Facebook Ads Library Scraper · No Login · $0.3/1k
Pricing
from $0.30 / 1,000 ads
Search Meta ads by keyword, page or URL. Export copy, creatives, CTA, dates, platforms and optional EU reach to CSV, Excel or JSON. Core fields align with Apify's official scraper. $0.30/1k ads on Business or $0.45/1k on Free; no start fee or platform charge.
Words or phrases to search for in ad texts and advertiser names, one per line, e.g. running shoes or Nike. Each keyword is one search. Combine with Country and the filters below.
Numeric Facebook page IDs of advertisers whose ads you want, e.g. 15087023444 (Nike). Find the ID in the Ad Library URL after view_all_page_id=. Returns every ad of the page in the selected country.
URLs copied from the Facebook Ad Library, in the same form the official Facebook Ads Scraper takes: a keyword search (...?q=nike&country=US...), an advertiser page (...?view_all_page_id=15087023444...) or a single ad (...?id=1785179045778580). Filters in the URL (country, active status, media type, dates, platforms, languages, sorting) override the fields below. Plain Facebook page URLs work too: https://www.facebook.com/Nike returns every ad of that page.
Country of the audience: two-letter ISO code such as US, GB or TH, the English country name (Thailand), or ALL for every country.
All ads, only ads that are Active now, or only Inactive ones.
Only ads with this kind of creative.
All ads, or only the special categories Facebook keeps for years: political and issue ads, housing, employment or credit ads.
Only ads shown on these Meta platforms. Empty = all platforms.
[]Unordered finds ads containing all the words in any order (Facebook's default); Exact phrase needs the words together, in order.
Date puts the newest ads first (the Ad Library's "Most recent" order, by the month the ad started). Impressions puts the most viewed ads first (the Ad Library's "Impressions: high to low"); Facebook publishes impression ranges only for issue, election and political ads, so combine it with that ad category.
Only ads whose text is in these languages: two-letter ISO 639-1 codes such as de, fr or pt, one per line. Empty = any language.
[]Only ads that were shown on or after this date (the Ad Library's Impressions by date filter; an ad that started earlier but was still running counts). A date YYYY-MM-DD, or a period before the run: 7d, 30 days, 3 weeks, 2 months, 1 year, today, yesterday.
Only ads that were shown on or before this date. Same forms as Shown after: YYYY-MM-DD, yesterday, 7d and so on.
Maximum number of ads per keyword, page or URL. Each ad is one record; you pay only for records written.
Add the ad_details block to every ad: advertiser page info (About text, likes, Instagram followers, verification), payer and beneficiary, EU transparency with total reach and the reach by country, age and gender, plus spending for political advertisers. Slower (one extra lookup per ad), no surcharge.
Name of a watchlist that remembers every ad this run delivers, e.g. nike-us. The IDs are kept in a key-value store called watchlist-<name> in your account and updated after each run. Use the same name in every run of a schedule; leave empty to keep nothing.
Skip ads that an earlier run with the same Watchlist name already delivered. They are neither written to the dataset nor charged; the log says how many were skipped. With Date sorting, each search is read to the end once a day and the runs in between read only the newest ads. Needs a watchlist name.
How many keywords, pages or URLs are searched in parallel (default 5). Results of one search are always read in order.