Originally published on The Searchless Journal
The Information published screenshots on April 21, 2026, showing OpenAI's ads manager with cost-per-click bidding enabled. The range: $3 to $5 per click.
That is not a test. That is a price tag.
Ten weeks earlier, ChatGPT ads launched at $60 CPM. Today, CPMs sit in the $15-45 range. Minimum spend dropped from $250,000 to $50,000. OpenAI is hiring its first advertising marketing science leader. The company projects $2.5 billion in ad revenue for 2026, $11 billion in 2027, and $100 billion by 2030.
The signal is clear. ChatGPT advertising is no longer an experiment. It is a performance media platform in the making.
But the deeper story is what happens after the click. ChatGPT users are in a conversational flow, not a search-intent funnel. OpenAI's clicks are not Google's clicks. The measurement gap—whether a conversational click is worth as much as a search click—will determine whether AI-native advertising becomes a real budget category or stays a test budget line item.
This is what the CPC launch means, why the pricing model matters, and what advertisers should actually do.
The CPC Launch: What Changed and Why It Matters
The Information's report with verified screenshots is the most important signal in ChatGPT advertising since the February launch. The screenshots show an ads manager interface where advertisers can set CPC bids between $3 and $5 per click.
Digiday confirmed the shift on April 21, adding context from agency executives. Gartner's analyst noted that CPC pricing "will allow advertisers to directly compare ChatGPT's results to other platforms" in a way that CPM-only models could not. That comparison capability is the bridge between experimental spend and performance-eligible budgets.
The timing is strategic. ChatGPT ads launched February 9, 2026, as CPM-only inventory at $60. By mid-April, CPMs had compressed to $15-45 depending on category and demand. ExchangeWire reported that price drop on April 20, citing data from Jellyfish and Criteo. Minimum spend requirements dropped from $250,000 to $50,000, opening the platform to mid-market advertisers.
Impression volumes tell the same story. Adweek reported on April 20 that ChatGPT ad impressions increased approximately 600% from early March to mid-March. More inventory, lower CPMs, accessible minimums, and now CPC pricing. That is the formula for scale.
OpenAI expanded ChatGPT ads internationally on April 16-17 to Australia, New Zealand, and Canada. The company now has 900 million weekly active users. The platform is building toward a performance advertising system.
The CPC launch completes the infrastructure loop. CPM was fine for brand experimentation. CPC is required for performance budgets. The conversion tracking pixel—reported by Digiday on April 16—means advertisers can connect clicks to actions. The missing piece is the marketing science layer, which is why the new marketing science leader hiring is so important.
The Revenue Projections: $2.5B in 2026, $100B by 2030
Reuters and Axios reported on April 17 that OpenAI projects $2.5 billion in ad revenue for 2026, $11 billion in 2027, and $100 billion by 2030. Those numbers are not guesses. They come from investor presentations and internal roadmaps.
The $2.5 billion 2026 target means OpenAI expects advertising to become a significant revenue stream this year, not next. The $100 billion 2030 projection is more ambitious—equal to what Alphabet's Google advertising business generates today.
That projection tells you how OpenAI thinks about ChatGPT advertising. This is not a side project. This is a core revenue pillar.
The revenue targets explain the infrastructure investments. CPC pricing, conversion tracking, ads manager improvements, international expansion, and now a marketing science leader—all of these are the building blocks of a $100 billion ad platform.
The StackAdapt partnership adds another signal. Adweek reported on April 20 that StackAdapt is selling ChatGPT placements through its platform, using "prompt relevance" as the targeting mechanism. The leaked deck shows CPMs ranging from $15 for niche prompts to $60 for competitive categories. That partnership expands ChatGPT's distribution reach and gives OpenAI access to programmatic buying infrastructure.
Omnicom has more than 30 clients running ChatGPT ads. Dentsu reports rising weekly volume. Agencies are testing, learning, and scaling. The early adopter phase is ending.
The Pricing Context: Where $3-5 CPC Sits in the Market
Adthena's data on search advertising provides useful context. Meta's CPCs typically run 3-5x cheaper than Google Search because the intent gap is lower. Google Search users have explicit intent. Meta users are browsing. The price difference reflects conversion probability.
ChatGPT's $3-5 CPC range sits between Meta and Google Search. That positioning is deliberate. If CPCs were $0.50-1.50 like Meta, advertisers would expect Meta-like conversion rates. If CPCs were $5-15 like Google Search for competitive terms, advertisers would expect Google-like intent.
The $3-5 range suggests OpenAI is pricing for mid-funnel intent—users who are researching, considering, and comparing, not necessarily ready to transact immediately. That matches the conversational nature of ChatGPT usage. People ask questions, explore options, and gather information. Some convert immediately. Many do not.
The CPM compression from $60 to $15-45 tells another story. Higher supply and more advertisers entering the pilot drove prices down. That is healthy market behavior. It suggests ChatGPT has more inventory than early buyers could absorb, which means the platform can scale without hitting inventory constraints.
The minimum spend drop from $250K to $50K is equally important. That threshold change expands the addressable market from enterprise-only to mid-market brands. More advertisers means more demand, which should support stable CPCs as impression volumes grow.
The Marketing Science Hire: Why Attribution Is the Next Battleground
OpenAI's job posting for its first advertising marketing science leader reveals where the company is focusing next. The role includes attribution modeling, incrementality testing, media mix modeling, and geo experimentation.
That job description is the blueprint for proving that ChatGPT clicks are worth buying.
Attribution modeling will help advertisers understand which ChatGPT interactions drive conversions. Is it the first touch in a research journey? The last touch before purchase? An assist touch that matters but does not get credit? ChatGPT's conversational interface means users might engage multiple times across days before converting. Last-click attribution will undervalue that. Multi-touch or data-driven attribution will be required.
Incrementality testing is the gold standard for measuring lift. The question is not "what percentage of ChatGPT-referred visitors converted?" The question is "how many more conversions happened because of ChatGPT ads versus if those ads had not run?" That requires holdout tests, control groups, and statistical rigor. OpenAI is building that capability.
Media mix modeling helps advertisers understand ChatGPT's role in the broader media ecosystem. Should budget shift from Google Search to ChatGPT? From Meta to ChatGPT? From programmatic display to ChatGPT? The models will show which combinations drive the best outcomes.
Geo experimentation allows for market-level testing. Run ChatGPT ads in one metro area, hold out in another, compare results. That is how Meta proved the value of its ad platform early on. OpenAI is following the same playbook.
The marketing science hire is the signal that OpenAI understands the measurement gap. The company knows that performance advertisers will not scale spend without confidence in measurement. The infrastructure is being built.
The Measurement Gap: Conversational Clicks vs Search Clicks
The fundamental question is whether a conversational click is worth as much as a search click.
Google Search users type explicit queries. "Best running shoes for flat feet." "CPA near me." The intent is clear. The user is actively looking for something. A click from a search result has a high probability of converting because the user signaled intent before clicking.
ChatGPT users ask questions in conversation. "I'm training for a marathon and my feet hurt. What should I do?" That query might lead to a discussion about shoes, stretching, training plans, or medical advice. A sponsored link to running shoes might appear. The user might click. But the intent is less transactional. The user was exploring a problem, not explicitly shopping for shoes.
That intent difference matters for conversion rates. Search clicks typically convert at higher rates than conversational clicks because the starting point is more purchase-oriented. OpenAI's $3-5 CPC pricing reflects that. The platform is not charging Google Search prices because the intent is not the same.
But the counter-argument is that conversational discovery happens earlier in the funnel. A user asking about foot pain during marathon training might not have searched for running shoes yet. The ChatGPT ad introduces the option. That creates demand that did not exist. The value is not just in immediate conversion. It is in top-of-funnel awareness and consideration.
The challenge is measuring that value. If a user sees a ChatGPT ad for running shoes, clicks, browses, and leaves without buying, then searches for "best running shoes" two days later and purchases from a different brand, how does ChatGPT get credit? Standard last-click attribution gives ChatGPT nothing. Data-driven attribution might give it partial credit. Incrementality testing would show whether ChatGPT ads increased total conversions in the market.
This is why the marketing science hire is so important. OpenAI needs to build the measurement framework that proves the value of conversational clicks. If the company can demonstrate that ChatGPT ads drive incremental conversions—not just the same conversions that would have happened through search anyway—then performance budgets will follow.
What Advertisers Should Do Now
The CPC launch changes the calculus for performance advertisers. CPM-only inventory was experimental. CPC inventory is performance-eligible. Here is what to do.
Test with a clear hypothesis. Do not just "try ChatGPT ads" and see what happens. Define what success looks like. Is it CPA below a threshold? ROAS above a benchmark? Incremental lift in a holdout test? The hypothesis should be specific, measurable, and tied to business outcomes.
Isolate the funnel stage. ChatGPT's strength is in consideration and research, not bottom-funnel transaction. Structure campaigns around mid-funnel goals—education, comparison, consideration—rather than demanding immediate purchase conversions. Measure engagement metrics like time on site, pages per session, and add-to-cart rate alongside conversion rate.
Use attribution that captures assisted value. Last-click attribution will underreport ChatGPT's impact. Implement data-driven attribution or multi-touch attribution if your analytics platform supports it. Track view-through conversions. Measure the full customer journey, not just the last click.
Compare across platforms, not in isolation. Run parallel campaigns on Google Search, Meta, and ChatGPT for the same product or offer. Compare CPA, ROAS, and incremental lift. The goal is not to determine whether ChatGPT is "better" than Google. The goal is to understand where ChatGPT fits in the media mix and what role it plays.
Start small, scale what works. The $50,000 minimum spend is accessible but still significant for testing. Begin with a focused test—single category, clear audience, defined time window. Use the results to inform a broader strategy. The platform is evolving rapidly. What works in April 2026 might not work in July 2026. Treat this as an ongoing learning process, not a one-time test.
Watch the measurement roadmap. OpenAI is building attribution, incrementality, and media mix modeling capabilities. Those tools will make ChatGPT ads more measurable and more performance-eligible over time. Early adopters who learn the platform now will have an advantage when the measurement infrastructure matures.
The Bigger Picture: AI-Native Advertising as a Category
ChatGPT's CPC launch is not just about one platform. It is a signal that AI-native advertising is becoming a real category.
Google is building AI Overviews ads. Perplexity is experimenting with sponsored commerce units. Meta is integrating generative AI into its ad targeting and creative. The $100 billion revenue projection for ChatGPT advertising by 2030 suggests OpenAI believes AI-native advertising will rival traditional search and social advertising within five years.
The defining characteristic of AI-native advertising is that ads appear inside AI-mediated interactions. The user is not searching a result page. The user is not browsing a feed. The user is in conversation with an AI. The ad is a recommendation within that conversation.
That context changes everything. Targeting works differently. Measurement works differently. Creative works differently. The entire discipline of performance marketing has to adapt.
The CPC launch is the first step in that adaptation. Pricing moves from impressions to clicks. Measurement moves from brand metrics to performance metrics. Budget moves from experimental spend to core media allocation.
The measurement gap is the remaining hurdle. OpenAI is building the infrastructure to close it. The marketing science hire, the attribution models, the incrementality tests—those are the tools that will prove conversational clicks have value.
If OpenAI succeeds, AI-native advertising becomes a permanent budget line item. If the company fails to prove the value of its clicks, ChatGPT ads stay a niche experiment.
The CPC launch is the pivot point between those two futures.
Get an AI Visibility Audit to measure how your brand appears across ChatGPT, Perplexity, Gemini, and Google AI Overviews.
Sources
- The Information, "OpenAI turns on cost-per-click ads inside ChatGPT" (April 21, 2026) – Verified screenshots of ads manager CPC interface with $3-5 bid range
- Digiday, "OpenAI's ChatGPT ads get CPC pricing as it pushes for ad revenue" (April 21, 2026) – Agency executive quotes, Gartner analyst commentary on CPC comparison value
- Adweek, "StackAdapt to sell ChatGPT ad placements" (April 20, 2026) – Leaked deck showing prompt relevance targeting, CPM ranges from $15-60
- ExchangeWire, "ChatGPT ad CPMs drop from $60 to $15-45" (April 20, 2026) – CPM price compression data from Jellyfish and Criteo
- Reuters/Axios, "OpenAI projects $2.5B ad revenue in 2026, $100B by 2030" (April 17, 2026) – Revenue projections from investor presentations
- The Keyword, "ChatGPT ads expand to Australia, New Zealand, Canada" (April 20, 2026) – International expansion details, 900M weekly active users
- OpenAI ChatGPT release notes, Official documentation of ads features and expansion (April 16-17, 2026)
- PYMNTS, "OpenAI begins offering cost-per-click ad campaigns on ChatGPT" (April 21, 2026) – CPC summary with Ad Context Protocol context
- Adthena, "Meta CPCs vs Google Search CPCs" (Industry benchmark data, 2026)
Top comments (0)