How OpenAdLibrary captures public ads
How ads are observed
Native ads: OpenAdLibrary collects them API-first. Through geo-matched infrastructure it requests the public recommendation widgets that publisher pages load, without rendering the page; a browser crawler that renders public pages and detects served ad units remains as a fallback. It downloads and re-encodes the creative, deduplicates matching copies, and records where and when the placement appeared.
Meta and TikTok ads come from the platforms’ own public sources: Meta’s Ad Library, and TikTok’s ad library and Creative Center. OpenAdLibrary shows the figures the platform publishes, such as reach and targeting, and its own spend estimate, always marked Est.
How the advertiser is resolved
Native ads commonly route through tracking services. OpenAdLibrary follows those public redirect responses directly to the advertiser landing page and attributes the destination back to the creative. Known ad-tech and supply-side domains are filtered so an intermediary is not mislabeled as the advertiser.
No live third-party ad clicks
OpenAdLibrary does not auto-click live third-party ads. Following redirect responses directly avoids generating a paid click. Clicking is restricted to owned or test inventory, and captures are logged.
Evidence and auditability
An observation can retain the creative, captured landing page, supply chain, collection source, geography, device, and observation time. Public pages and the developer API expose only leak-safe fields.
Collection-source provenance
- Crawl: observed on a public surface by OpenAdLibrary collection systems.
- Official API: imported from an official platform source where available.
- Panel or extension: labeled separately when contributed through those collection methods.