspective. A $60 CPM means $60 for every 1,000 impressions. That's roughly 10x what you'd pay for Google Display Network. About 2-3x premium YouTube placements. And the $200K minimum means this is an enterprise-only channel right now.
If you're a local business spending $3,000 a month on Google Ads, this isn't for you. If you're an agency managing mid-market clients, this still probably isn't for you.
For comparison, Google Search Ads run on CPC where you might pay $2-8 per click depending on industry. ChatGPT's CPM model means you're paying for eyeballs, not actions. That's a fundamentally different value proposition and it's harder to tie back to revenue.
The Reach Problem That's Frustrating Advertisers
This is the part nobody's talking about honestly enough. The rollout has been painfully slow.
At the start of March, ads were reaching roughly 1% of mobile users. That's climbed to about 5% now, but advertisers spending $200K+ are genuinely frustrated with the limited scale. I've spoken to a few media buyers who described it as "paying premium rates for a beta product."
Even with ChatGPT processing over 1 billion queries per day, if only 5% of free-tier mobile users see ads, you're looking at a fraction of that audience. And remember, paying subscribers see no ads at all. OpenAI has over 11 million paying users, so a meaningful chunk of the most engaged, highest-intent users are completely ad-free.
For a $200K minimum buy, that reach limitation really stings.
But the Conversion Data Is Genuinely Compelling
Here's the number that made me stop and rethink everything. According to research from multiple sources tracking AI referral traffic, AI search traffic converts at 14.2% compared to Google's 2.8%. That's a 5x difference.
I've been watching this across client sites that get measurable AI referral traffic, and the pattern holds. People coming from AI chat interfaces tend to be further along in their decision-making process. They've already had a back-and-forth conversation about their problem. They've refined what they need. By the time they click through, they're not browsing. They're ready to act.
It makes sense when you think about it. A Google search for "best CRM for small business" might be someone at the very start of research. A ChatGPT conversation about the same topic usually involves several exchanges where the user narrows down requirements. The click-through is more qualified.
ChatGPT also drives 77% of all AI-driven website referral traffic globally (Press Gazette, March 2026). Perplexity holds about 15%. Everyone else is fighting over scraps. If you're going to invest in AI search visibility, ChatGPT is where the traffic actually is.
What This Actually Means for Your SEO Strategy
The bigger story isn't the ads themselves. It's what they signal about how AI platforms will monetise search behaviour going forward.
Your content needs to work in conversations, not just search results
Traditional SEO matches keywords to pages. AI search answers questions in context. If your content only works as a list of keywords stuffed into H2 tags, it's going to struggle in AI-generated answers, whether those answers are organic or sponsored.
I've been restructuring client content to answer specific questions directly, with clear data points and concrete examples. Pages that do this are showing up in AI citations more frequently.
Brand visibility in AI answers matters more than ever
When ChatGPT recommends your product in an organic answer, that's incredibly valuable. Now that paid placements exist alongside organic mentions, there's going to be a land grab for visibility. Brands that are already well-represented in AI training data have a head start.
The line between paid and organic is getting blurry
On Google, you can see the "Sponsored" label clearly. In a conversational AI interface, the distinction between a genuine recommendation and a paid placement is much harder to spot. OpenAI says ads are "clearly labeled as sponsored content, separated visually from organic responses." But in practice, inside a flowing conversation, that separation is less obvious than a search results page.
What I'd Actually Do Right Now
If you're spending under $20K/month on ads
Ignore ChatGPT ads completely. The $200K minimum is a non-starter. Focus on Google Ads and Meta where you have proven, measurable channels.
What you should do instead: make sure your content is structured to appear in AI-generated answers organically. That's free, and it's where the real opportunity is.
If you're spending $50-200K/month on ads
Watch closely but don't move budget yet. The 5% reach to free-tier mobile users isn't enough to justify pulling from channels that are working. But start tracking your AI referral traffic now. Set up proper UTM parameters for AI sources. Most businesses I audit have no idea how much traffic they're getting from ChatGPT because they haven't set up tracking.
If you're enterprise spending $200K+ monthly
Consider a test budget. You can afford the minimum, and there's genuine first-mover advantage in understanding a new channel before competitors. The 14.2% conversion rate suggests unit economics could work even at $60 CPM.
But go in clear-eyed. Demand transparent reporting on actual reach and click-through rates. Don't let novelty override basic media buying discipline.
For everyone, regardless of budget
Invest in AI search visibility through your content:
- Answer specific questions directly in your content, not just target keywords
- Include concrete data and numbers that AI models can cite
- Build topical authority across multiple pages
- Check your robots.txt to make sure AI training bots can crawl your site
- Monitor your brand mentions in ChatGPT and Perplexity regularly
Where This Is Heading
OpenAI needs revenue to justify its valuation. Advertising is the most proven business model in tech. The current limitations, slow rollout, high minimums, limited reach, are growing pains. Not permanent features.
I'd expect the minimum commitment to drop within 6-12 months as they build self-serve tools. Reach will expand. Targeting will get more sophisticated based on conversation context.
The bigger shift is that search advertising is no longer just a Google and Bing conversation. AI platforms are becoming a meaningful source of high-intent traffic, and where traffic goes, ad dollars follow.
The skills that help you rank in traditional search, creating genuinely useful content, building authority, understanding what people actually need, are the same skills that will help you succeed in AI search. The format is changing. The fundamentals aren't.
Start tracking your AI referral traffic today. That data will be invaluable when this channel matures. And it will. A billion daily queries and a 14.2% conversion rate are too compelling for advertisers to ignore, even with the current growing pains.
