Meta Ad Library: The Complete Guide for 2026

The Meta Ad Library (also called the Facebook Ad Library) is a transparency database that Meta launched in 2019, in the wake of regulatory scrutiny over political advertising. What began as a tool for political and issue ads has expanded to index every active ad running across Facebook, Instagram, Messenger, and Audience Network—a public record of the company’s entire live advertising surface.

It is hosted at facebook.com/ads/library and is completely public: no Facebook account, login, or authentication is required. Anyone—marketers, journalists, researchers, or competitors—can search it. Meta built it to be auditable by design, indexing ads from over 8 million advertiser pages worldwide.

For marketers, the library is the foundational source for competitive ad intelligence on Meta. Because every active ad a competitor runs is visible, you can see their creative, copy, calls-to-action, and how long each ad has run—without any special access. It is the Meta equivalent of the Google Ads Transparency Center, which performs the same role for Search, Display, and YouTube ads.

The library distinguishes between two ad categories. Commercial ads (the vast majority) show creative and metadata while active. Political and issue ads carry additional disclosures—who paid for them, spend ranges, and impressions by demographic—and are retained for up to 7 years even after they stop running. This split matters for what data you can expect to find.

Effective searching combines a lookup method with filters. There are three search methods, each suited to a different goal.

Brand name search is the most direct: type the competitor’s name, select their verified page from the dropdown, and browse all of their active ads. Keyword search surfaces ads containing specific terms in their copy—useful for finding every advertiser pushing a particular offer, angle, or product category. Page ID search uses the numeric Facebook page ID (visible in the page’s URL) for exact matching when a brand name is ambiguous or returns multiple results.

Once you have results, the filter system narrows them across five dimensions: country (any market where the advertiser runs ads), platform (Facebook, Instagram, Messenger, or Audience Network), media type (images, videos, memes, or no media), active status (active or inactive), and date range (by ad start date). Combining country and platform is the key move for regional analysis—filtering to “United States” + “Instagram,” for example, isolates a competitor’s US Instagram strategy.

A practical search workflow: start with a brand search to see a competitor’s full active set, then apply a country filter for your market, then sort by date to separate long-running evergreen ads (likely winners) from recently launched tests. Note an ad’s start date—ads live for months signal proven performers, while a burst of new ads signals a campaign push or product launch.

For Instagram specifically, the library covers all Meta placements by default; apply the Instagram platform filter to view only Instagram-eligible creative. There is no separate Instagram ad library—it is the same database, filtered.

Understanding the library’s boundaries is essential to using it well. For any commercial ad, you can see: the creative (image, video, or carousel), the primary text, the headline and description, the CTA button (Learn More, Shop Now, Sign Up, etc.), the platforms it runs on, and the date it started running.

What you cannot see for commercial ads is just as defining: audience targeting (who the ad is shown to), spend or budget, performance metrics (impressions, CTR, conversions), and A/B test variants grouped together. The library reveals what a competitor says and shows—not who they target or how well the ad performs.

The one exception is political and issue ads. These carry mandatory disclosures: a “Paid for by” label, spend ranges, total impressions, and audience breakdowns by age, gender, and location. This richer data exists because of election-integrity regulation, not commercial transparency—so most marketers researching brand competitors will only see the commercial-ad fields.

A second limitation is media longevity. Inactive ads remain listed for reference, but their creative media (images and videos) can expire after the ad stops running, leaving the metadata without the asset. For reliable access to historical creative, you need a tool that caches the media at collection time—which is one reason teams move from manual browsing to tools like Rival or AdSpy that store scraped creative independently.

The Meta Ad Library offers no native download or export feature—a frequent frustration for anyone building a swipe file or report. There are three ways to save ads, in increasing order of scale.

Manual saving. For a handful of ads, take screenshots or right-click to save individual images. Video ads are harder: you’ll need screen recording or browser developer tools to extract the media URL. This works for a quick reference set but does not scale and produces unstructured files.

Browser extensions and scrapers. A range of third-party tools and Chrome extensions promise to filter the library and bulk-download creative. Quality and reliability vary, and many break when Meta updates the page. Treat them as convenience utilities rather than dependable infrastructure.

Automated scraping tools. Purpose-built tools collect and store creative as part of their pipeline. Rival downloads every active ad’s media during scraping and stores it in its own database, so the creative remains available for analysis and export even after the original ad goes inactive and Meta’s copy expires. AdSpy similarly maintains an independent cache of historical ad media. The advantage over manual saving is structure: ads are stored with their metadata, organized by competitor, and exportable to PDF or Excel for sharing.

On legality: the 2024 Bright Data v. Meta Platforms ruling confirmed that scraping publicly available data does not violate the Computer Fraud and Abuse Act. The Ad Library is public by design, so collecting and saving its contents for competitive research is permissible—standard fair-use principles apply, meaning you analyze the creative rather than republish it as your own.

The Meta Ad Library API lets developers query ad data programmatically rather than through the web interface. In principle, it could power bulk competitive analysis—but in practice its scope is narrow.

The API is primarily restricted to ads about social issues, elections, and politics. To use it, you must complete identity confirmation and agree to Meta’s terms. For political and issue ads, it returns rich fields: spend, impressions, demographic distribution, and delivery dates. This makes it valuable for researchers and watchdogs studying political advertising.

For commercial ads—the category most marketers care about—API access is limited. The general web library remains the most complete view of commercial advertisers’ active ads, and there is no comprehensive commercial-ad endpoint equivalent to the political-ads API. This gap is the core reason competitive ad intelligence tools rely on browser-based scraping of the public library rather than the official API.

Rival takes the scraping approach: a Playwright-based scraper loads each competitor’s public Ad Library page and extracts structured data—creative URLs, headlines, body copy, CTA text, and platform distribution—directly from the rendered page and its underlying data. This captures the full commercial-ad picture the API does not expose, and feeds it into AI analysis. The result is the structured, queryable competitive dataset the official API would provide if commercial coverage existed.

The Meta Ad Library is excellent for browsing one competitor occasionally. It breaks down for systematic analysis: there’s no cross-competitor view, no export, no saved searches, and no analysis layer. Reviewing a single competitor’s 30–100 active ads and noting patterns by hand takes 3–4 hours, and the output is subjective notes that degrade over time.

Rival automates the full path from collection to insight. After AI competitor discovery, its scraper collects every active ad from each competitor’s library page, then runs three parallel analysis tracks. Copy analysis (GPT-5-nano, batched 5 ads per call) extracts messaging themes, tone, value propositions, and CTA patterns. Image analysis (GPT-4o vision) annotates composition, brand elements, and text overlays. Video analysis extracts 6 keyframes via ffmpeg for scene-by-scene breakdown.

The aggregated output is an Ad Intelligence Report with an Ad Strategy Fingerprint (the competitor’s signature pattern of formats, themes, and CTAs) and a Creative Gap Score (a 0–100 metric measuring exploitable gaps). Where manual browsing answers “what is this one ad doing?”, the report answers “what is this competitor’s overall strategy, and where is it weak?”

Pricing scales with need: Rival’s free tier covers 1 competitor (10 AI-annotated ads); Pro ($19/mo) covers 5 with unlimited ads and monthly auto-refresh; Team ($49/mo) covers 15 with 5 seats and PDF/Excel exports. The full pipeline runs in under 5 minutes per competitor at roughly $0.02 per ad analyzed—turning the raw library into structured, comparable intelligence.