Best ad library API in 2026: 9 options compared

The best ad library API depends on what you are building. If you need cheap Meta search, three vendors sell it for about a dollar per thousand results at volume. If you need to know what a competitor ran last spring across six platforms, almost none of them can help. This guide compares nine options as of October 2026 and tells you which one we would pick for each job, including the jobs where we are not the answer.

A disclosure up front: we build AdLib, and we are new. Our archive starts with our first crawl, we have no SLA yet, and we are not listed in n8n, Zapier or Clay. Where a competitor is better, we say so.

What an ad library API actually returns

An ad library API returns the ads a platform publishes in its public transparency library: Meta Ad Library, Google Ads Transparency Center, TikTok Ad Library, LinkedIn Ad Library, Pinterest Ads Repository and the X Ads Repository. You get advertiser, creative text and media, landing URL, dates and, for some ads, reach or spend ranges.

Almost every vendor we checked is a live wrapper. It fetches what the library shows at that moment and bills per request or per record. That is fine for search. It is a problem when the library forgets: outside the EU, Meta only shows commercial ads while they run, and the EU's Digital Services Act requires large platforms to keep ads only for their run time plus one year (DSA Art. 39). We wrote about that in why ad libraries forget ended ads.

Ad library API comparison table (October 2026)

All competitor facts below come from vendor pages and docs checked on 26 Oct 2026. Prices change; follow the links before you buy.

"Not found" means we looked and did not find it, which is not proof that it does not exist.

Where the competitors are better than us

Credibility matters more than a clean sweep, so here is the other column of the scorecard.

Archive size

Foreplay advertises an archive of tens of millions of ads (its own pages say 54 million, 100 million and 200 million, see Foreplay). Our archive starts with our first crawl. For non-political Meta ads outside the EU, nobody can backfill what ran before they started watching, us included.

Enrichment

ScrapeCreators returns transcripts for Meta video ads through a /ad/transcript endpoint (docs). Foreplay includes transcripts in its API. Adyntel flags UGC videos with check_ugc. We return raw data without AI tags or transcripts.

Reliability on paper

SearchAPI.io offers a 99.9% SLA on all plans (pricing). ScrapeCreators publishes a 98.2% success rate and 3.12 s latency over 30 days on its homepage. We have no SLA yet, and our X adapter is beta: X search often answers 503 "over capacity".

Ecosystem

SerpApi has a verified n8n node. Apify connects to Zapier, Make, n8n, Sheets and Slack. Adyntel is a data provider inside Clay. ScrapeCreators sits on Composio and Pipedream. We are not listed anywhere yet; you connect us with an HTTP step, as shown in ad library data in Clay and n8n.

Meta EU depth

SearchAPI.io's Meta detail engine returns eu_transparency and uk_transparency with targeting, eu_total_reach and breakdowns by country, age and gender, plus payer_beneficiary_data (docs). We return EU fields on the ad detail, not yet in search.

Which ad library API for cheap raw Meta search

For high-volume Meta keyword search where you only need today's results, pick SearchAPI.io, ScrapeCreators or an Apify Meta actor. SearchAPI.io goes down to $1.00 per 1,000 searches and backs it with an SLA. ScrapeCreators goes down to $0.99 per 1,000 requests, and its credits never expire, which suits bursty usage. The curious_coder actor on Apify starts at $0.75 per 1,000 ads and has the most users of any community actor.

As a facebook ad library api alternative for pure search, these three are mature and cheap. We are competitive on price here (a live search page costs at most 5 credits per platform, however many ads it returns, and cached results under 24 hours are free), but price alone is not a reason to switch from a vendor with a longer track record.

Which one for lead enrichment by domain

For "is this company advertising, and where?", we would pick AdLib, with Adyntel as the established alternative. Adyntel takes a domain as input for Meta, Google, LinkedIn and TikTok and is already a Clay data provider, which is a real advantage if your team lives in Clay. Each platform is a separate call there.

GET /v1/advertisers/resolve?domain= returns advertiser ids on all six platforms in one call, each with a confidence score. On 26 Oct 2026, nike.com resolved to Meta page meta:15087023444 (score 1.0, matched because 11 of 12 sampled ads link to nike.com), two Google advertisers at 0.97, TikTok, LinkedIn and Pinterest accounts, and an X account at 0.75 on a name match only. A resolve costs 10 credits, and 0 when nothing is found. For a pure yes or no, GET /v1/domains/{domain}/check costs 3. The details are in find advertiser ids from a domain.

Empty results, timeouts and upstream errors cost 0 credits with us. That matters for lead lists, where most domains run no ads at all. Adyntel does not charge for a 204 (no page found), but its FAQ is ambiguous about results with zero ads. One Apify actor, s-r/cross-platform-ad-library, also resolves advertisers across five platforms with a confidence score, at $8 per 1,000 ads.

Which one for monitoring competitors over time

For tracking what competitors launch, stop and change, this is the job we built AdLib for, and we would pick it. A watchlist (POST /v1/watchlists with advertisers, webhook_url and interval_hours) sends HMAC-signed advertiser.changes webhooks for new, stopped and changed ads. Among the API vendors we checked, Adyntel's webhook_url only delivers large results asynchronously, Foreplay alerts in its UI, and Atria says webhooks are "not yet available" (Atria).

Monitoring only works if ended ads stay retrievable. We store a new version only when an ad changes, keep ended ads and their media under the same id, and expose /history, /timeline and /changes?since= per advertiser. Media URLs are permanent, while ScrapeCreators and the Apify Meta actors return platform CDN links that expire. That history is what makes run length usable as a signal, as in ad run length as a performance signal: Nike's Meta ads in our DE index ran from 8 days (a Black Friday promo, 18 to 26 Nov 2025) to 86 days.

Metapi is the alternative to look at if you only care about Meta. As of October 2026 it advertises over two years of history including removed ads and real-time webhooks. We could not verify those claims beyond its marketing page, and its sample response shows Meta CDN URLs. If you need a deep Meta back catalogue today, Foreplay's archive is larger than ours.

Which one for AI agents

For agents that research competitors in Claude, Cursor or similar, we would pick AdLib: the same server runs locally with an API key or as a remote connector with OAuth, so Claude.ai and ChatGPT users don't paste a key. Our adlib-mcp package has 20 tools, including history, timelines, watchlists, batches and webhook tests, so an agent can set up monitoring and not only search. Adyntel's MCP has 3 tools.

The honest gap: ours runs over stdio with an API key. Foreplay and ScrapeCreators offer remote MCP with OAuth, which is what Claude.ai connectors and ChatGPT expect. Setup and example prompts are in ask Claude about competitor ads with MCP, and the agent docs live at /for/ai-agents.

claude mcp add adlib --env ADLIB_API_KEY=your-key -- npx -y adlib-mcp

Which one for enterprise datasets

For bulk data delivered into a warehouse, pick Bright Data. Its Facebook ads scraper takes up to 5,000 URLs per call, delivers JSON, NDJSON or CSV to S3, GCS, Azure, Pub/Sub, SFTP or Snowflake, and offers a premium SLA on its enterprise plan. We have an async batch endpoint (POST /v1/batch) but no file export or warehouse delivery yet.

If you need modeled spend estimates outside political and EU data, none of the APIs in this table provide them. That is the territory of enterprise ad intelligence vendors like Pathmatics, at a very different price point.

Meta's official API and Google: when free or single-platform is enough

Meta's official Ad Library API is free and the right choice for political and issue ad research. Its reference says ads that did not reach any EU location "will only return if they are about social issues, elections or politics" (Meta docs). For commercial ads outside the EU it returns nothing, which is why third-party APIs exist. We cover the details in Meta Ad Library API alternatives.

For a google ads transparency api with nothing else attached, SerpApi is a strong pick. Its engine already returns total_days_shown, first_shown and last_shown per creative (SerpApi), and its n8n node is verified. For a tiktok ad library api, TikTok's own Commercial Content API needs an approved application, so most commercial users pick between SearchAPI.io, ScrapeCreators, Adyntel and us.

How to choose

Pick by the job, not the logo:

If your job is in the AdLib rows, start with the free tier (1,000 credits once, then 100 a month) and run a resolve on your own domain. Developers can begin at /for/developers, teams buying data at /for/companies, and the per-platform coverage is on /platforms.