What is the Facebook Ad Library and how do I use it for research?
The Facebook Ad Library at facebook.com/ads/library is Meta's free public archive of every ad currently running across Facebook and Instagram, searchable by advertiser Page or keyword with no login required. It shows each ad's creative, copy, start date, formats and platforms, which makes it the primary tool for competitor creative research. It does not show spend, reach or performance for commercial ads; those figures exist only for political and issue ads.
Last updated 2026-08-11
What it is and why it exists
Meta built the library as an ad transparency measure: anyone can inspect what any advertiser is currently running, across Facebook, Instagram and Meta's other surfaces, without an account. For political, electoral and social-issue ads it goes further, with spend bands, reach figures and a multi-year archive, because that is the category regulation targets. For commercial ads the archive is shallower, active ads only, no performance data, but the creative transparency alone changed marketing research: your competitors' entire live creative strategy, every angle, hook, format and offer they are paying to distribute right now, is public, legally inspectable, and free.
Searching it effectively
Search by Page name rather than keyword to get a specific advertiser's full set, and verify you found the real Page, since lookalike-named pages and regional variants abound. Set the country filter deliberately, because results differ per market. Filter by platform, media type and date range to isolate what you actually want, for example video ads launched in the last month. Expect many near-duplicate entries, since advertisers run variants and dynamic formats expand into multiple renditions; the header count is ads, not concepts. Click into individual ads to see all versions and placements, which the tile view compresses, and note the start date on anything interesting.
What it will not tell you, and the honest proxies
For commercial advertisers there is no spend, reach, CTR or conversion data, and any third-party tool quoting a competitor's exact ad spend is modelling, not reporting. The library still leaks strategic signal through proxies. Longevity is the strongest: advertisers kill losing ads quickly, so an ad running unchanged for months is very likely paying for itself. Variant density is next, since a concept with many active variations is one the advertiser believes in enough to iterate. Ad-count trends mark scaling or testing pushes. And launch-date clusters reveal each quarter's new bets. Read patterns and trajectories, and distrust any number the library itself cannot show you.
Reading a competitor like an analyst
Single ads inspire; portfolios inform. Pull a competitor's full active set and cluster it: which angles recur, what the format mix looks like, how hooks open, how offers are framed, what the oldest surviving ads share. The old survivors are the proven core, the newest cluster is the current experiment, and the ratio between them tells you whether the account is harvesting or searching. Do this across several competitors and category-wide patterns emerge, angles everyone has converged on, and gaps nobody is exploiting, which are often the more valuable finding. An hour of this quarterly beats daily casual browsing, because the unit of insight is the comparison, not the ad.
From research to launches
The library earns its time when findings change what you test. Translate patterns, not assets: keep the structural insight, open on the problem, demo inside three seconds, anchor the price in the copy, and rebuild it entirely with your own product, claims and footage. Direct copying is a legal risk and a strategic error, since you inherit a competitor's positioning without the context that makes it work. Feed translated concepts into your normal test structure with the same budgets and kill rules as any other idea; a pattern that works for their offer is a hypothesis for yours. The compounding version of this practice is a tagged swipe file that feeds a test calendar, quarter after quarter.
Preserving what you find
The library shows only what is live, so an ad you found today can vanish tomorrow along with the answer to when it was killed and how long it really ran, which is precisely the longevity signal you wanted. Serious research therefore captures at the moment of finding: the creative, the copy, the dates, tagged by advertiser, angle and format so it stays searchable. Volume Creatives does this with a Chrome extension that saves ads from the library into your own creative library and rehosts the media so nothing rots. However you do it, the discipline is the same, capture immediately and tag consistently, because the library has no memory and offers no second chances.
The Ad Library's interface, filters and data coverage change over time, and scraping it at scale is against Meta's terms even though browsing it is explicitly public. Research through the interface or through tools that capture as you browse, not through bulk extraction.