Advertiser-supplied context hint
A phrase supplied inside an ad group to add genuine information about the product, who it helps, or when it is useful. This is private campaign configuration unless the advertiser publishes it.
Targeting explainer
ChatGPT Ads context hints
Context hints describe the conversations, needs, or situations where an ad group may be useful. They guide matching, but they are not exact-match keywords, audience rules, or a promise that an ad will appear.
OpenAI defines context hints as advertiser-provided information that helps its systems understand when an ad may be relevant to conversations or user journeys. A hint can describe what the product offers, who it helps, or when it may be useful.
Three different objects
Do not confuse research with campaign settings
Only the first item below is an advertiser's actual context hint. The other two belong to the library's observation and interpretation workflow.
Collector probe prompt
A controlled or curated prompt used by this library to observe sponsored placements. It is not a verified real-user query, a paid keyword, or proof that an advertiser selected the prompt.
Reconstructed context hypothesis
A directional description inferred from public creative, landing-page, category, and observation evidence. It is a hypothesis for research, never a claim about private advertiser settings.
Mechanism
Context hints are not keywords
Describe meaning and usefulness in natural language. Do not build a disguised keyword list or imply control over the full selection system.
Need, product fit, and situation
- Add information beyond the title, copy, and landing page.
- Connect a genuine product benefit to a specific customer need.
- Keep one ad group focused on a coherent product, theme, or intent.
Broad labels and unrelated lists
- Avoid isolated category words with no useful situation.
- Do not combine products that need different creative or landing pages.
- Do not treat a hint as an audience rule, exact match, or delivery command.
Public examples
From observed category evidence to a hypothesis
These examples stay at category level. Each evidence link opens a public ad capture. No advertiser's private hint is shown or claimed.
A collector might probe a beginner comparing an espresso setup around a fixed budget.
Category hypothesis: a shopper choosing a first home espresso setup that can support later upgrades.
Inspect the exact public ad evidenceA collector might probe a shopper comparing whole-genome services and their practical tradeoffs.
Category hypothesis: a customer evaluating deeper DNA analysis after using a consumer testing service.
Inspect the exact public ad evidenceThe observation supports only that the linked sponsored creative appeared with a controlled probe in the captured sample. It does not reveal the advertiser's submitted hints or prove why the system selected the ad.
Methodology
Observe the public placement, preserve the probe and creative, then label any reconstructed context as a hypothesis.
The library runs controlled or curated probes, records sponsored placements, and groups public evidence by category. It does not access advertiser accounts, private campaign settings, user chat logs, bids, or OpenAI's internal selection rationale.
DirectionalWhat reconstructed hints can support
Coverage is a captured sample, not a complete inventory. An appearance is not proof of targeting, demand, performance, spend, or future delivery.
Read the full methodologySources
Primary guidance and public evidence
Official OpenAI documentation governs platform behavior. Library pages document only this project's observations.
FAQ
Context hints, answered
The visible answers below are identical to the FAQ schema on this page.
What is a context hint in ChatGPT Ads?
A context hint is information an advertiser supplies at the ad-group level to describe what a product offers, who it helps, or when it may be useful. OpenAI says context hints help its systems understand when ads may be relevant to conversations or user journeys.
Are context hints the same as keywords?
No. OpenAI says context hints are not exact-match keywords. They can describe broader needs, situations, and natural variations, and they do not guarantee delivery for particular words, topics, audiences, or conversations.
Does ChatGPT Ad Library show an advertiser's real context hints?
No. The public library shows captured sponsored creatives and controlled probe prompts associated with those observations. Any context description reconstructed by the library is a research hypothesis, not the advertiser's private ad-group configuration.
Are the prompts in ChatGPT Ad Library real user searches?
No. They are controlled or curated collector probe prompts used to observe whether sponsored placements appeared. They are public research inputs, not verified queries from ChatGPT users and not proof of an advertiser's targeting.
Can a context hint guarantee that my ad appears?
No. OpenAI says hints guide matching but do not guarantee delivery in a specific conversation. Selection considers multiple signals, including conversation context and intent, the landing page, creative, advertiser-provided hints, and other eligible signals.
Where can I build a company-specific context hint?
Use ChatGPT Ad Library to inspect public evidence and ContextHint to turn that research into an editable, company-specific planning workflow. Validate every proposed hint against your actual offer, landing page, audience, and campaign structure before launch.
Turn public evidence into your own plan
ChatGPT Ad Library is the public evidence source. ContextHint is the personalized workflow for drafting, validating, and organizing company-specific hints.