ChatGPT ads vs Perplexity ads: where to spend first
ChatGPT runs 89,133 captured placements across 108 niches and 2,655 advertisers. Perplexity sells a smaller, question-led format. Here is how to choose.
As of July 2026, ChatGPT runs the larger and more mature AI answer-engine ad program, with 2,655 advertisers buying 89,133 placements across 108 niches. Perplexity sells a smaller, question-led format routed mostly through direct sales. Marketers usually start on ChatGPT for reach, then layer Perplexity for niche question audiences.
What are ChatGPT ads?
ChatGPT began showing sponsored recommendations inside answer threads in 2025. The ads look like a short labeled block, often tagged 'Sponsored', that sits beneath or beside an assistant's response. They are bought through OpenAI's own ads manager, priced on a cost-per-click or impression basis, and shown only on queries where OpenAI believes a product or service could be useful to the user. Because the ad sits next to a generated answer rather than a list of blue links, the click is usually from a reader who has already seen the AI's recommendation.
What are Perplexity ads?
Perplexity sells a different kind of inventory. Its main paid unit is a 'Sponsored' follow-up question that appears under an answer, inviting the reader to ask a related brand-led query. Perplexity also runs display-style media buys on adjacent publisher inventory. The format is built for question-led discovery: a reader asks something, gets an answer, then sees a brand-paid follow-up. Broad self-serve buying is still limited, and most spend moves through direct sales.
ChatGPT vs Perplexity at a glance
| Dimension | ChatGPT | Perplexity |
|---|---|---|
| Inventory size (captured) | 89,133 placements | Smaller, direct-sales-led |
| Distinct advertisers | 2,655 | Limited public count |
| Niches covered | 108 tracked | Concentrated in tech and finance |
| Primary ad format | Sponsored recommendation in answers | Sponsored follow-up question |
| Buying model | Self-serve + managed | Mostly managed, invite-based |
| Best fit | Broad B2B SaaS, fintech, enterprise | Technical and comparison-led queries |
Who's actually advertising on ChatGPT?
Which niches buy the most ChatGPT inventory?
Where Perplexity tends to win
Perplexity's audience skews toward technical and research-oriented queries, the kind a developer, analyst, or founder types when comparing tools. Its sponsored follow-up question format rewards brands with a clear adjacent question to seed. For consumer products where the buying journey starts with a comparison ('best noise-cancelling headphones under $300'), Perplexity can outperform because the answer itself acts as pre-qualification. For broad B2B SaaS, where ChatGPT's 108-niche coverage already captures enterprise buyers, ChatGPT usually wins on raw volume.
How to choose where to test first
Start with ChatGPT if your buyer uses it for product research, your category maps to one of the 108 tracked niches, and you want self-serve buying today. Add Perplexity once the ChatGPT test is producing measurable clicks or pipeline, especially if your category is technical and your buyer asks comparison-style questions. The simplest rule: go where the buyer already asks, and pick the platform that matches the question format your category needs to win. You can browse every captured ad across niches in our ad library and track live shifts in our intelligence hub.
Frequently asked questions
We probe ChatGPT with realistic consumer prompts and capture the sponsored ad cards it returns — every creative, the triggering prompt, and the advertiser. Figures reflect our captured sample, not OpenAI's internal data. Explore the live ad library and market intelligence.