ChatGPT Ads Advertising Platform: Targeting Options, Industry Measurement, and Ad Transparency
We scored OpenAI's ChatGPT ads platform on targeting options, industry fit, measurement, and ad transparency using only OpenAI's own documentation, then ranked the tools that cover its gaps.
We usually rank tools. This time we scored a platform: OpenAI's ChatGPT ads system, as documented in its help center, developer docs, and ad policies on 2026-10-21. Four criteria, the ones people search for together: targeting options, industry fit, measurement, and ad transparency. Then we ranked the monitoring tools that cover what the platform leaves out. The full tool directory is on the rankings page.
How we scored
Each criterion gets a short verdict based only on what OpenAI publishes. No anecdotes from ad buyers, no leaked rate cards. If OpenAI's pages do not describe a capability, we treat it as absent.
1. Targeting options: narrow but sensible
Campaign-level targeting covers three things (targeting docs):
- Geography. Country codes, plus regions and markets through location IDs, with exclusions. Up to 2,500 IDs per list. Product-feed campaigns target at country level only.
- Platform. iOS app, Android app, or web. Web includes mobile browsers, so "web" is not the same as desktop.
- Custom audiences. Uploaded customer or prospect identifiers. Inclusion and bid multipliers need at least 25,000 matched users. Exclusion audiences have no minimum, and exclusions win when someone is in both lists.
Ad-group level adds context hints: up to 2,000 descriptions of the conversations, topics, or use cases where the ad is relevant. OpenAI says plainly they are not exact-match keywords and do not guarantee delivery (The Basics). Qualifying advertisers can also use Negative Phrases for brand-specific placement limits.
Verdict. No demographic, interest, or job-title targeting is described. Intent comes from the conversation itself. That is the point of the product, and it means you control relevance mostly through copy, landing page, and hints.
2. Industry fit: depends entirely on your category
There is no industry selector. What decides industry fit is the ad policy. Early phases focus on household and consumer goods, local services, travel and entertainment, and digital products and education. Financial services, health services, and legal services are restricted to approved US advertisers on a case-by-case basis and are generally prohibited outside the US. Political ads are not allowed. Individual job and housing listings are prohibited.
Placement rules also matter by industry. Ads are not eligible near sensitive conversations such as personal health, mental health, or politics. Since April 2026, medical, legal, and financial advice contexts are no longer blocked by default, per the policy changelog.
Verdict. Strong fit for consumer goods, travel, local services, and digital products. Gated for finance, health, and legal. Closed for politics.
3. Measurement: solid basics, no benchmarks
Buying options are Views (CPM), Clicks (CPC), and Conversions, which optimize toward a conversion event while billing per click or per thousand impressions. You never pay per conversion (Create Campaigns). OpenAI suggests starting CPC bids at $3 to $5. The auction is relevance-weighted and second-price.
Reporting includes impressions, clicks, spend, CTR, average CPC, average CPM, and conversions. Impressions and clicks refresh about every 15 minutes. Spend can lag seven to eight hours. Conversions come from a measurement pixel or a server-side Conversions API. Default attribution is a 30-day click window and a 1-day view window, and reports can break out by country or device (reporting docs).
The gap is industry measurement. OpenAI's FAQ says the platform "does not yet have performance benchmarks across advertisers, industries, or campaign types" (FAQ). You can measure yourself. You cannot measure yourself against your vertical.
Verdict. Good enough to run a test properly. Not enough to know whether your CTR is good for your industry.
4. Ad transparency: strong per ad, absent per market
Per ad, OpenAI does well. Each ad is labeled, sits below the answer, and has a menu with About this ad, Hide, Report, and Ask ChatGPT. Users can view their ad history and topics and delete ad data (Ads in ChatGPT). Advertisers get aggregated results only, never chats.
Per market, we found nothing. None of OpenAI's ads help pages or policies describe a public library of who is advertising, comparable to Google's Ads Transparency Center. An outside review by the Network Advertising Initiative also found, in May 2026 screenshots, that the "Learn about ads and personalization" link did not appear under every ad.
Verdict. A user can learn why they saw one ad. A brand cannot look up a competitor's campaign.
The tools that cover the gaps, ranked
Two of the four gaps (industry benchmarks and market transparency) can only be filled by watching answers yourself. These are the tools in our catalog, or with sourced product pages, that do it.
1. Promptwatch. Ads Radar captures sponsored ads inside ChatGPT Search, Google AI Overviews, and Google AI Mode answers to your tracked prompts, with advertiser, domain, copy, image or price, target URL, and the prompt that triggered it (Ads Radar docs). Build a prompt set for your industry and you get your own industry benchmark: which questions carry ads, who buys them, and how often. It ranks first because the same prompts also carry your organic mentions, citations, crawler logs, and AI-referred conversions. Ads Radar is on Business ($579/mo) and all agency plans (from $199/mo), not on Essential or Professional, per llms.txt on 2026-10-21.
2. Gauge. Lists ChatGPT Ads tracking from $599/mo. Its feature page describes competitor placements plus a connection to your own OpenAI Ads account. No published crawler logs or conversion-level visitor analytics in our catalog.
3. Adthena. Not in our directory. Its ChatGPT ads page describes competitor tracking, a searchable ad library, and share-of-search metrics, with a 21-day free trial and no published price. Ads only, no organic visibility.
4. LLM Pulse. Lists AI ads tracking from €49/mo with a weekly default cadence. The cheapest entry, the slowest refresh.
OpenAI built a reasonable ad platform for buying. It did not build one for watching. That half is on you, and Promptwatch is where we would start.