TL;DR: The Meta Ad Library is a free public database of every active ad on Facebook, Instagram, Messenger, and Audience Network. You see the creative, the advertiser, the start date, and the platforms — not the spend, impressions, or targeting. Political and issue ads show more. The most useful signal on a general ad is the start date: anything running 60+ days is a winner. Long-runners are the ads competitors are scaling.

How the Meta Ad Library works

The Library lives at facebook.com/ads/library. Meta launched it as the Ad Archive in 2018 under election-interference pressure, expanded and rebranded to the Ad Library in 2019, and has run it as a public-facing transparency tool ever since.

graph LR
    A[Anyone with a browser] --> B[facebook.com/ads/library]
    B --> C{Search by}
    C -->|Keyword| D[Active ads matching<br/>text in copy]
    C -->|Advertiser| E[Page → full active<br/>ad set for that brand]
    D --> F[Filter by platform,<br/>language, media type]
    E --> F
    F --> G[Click ad card →<br/>creative + start date<br/>+ variants]

No Facebook account required. No login. No payment. For an ad to appear, it has to be currently delivering impressions — once paused or out of budget, it drops out of the Library unless it's a political/issue/election ad in a region with extended retention rules.

What you see vs. what you don't

The Library is most useful when you're clear about its scope. The gap between what it shows and what advertisers wish it showed is wide.

Field General ads Political / issue / election ads
Creative (video, image, copy, CTA) ✅ ✅
Advertiser Page ✅ ✅
Start date ✅ ✅
Platforms running on ✅ ✅
Variant count under one Library ID ✅ ✅
Spend ❌ Banded ranges
Impressions / reach ❌ Banded ranges
CTR / performance ❌ ❌
Audience targeting ❌ Top demographics only
Funding source ❌ ✅
Historical retention Dropped at end 7+ years

The political/issue tier is why journalists and researchers care about the Library. For most marketers, the general-ad tier is what matters — and the start date is the single most useful piece of metadata on it.

How to search the Library

Five steps. This is the part you'll come back to.

1. Open facebook.com/ads/library

The search bar opens with a country selector (default: your location) and an ad-category dropdown (default: "All ads"). Confirm both before searching. "Nike" in "US, All ads" returns different results than "Worldwide, All ads".

2. Pick your search type

Type Use for
Keyword search A specific phrase, slogan, or product term. Returns active ads whose copy contains the term.
Advertiser search A specific brand. Returns the Page, then the brand's full active ad set when you click in.

For competitor research, search by advertiser. Keyword search misses ads where the brand name isn't in the visible copy (which is most of them).

3. Filter the result

Once on a Page, four filters tighten the view:

  • Active status (active only / inactive only / both)
  • Platform (Facebook / Instagram / Messenger / Audience Network)
  • Languages
  • Media type (image / video / with link / no link)

Filter to active video ads only for the cleanest signal on what a brand is currently scaling. 80 active video ads on a Page means real budget. 4 active means testing or maintaining.

4. Read each ad card

Click an ad. You get the full creative, the start date, the platforms, the Library ID, and any other versions of that ad (Meta groups creative variants under a single Library ID).

The single most useful piece of metadata is the start date.

  • 60+ days running = winner. Meta's algorithm is allocating budget because it's converting.
  • 3 days running = test. Most won't survive.
  • Library has 8+ variants under one ID = serious A/B budget behind it.

5. Capture what matters

No built-in export. Three options:

  1. Right-click → save individual creative files.
  2. Browser screenshot extension for quick archival.
  3. Third-party tooling (AdSpy, Foreplay, Superscale's Competitor Tool) for ongoing storage and cross-advertiser search.

Using the Library for competitor research

Most operators "browse" the Library when they need it. The teams that get real value run a structured weekly process instead.

Build the watch list

Pick 8–12 brands you compete with directly. Track them in a Notion table or spreadsheet:

| Brand | FB Page URL | Ad Library URL | Active ads (last wk) | Active ads (this wk) | Long-runners (60+ days) | New tests (≤7 days) | Notes |

Score the long-runners weekly

The ads running 60+ days are the ones that work. Pull the creative, watch the video, read the copy. These become your reference library for creative direction — not to copy, but to study what hook structure, offer framing, and visual format look like at the working end of your category.

Score the new tests

The 1–7 day window is where your competitor is making this month's bets. Most will fail. Tracking what they try gives you a forward-looking read on category strategy.

Cross-reference against your own creative

If a competitor is testing a format you haven't, that's a signal. If they've been running the same hook for 90 days and you're now testing the same hook, you're late.

The limits, and how to work around them

The Library's biggest gap is performance data. You see the creative; you don't see whether it works. The closest proxies:

  • Time-on-Library. Long-runners are working. Ads that drop after 5 days probably weren't.
  • Variant count. Ads with 8+ variants under one Library ID signal a budget worth A/B-testing.
  • Cross-platform presence. Ads running on all four placements (FB / IG / Messenger / AN) usually carry a bigger budget than IG-only ads.
  • Page-level activity over time. The count of active ads on a Page tracks roughly with spend.

These proxies aren't precise, but they beat guessing. Third-party tools (AdSpy, Foreplay, PowerAdSpy, Atria) layer more metadata on top: estimated spend, engagement counts, "trending" rankings. Useful, but the underlying data still comes from the same public Library.

How AI agents use the Library programmatically

The Ad Library is structured public data. That makes it a clean input for an AI marketing agent.

Instead of a human clicking through 80 competitor ads a week, the agent pulls active ads matching a category or brand list, parses the creative (computer-vision on images and videos, transcription on audio), clusters the patterns (hook structure, offer framing, visual format), and surfaces what's working.

Superscale's Competitor Tool runs exactly this loop. Configure a brand list; the agent monitors the Library, pulls active ads, scores them, and feeds the patterns into your own creative production pipeline.

Two case-study patterns:

  • Solo operator: Advercy. A one-person performance marketing consultancy running 5 client brands in a single Superscale workspace. Result: 95% drop in UGC production cost, 10× faster ad creation, 5× more creative shipped — because the inputs were structured competitive research rather than ad-hoc browsing.
  • Agency: marketbirds. Uses the Competitor Tool to bring competitor breakdowns into client review calls. The client sees what the competition is running, signs off on directional creative, and the agency frontloads a month of variants for bulk approval. The Library shifted from a research stop to a continuous input.

Library vs. third-party ad-spy tools

Three tiers of tooling. Pick the one that matches what you actually do with the data.

graph TD
    A{What do you do<br/>with the data?} --> B[Occasional research<br/>One-off checks]
    A --> C[Weekly structured<br/>competitive intel]
    A --> D[Also produce creative<br/>off the back of it]
    B --> B1[Meta Ad Library<br/>Free, native]
    C --> C1[Paid spy tools<br/>AdSpy, Foreplay,<br/>PowerAdSpy, Atria]
    D --> D1[Integrated AI agent<br/>e.g. Superscale<br/>Competitor Tool]
Tier Best for Cost
Native (Meta Ad Library) Occasional research, journalists, one-off competitive checks Free
Paid spy tools Weekly structured competitive intel, engagement counts, estimated spend $50–$300/mo
Integrated AI agent Turning competitor research into produced creative, not just dashboards Varies (Superscale plans from $49/mo)

If you research and don't act on it, the free Library is enough. If you research and produce creative weekly, integrated tools save more time than they cost.

FAQ

Is the Meta Ad Library free?
Yes. It's a public tool. No account needed, no payment, no sign-up.

Do I need to be logged in to Facebook to use the Meta Ad Library?
No. The Library is accessible to anyone with a browser. Some advanced filters surface more cleanly when logged in, but the core search and result view doesn't require an account.

Can I see how much a competitor is spending on Facebook ads?
For general ads, no. You see the creative, the start date, and the platforms — not the spend or impressions. For political, social-issue, and election ads, the Library shows banded spend and impression ranges plus demographic splits.

How do I know if a competitor's ad is performing well?
The Library doesn't show performance metrics directly. The best proxies are time-on-Library (long-runners are working), variant count (ads with many creative variants usually have more budget behind them), and cross-platform presence (ads running on all of FB / IG / Messenger / Audience Network usually mean a larger spend).

Does the Meta Ad Library show TikTok or Google ads?
No. The Library is Meta-only. TikTok has its own Creative Center for ad inspiration. Google has the Ads Transparency Center, launched in 2023, which works similarly to the Meta Ad Library for Google's ad inventory.

Is there a Meta Ad Library API?
Yes, but with limits. Access is granted to researchers, journalists, and approved third-party tools. It's not a self-serve API for marketers. Most ad-spy products in the market use a combination of the official API (for political ads) and large-scale scraping (for general ads).

How long are political ads kept in the Library?
At least 7 years in most regions with transparency requirements (EU, UK, US, Canada). General ads drop out as soon as they stop running.