Facebook and Instagram in the same query. Every ad with its full copy, its landing page, its format and the exact dates it was running.
The Meta Ad Library is the richest of the four by volume and the most generous with text: it returns the full ad body, not a truncated preview. In exchange it is the quietest on performance — outside political advertising, Meta publishes neither impressions nor spend. Use it to read creative strategy, not to estimate budget.
We would rather say it before you build an analysis on top of it.
Records from this platform with the fields we deliver. Click a row to see the raw JSON.
| Platform | Advertiser | Creative | Country | Format | From | To | Volume |
|---|
Click any row to see the full record exactly as it arrives from the platform.
Structure, field names and formats are the real ones from each Actor. The LinkedIn records come from real runs; the rest use example advertisers so no invented figures are attributed to real companies.
Exactly as the Actor delivers them, before normalisation.
| Field | Type | Example |
|---|---|---|
| ad_archive_id | string | 1284470119832255 |
| page_id | string | 10228441907 |
| page_name | string | Casa Nordia |
| page_url | string | https://facebook.com/casanordia |
| status | string | active |
| body | string | Summer sale: up to 40% off sofas… |
| link_url | string | https://casanordia.es/sale |
| link_title | string | Shop the collection |
| start_date | date | 2026-07-03 |
| end_date | date | null |
| platforms | array | ["facebook", "instagram"] |
| media_type | string | video |
Need these aligned with the other three platforms? We normalise them onto one schema. See the shared schema →
No, and be sceptical of anyone who claims otherwise. Meta only publishes amounts and impressions for political and social issue ads. For commercial advertising the available signal is how long a creative has been live and how many surfaces it runs on.
Yes. It is one library for the whole Meta ecosystem, and the platforms field says which surfaces each ad is running on: facebook, instagram, audience_network or messenger.
You can start from their domain or their page. Domain is usually more reliable, because many brands run several pages per country or product line.
Another question? Write to us and you'll get the technical detail.
Run a test extraction and look at the dataset before committing to anything. Pay per result: no data, no charge.