ChatGPT ad results may contain ads that were returned in the response but were not actually displayed to the user. A new field in ChatGPT LLM Scraper API now helps you identify which ChatGPT ads actually made it onto the screen to measure real user exposure more accurately.
What’s new
We added a new boolean is_rendered field to ad results returned by:
The field indicates whether a served ad was rendered as a user-visible card in the captured ChatGPT response.
How it works
The is_rendered field returns true when an ad is present as a visible element in the response DOM at capture time.
Because the returned ads list can include ads that were served but never displayed, is_rendered helps distinguish between served ads and user-visible ads. For impression analysis, counting ads with is_rendered: true provides a more accurate indication of actual exposure than relying on the total number of ads returned.
Why it matters
With the new is_rendered field, you can:
- measure user-visible ad impressions more accurately,
- distinguish between ads that were served and ads that were actually displayed,
- improve ChatGPT ad reporting and exposure analysis,
- build more precise ad monitoring and analytics workflows.
Check our documentation to start working with the new is_rendered field in ChatGPT LLM Scraper API responses.