Ad Spy API Comparison 2026: AdLibrary, BigSpy, Apify and Friends

Every ad spy API pitch reads the same. Millions of ads. Every network. Instant access. Then you open the docs and discover that "API" meant a CSV export button, or a community scraper that died the last time Meta touched its markup. This ad spy API comparison exists so you find that out before you commit a sprint to the wrong vendor, not after.

TL;DR: Ad spy APIs differ less on raw data volume and more on three boring axes: platform breadth, field depth, and whether the thing is a real API or a scraper in a trench coat. Meta's free Ad Library API is the legitimate baseline but returns political and social-issue ads only outside the EU. The AdLibrary API covers 11 platforms with performance signals behind one key. BigSpy and PowerAdSpy are web tools first, APIs as an afterthought. Apify actors are scrapers with an API-shaped wrapper. Google's Transparency Center has no REST API at all, but it does have a BigQuery dataset.

If you arrived here from a general roundup like our best ad spy tools comparison, this piece goes one layer deeper. Programmatic access only. No dashboards, no Chrome extensions. We compare what your scripts and agents can actually call.

What Counts as an Ad Spy API (and What Just Pretends)

An ad spy tool shows you competitor ads in a browser. An ad spy API returns the same data as structured JSON your own code can act on. That difference sounds cosmetic. It is the entire purchase decision, because the market sells four different things under one label:

1. Native transparency APIs. Meta's Ad Library API is the canonical example. Built by the platform for regulatory transparency, free, and limited by design.
2. Commercial ad intelligence APIs. Paid products like the AdLibrary API that aggregate multiple ad libraries and layer ad intelligence signals on top.
3. Web tools with an API bolt-on. BigSpy and PowerAdSpy built browser products. API access exists somewhere near the top of their pricing pages, often as a sales conversation rather than a signup button.
4. Scrapers with an HTTP wrapper. Apify actors fetch Meta's public Ad Library pages and hand you the parsed result. You call them like an ad spy API. They behave like the scrapers they are.

Evaluate these four as one category and you will get burned on at least one axis. A scraper's pricing looks unbeatable until it returns empty arrays for a week. A native API's coverage looks fine until you need commercial ads in the US.

One scoping note: "ad spy API" as a search usually means Facebook first. Meta is where most paid-social budgets live, so every vendor leads with Meta coverage. The gaps show up on platform two.

Ad Spy API Comparison Table: Six Options Side by Side

Here is the short version of the whole ad spy API comparison. Each row gets a full section below.

Two things jump out. First, only one row is a genuinely multi-platform ad spy API, and the scraper row covers a single network. Second, "free" appears twice, and both free options are transparency surfaces with no commercial performance signals. If you want the deeper head-to-head on the web tools themselves, we covered BigSpy vs PowerAdSpy vs AdSpy and Meta Ad Library vs BigSpy vs AdLibrary separately.

The Three Axes That Actually Separate an Ad Spy API

Vendors compete on database size because it makes good landing-page copy. "One billion ads" is a vanity metric. In real competitive intelligence work, the differences that decide whether your integration survives are duller.

Axis 1: Platform breadth

Most ad spy data products are Meta databases with extras. That was defensible in 2020. In 2026, TikTok takes a serious share of DTC budgets and YouTube takes a serious share of everything else. The platforms know it: TikTok now runs its own Commercial Content Library and LinkedIn ships an Ad Library, both web-only. A multi-platform setup matters because your competitors do not run single-channel, and an ad spy API that only sees Meta will tell you a brand "went quiet" when it actually moved budget to TikTok. Check breadth at the API level, not the product level. BigSpy's web app browses nine networks. What its data API exposes, at which tier, is a question for sales.

Axis 2: Field depth

Two ad spy APIs can return "the same ad" and give you wildly different objects. The shallow version: creative URL, advertiser name, dates. The deep version adds full copy (title, body, caption, CTA), engagement counts, impressions, an estimated spend figure, and runtime. Depth decides what you can compute downstream. You cannot rank a competitor's portfolio by traction if every ad arrives as a thumbnail and a date. Note the word estimated: no third-party product knows true spend. Meta's own API reports ranges, and honest commercial vendors label spend as an estimate.

Axis 3: API or scraper in a trench coat

The structural question almost nobody asks during a demo. A real ad spy API is a contract: documented endpoints, stable fields, rate limits, error codes, someone paid to keep it working. A scraper is a bet that a third party's HTML stays still. The wrapper can look identical. The failure modes are nothing alike. When Meta ships a frontend change, an ad transparency scraper returns garbage until a maintainer notices, and your nightly job fails silently in the meantime. Ask one question of any vendor: "When the source page changes, who fixes it, and under what SLA?" A real API vendor has an answer. A scraper marketplace has a refund policy.

AdLibrary API: The Multi-Platform Commercial Option

Full disclosure up front: this is our product, so apply skepticism and check the live documentation yourself. Here is what it is, factually.

The AdLibrary API is a paid REST ad spy API that returns commercial ads across 11 platforms from one endpoint: Facebook, Instagram, TikTok, YouTube, Google, LinkedIn, Twitter, Pinterest, Yahoo, Unity Ads, and AdMob. Authentication is a single bearer key starting with adl_, created in the dashboard. No app review, no identity verification, no OAuth flow, no 60-day token expiry.

Field depth is the main argument for paying. Every ad comes back with the creative (image, video, carousel, or collection, with direct URLs), full copy including title, body, caption and button text, engagement counts, impressions, an estimated spend figure, a 0 to 1000 heat score for momentum, runtime in days, geo, and the landing page URL. Runtime deserves a highlight: it is the most reliable public tell that a creative works, since advertisers kill losers in week one.

Beyond search, the AI enrichment endpoint turns any single ad into a structured teardown plus a 1:1 replication brief. There is also a free cross-platform advertiser resolver, a curate endpoint that pulls every recent ad for a saved brand across platforms in one request, and a winners scan that scores an advertiser's whole portfolio to surface the creative intelligence answer everyone actually wants: which concepts did this brand scale.

Pricing is credit-based and published in euros. One credit per search, one per AI creative brief, advertiser resolution free, and a failed search refunds its credit automatically. API access sits on the Business plan at €329/mo with 1000+ credits per month, which also includes free integration help. Rate limits are 10 requests per minute and 10,000 per day per key, with a Retry-After header on 429s. The search response includes a total count, so you can size a niche before paging through it, and our ad spend estimator helps translate competitor activity into budget context.

The honest limitation: it is not free, and Meta's API is. Which brings us to the baseline.

Meta's Free Ad Library API: The Baseline Every Ad Spy API Gets Measured Against

Meta built the Ad Library after 2018 for political transparency, and the Ad Library API exposes that archive through the Graph API's ads_archive endpoint. It is free, official, and authoritative for what it covers. It deserves to be named the originator of this whole category, and if it covers your use case you should use it and pay nobody.

What it covers is the catch. Worldwide, the API returns political and social-issue ads. All ad types are queryable only for the EU and UK, a consequence of the Digital Services Act, and those commercial records stay available for about a year. A US e-commerce brand's ads are visible in the web UI but are not returned by the API. That single fact disqualifies Meta's free option for most commercial research, and it surprises nearly every developer who reads the marketing page before the limitations.

Implementation friction is real. You confirm your identity and location (this can take days), register a developer app, and generate an access token that expires every 60 days unless you build a refresh flow. Standard apps get roughly 200 calls per hour, shared across search and pagination, which makes deep crawls slow. Performance data is thin by design: impressions and spend arrive as bucketed ranges like 1000-5000, and only for political and issue ads.

None of this is a flaw. Meta built a transparency instrument, not a research product, and as a transparency instrument it is excellent. We wrote a full Meta Ad Transparency Center guide for the manual side. The mistake is expecting a free compliance API to behave like a commercial ad spy API. Meta's free API is fine for one platform and one ad category. The moment you need TikTok, YouTube, or LinkedIn data in the same query, you need something else.

BigSpy and PowerAdSpy: Web Tools With an API Somewhere in the Building

BigSpy and PowerAdSpy are the two names that always show up in an ad spy api comparison, and both are real businesses with real databases. The question for this article is narrower: what do they offer a developer?

BigSpy runs one of the largest ad creative databases in the category and its web app tracks around nine networks. Entry-level subscriptions are priced for casual browsing and are genuinely cheap. Programmatic access is a different animal. BigSpy's data API is positioned as an enterprise offering, gated behind its top-tier and custom plans, with scope and quotas settled through sales rather than a pricing table. Plenty of teams accept that. Just know going in that "BigSpy has an API" translates to "BigSpy will discuss an API," and budget calendar time accordingly. Our AdSpy review covers a similar pattern in an adjacent tool.

PowerAdSpy covers multiple networks in the browser and markets hard on filter count. It is a web product through and through. There is no public, self-serve ad spy API here as of this writing. Power users work around it with CSV exports and browser automation, which converts a paid web tool into an unofficial scraping target, with the same fragility problems as the Apify route minus the marketplace.

Neither point is a knock on the products as browsing tools. For a freelancer building a swipe file by hand, they can be exactly right. This is a programmatic comparison, and on the axis of "can my pipeline call it today," both score poorly. The deeper pattern: web-first products treat the API as a retention feature for big accounts, while API-first products treat the web app as one client among several. You can usually tell which one you are buying from whether the API docs are public.

Apify Facebook Ads Scrapers: A Scraper in a Trench Coat

Apify is a marketplace of "actors," community-built scrapers you trigger over HTTP and pay for by usage. Several actors target Meta's Ad Library, and they are popular for an obvious reason: you can start in ten minutes for pocket change. The actor loads the public Ad Library web pages, parses what renders, and writes the result to a dataset you fetch through Apify's well-documented platform API.

Be precise about what is an API here. Apify's platform is a real API with excellent engineering. The data acquisition underneath is a scraper, and inherits every structural weakness of scraping:

• Field depth is capped at the web page. The public Ad Library UI shows creatives, copy, and dates. It does not render engagement counts, spend estimates, or any performance signal for commercial ads. No scraper can return data that is not on the page.
• Fragility is built in. When Meta changes markup or rolls out new bot defenses, actors break. Fixes depend on an individual maintainer's availability, not an SLA. Read the reviews on any popular Ad Library actor and you will find the outage complaints between the five-star ratings.
• Terms of service. Meta's terms prohibit automated collection from its services without written permission. Enforcement against research-scale scraping is rare, but "rare" is doing heavy lifting if you are building a commercial product on top. Agencies with risk-averse clients should read that sentence twice.
• Silent failure is the expensive part. A scraper that returns 80% of the ads it used to does not throw an error. Your automation keeps running and your analysis quietly degrades.

The fair verdict: as an ad spy API for a one-off research dump, an academic project, or validating an idea before committing to a vendor, Apify actors are a fine choice. As production infrastructure for competitor monitoring, they are the option you will replace within a year. We went deeper on this whole category in Meta ad library scraping tools.

Google Ads Transparency Center: The Other Native Route

Google's answer is the Ads Transparency Center, covering verified advertisers across Search, YouTube, and Display. For YouTube it is the primary public source, which earns it a row here despite the headline fact: there is no public REST API for it.

What Google offers instead is arguably better for analysts: a BigQuery public dataset of Transparency Center data. If your team already lives in SQL, you can query advertiser-level creative records, regions, and run dates without managing any HTTP plumbing, and join the results against your own warehouse. It costs nothing beyond BigQuery query processing.

The limits mirror Meta's transparency surface. You get creative previews and run metadata, not engagement or spend signals for commercial ads. Coverage is verified advertisers. Freshness and field availability follow Google's transparency obligations, not your research needs. And because it is a dataset rather than a search API, "alert me when this brand launches a new YouTube creative" turns into scheduled queries you build yourself. We covered the YouTube angle in detail in our ad library alternative for YouTube ads piece.

Treat Google ATC the way you treat Meta's free API: an authoritative, free, single-vendor transparency source that pairs well with a commercial ad spy API rather than replacing one.

Implementation Effort: Your First Call on Each

Talk is cheap. Here is what minute one looks like on each ad spy API that accepts an HTTP request today.

AdLibrary API. Create a key in the dashboard (Business plan), then:

curl "https://adlibrary.com/api/search" \
  -H "Authorization: Bearer adl_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "keyword": "collagen supplement",
    "appType": "3",
    "platform": ["facebook", "tiktok"],
    "adsType": ["2"],
    "sortField": "-impression",
    "daysBack": 30
  }'

That is video ads mentioning "collagen supplement" on Facebook and TikTok from the last 30 days, sorted by impressions, one credit. The same call in Python:

import requests

r = requests.post(
    "https://adlibrary.com/api/search",
    headers={"Authorization": "Bearer adl_your_api_key"},
    json={"keyword": "collagen supplement", "appType": "3", "sortField": "-heat_degree"},
)
data = r.json()
print(f"{data['total']} ads, {data['_credits']['remaining']} credits left")

Meta Ad Library API. After identity verification and app setup, the ads_archive call itself is clean:

curl -G "https://graph.facebook.com/v23.0/ads_archive" \
  -d "access_token=YOUR_60_DAY_TOKEN" \
  -d "search_terms=collagen" \
  -d "ad_type=POLITICAL_AND_ISSUE_ADS" \
  -d "ad_reached_countries=['US']" \
  -d "fields=page_name,ad_creative_bodies,ad_delivery_start_time,impressions,spend"

The request is easy. The two weeks of verification and the token refresh flow before it are the actual implementation cost.

Apify actor. Trigger a run, then poll its dataset:

curl -X POST "https://api.apify.com/v2/acts/ACTOR_ID/runs?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"startUrls": [{"url": "https://www.facebook.com/ads/library/?q=collagen"}]}'

Input schemas vary per actor, so check the actor's README rather than assuming. Day one is painless. The cost shows up on the day the actor breaks.

Where this gets interesting in 2026 is agentic use. JSON with stable fields slots directly into AI agent workflows, and we publish end-to-end patterns for Claude Code with the adlibrary API and for building your own MCP server on top of it. Scrapers fit those loops badly because an agent cannot tell degraded output from a real market change.

Pick by Use Case: Six Verdicts

No single winner. Honest routing instead:

• Political or issue-ad research on Meta. Meta's free Ad Library API, no contest. It is the system of record, the price is zero, and bucketed ranges are acceptable for transparency work.
• EU-only commercial Meta research. Meta's free API is viable thanks to DSA coverage, if you accept thin fields and the verification process. Pair it with patience.
• Multi-platform competitor monitoring on a schedule. A commercial ad spy API is the only real answer, and this is the AdLibrary API's home turf: one key, 11 platforms, performance signals, and recipes for automated competitor ad monitoring through n8n or Zapier.
• Manual browsing and swipe files. You may not need an ad spy API at all. BigSpy or PowerAdSpy subscriptions, or AdLibrary's own web app on the Starter tier, cover hand research fine.
• One-off Meta dataset for a study. An Apify actor is defensible. Accept the fragility, verify the output sample by hand, and do not wire it into anything recurring.
• YouTube-first analysis with a SQL team. Google's Transparency Center BigQuery dataset, possibly augmented by a commercial API for engagement signals.

Whatever you pick, run the spend math first. If competitor intelligence informs a budget that a ROAS calculator says breaks even at 2.1, the monthly cost of any option here is noise next to one avoided losing creative test.

Frequently Asked Questions

What is an ad spy API?

An ad spy API is a programmatic interface that returns competitor ad data as structured JSON instead of a browser dashboard. You query by keyword, brand, platform, or format, and get back creatives, copy, and metadata your scripts, dashboards, or AI agents can process. It is the difference between reading competitor ads and computing on them.

Is there a free ad spy API?

Two free native options exist. Meta's Ad Library API returns political and social-issue ads worldwide, plus all ad types for the EU and UK only. Google's Ads Transparency Center has no REST API but offers a BigQuery public dataset. Both are transparency surfaces without commercial performance signals. Commercial multi-platform coverage with engagement and spend estimates is paid everywhere.

Are Apify ad library scrapers legal and reliable?

Apify's platform is legitimate, but Meta's terms of service prohibit automated collection from its services without written permission, so actors scraping the Ad Library operate in a gray zone. Reliability is the bigger practical issue: actors break when Meta changes its markup, and fixes depend on individual maintainers. They suit one-off research, not production pipelines.

Which ad spy API covers TikTok and YouTube?

Native transparency APIs do not: Meta's API is Meta-only and Google has no REST API. The AdLibrary API covers TikTok and YouTube alongside Facebook, Instagram, Google, LinkedIn, Twitter, Pinterest, Yahoo, Unity Ads, and AdMob through one endpoint. Web tools like BigSpy browse multiple networks, but programmatic access is gated to top tiers.

How much does the AdLibrary API cost?

API access is included in the Business plan at €329/mo with 1000+ credits per month. One credit per search, one per AI creative brief, and advertiser resolution is free. Failed searches refund their credit automatically. Lower tiers let you browse the same data in the web app first, so you can verify coverage before automating.

The Bottom Line on This Ad Spy API Comparison

Strip the marketing and the ad spy API field sorts cleanly. Meta's free Ad Library API is the originator and the right tool for political ads and EU transparency work. Google's BigQuery dataset is the quiet gem for SQL teams watching YouTube. BigSpy and PowerAdSpy are browsing tools whose APIs live behind sales doors. Apify actors are honest about their price and quiet about their fragility. And the AdLibrary API is the option built API-first for commercial, multi-platform research, with the performance fields that make ranking competitors possible at all.

The decision rule from the top of the article holds: ignore database-size claims and score every candidate on platform breadth, field depth, and the trench-coat test. The vendors that survive all three are a short list.

If your shortlist includes us, start where the data is visible: browse the coverage in the web app, then create a key on the Business plan (€329/mo, 1000+ credits, full API access) when you are ready to automate. The API access feature page has the three-step setup, and integration help from our team is included, so your first scheduled competitor pull can be live this week.