Ad Transparency Data in 2026: Every Platform's Library Compared

Five years ago, "ad transparency data" meant one thing: Meta's archive of political ads. In 2026 it means seven public libraries, each built under different pressure, exposing different fields, in different formats, with different retention rules. Nobody documents the deltas. Platforms publish their docs in isolation, and the cross-platform comparison of ad transparency data you actually need (what can I get, where, for how long) doesn't exist in one place.

TL;DR: Every major platform now runs an ad library, but the ad transparency data inside varies wildly. Meta exposes the most fields and the only mature API. Google's Transparency Center covers Search, YouTube, and Display but keeps records thin and short-lived. TikTok, LinkedIn, Snapchat, Pinterest, and X built theirs mostly because the EU's Digital Services Act forced them to, so EU researchers get reach and targeting data the rest of the world never sees. None expose performance. This guide compares all seven libraries field by field, then covers unified access to ad transparency data for teams who refuse to maintain seven integrations.

This guide is for the strategist mapping what competitor data is even available, and the developer who has to pull it. Both questions have the same answer, and it starts with a table.

What Counts as Ad Transparency Data in 2026

The term covers everything a platform discloses publicly about the ads it serves: the creative itself, the advertiser behind it, when it ran, where it ran, who it reached, and sometimes what the advertiser paid. Ad transparency started as an electoral-integrity project after 2016, which is why political ad transparency data remains the richest tier almost everywhere.

Regulation changed the landscape. The EU's Digital Services Act took full effect for very large online platforms in August 2023, and its Article 39 requires each of them to maintain a public ad repository covering all ads shown in the EU — content, advertiser, payer, run dates, targeting parameters, and aggregate reach, retained for one year after the ad last appeared. The DSA even mandates that repositories be queryable through APIs, though "API on paper" and "API you'd build on" turn out to be very different claims.

So ad transparency data in 2026 splits into three tiers:

1. Political ad data: the deepest tier. Spend, impressions, and funding entities, often retained for seven years (Meta, Google).
2. EU commercial ad data: DSA-mandated. Every ad shown to EU users, with targeting and reach, kept for a year.
3. Everything else: active ads only, minimal fields, and on some platforms a window as short as 30 days.

That third tier is the one most competitor research lives in, and it's the thinnest. A platform's "ad library" means different things depending on which side of the Atlantic you sit on. If you want the conceptual grounding first, our primer on what an ad library API actually is covers how programmatic access to ad transparency data works end to end.

Ad Transparency Data by Platform: The Table Nobody Publishes

Here is the side-by-side, current as of mid-2026. "Fields" describes a typical non-political record. Political ads get more almost everywhere.

Three patterns jump out of the ad transparency data comparison.

The EU gets a different internet. Five of the seven libraries are essentially DSA-compliance artifacts. Outside the EU, TikTok, LinkedIn, Snapchat, Pinterest, and X publish little to nothing about commercial ads. EU ad transparency data is the only place you'll find targeting parameters and reach for those platforms at all.

Ad transparency data retention is the silent killer. Google's commercial window is about 30 days. Meta shows only active ads to non-EU users, so the moment a campaign stops, it vanishes from your view. Anyone doing historical creative analysis is fighting the clock on every platform, which is why serious teams archive ads continuously rather than searching retroactively.

APIs are the exception, not the rule. One mature API (Meta), one BigQuery dataset (Google, political only), one application-gated research API (TikTok), and four platforms with nothing programmatic. The DSA's API requirement produced some technically-compliant endpoints and file dumps, but nothing a production system would call on a schedule.

The rest of this guide walks each library, then the obvious follow-up: working across all seven without burning a quarter on integrations.

Meta Ad Library: The Deepest Archive, With the Tightest Strings

Meta's Ad Library remains the reference implementation for ad transparency data. The web UI shows every active ad on Facebook and Instagram worldwide: creative, copy, start date, platforms, and multiple versions when an advertiser runs variants. For EU-served ads it adds total reach and demographic breakdowns. For political and social-issue ads, it adds spend bands, impression bands, and funding entities, retained for seven years.

The API is where strings attach. Access requires identity verification (a government-ID process that can take days) plus a registered developer app, and the token expires every 60 days unless you build a refresh flow. Once in, the API returns political and social-issue ads globally — but commercial ads only when they ran in the EU. A US-only Nike campaign is visible in the web UI while it's live and invisible to the API, period.

A typical call looks like this:

curl -G "https://graph.facebook.com/v21.0/ads_archive" \
  -d "access_token=YOUR_TOKEN" \
  -d "search_terms=running shoes" \
  -d "ad_reached_countries=['DE']" \
  -d "ad_type=ALL" \
  -d "fields=ad_creative_bodies,ad_delivery_start_time,page_name,eu_total_reach"

What comes back is honest but bounded: creative text, dates, page identity, EU reach. What never comes back is performance. No engagement, no clicks, and impressions only as bucketed ranges on political ads. Spend appears only as ranges, only for political ads. We've documented the full list of constraints in Meta Ad Library API limitations, and the practical workarounds for exporting Meta Ad Library data at scale.

One 2026 wrinkle worth knowing: facing the EU's incoming political-ads regulation, Meta stopped running political and social-issue ads in the EU entirely in late 2025. The political archive remains, but the richest tier of EU ad transparency data is now a historical dataset there rather than a live feed.

Google Ads Transparency Center: Wide Coverage, Thin Records

Google launched the Ads Transparency Center in March 2023, and its scope is genuinely impressive on paper: every ad from every verified advertiser across Search, YouTube, and Display, searchable by advertiser name or website.

Then you open a record and find almost nothing in it. The creative, the advertiser's verified name, the format, the regions it showed in, and roughly when it last ran. No run-length precision, no reach outside the EU, no copy variants grouped, no landing pages. The retention window for commercial ads is approximately the trailing 30 days, so a campaign's history evaporates within weeks. EU users get the DSA tier (targeting parameters, reach, one-year retention), and political ads live in a separate transparency report with multi-year retention and a public BigQuery dataset you can query with SQL.

There is no general-purpose API. If you want Google and YouTube ad transparency data in a pipeline, you're choosing between the political-ads BigQuery dataset (narrow), scraping (against terms, brittle), or a third-party provider. We've covered the programmatic angle in our guide to the Google Ads Transparency Center API situation, and what it means for YouTube ad research specifically.

The strategist's takeaway: Google's library answers "is this advertiser running ads right now?" and nearly nothing else. Treat it as a verification tool, not a research tool.

TikTok Commercial Content Library: Built for Brussels

TikTok's Commercial Content Library exists because the DSA demanded it, and it shows. The library covers ads and paid commercial content shown to users in the EU (plus a handful of regions TikTok has added since), with the advertiser's name, the run dates, the targeting parameters (age, gender, interests) and the number of users reached. Records persist for a year.

For an EU strategist, that's real ad transparency data. Targeting parameters in particular are something Meta only shows in aggregate and Google barely shows at all. The catch is geography: search for a US-only TikTok campaign and you'll find nothing, because the library's obligation stops at the EU border.

Programmatic access exists through TikTok's Research API, but it's application-gated. You describe your research purpose and wait for approval, and commercial competitive research isn't what the program was designed to approve. For most marketing teams the realistic options are the web UI or third-party coverage, which we've mapped in our complete TikTok Ad Library guide.

LinkedIn, Snapchat, Pinterest, and X: The Long Tail

The remaining four libraries are smaller, and each has exactly one thing worth knowing about its ad transparency data.

LinkedIn quietly ships one of the better DSA implementations. Its Ad Library covers every ad served since June 2023, searchable by company, keyword, and country, with records kept for a year after the last impression. EU-served ads show impression ranges broken down by country plus the targeting criteria the advertiser used — unusually granular for B2B research, where knowing a competitor targets "IT decision-makers, Germany, company size 200+" is the entire insight. No API though. Our LinkedIn Ad Library deep dive covers the search operators, and there's a programmatic workaround guide for teams who need it in code.

Snapchat splits in two. Its political ads library is genuinely excellent — full-precision spend, impressions, paying entity, and targeting for every political ad since 2018, downloadable as yearly CSV files. Its commercial transparency is EU-only DSA compliance. The gap between those two tiers is the widest of any platform, something we unpack in Snapchat ads transparency.

Pinterest runs an EU Ads Repository with creative, advertiser, dates, targeting, and reach, retained one year. Notable mostly for what's absent: Pinterest banned political ads in 2018, so there is no political tier at all. Commercial research options are covered in our Pinterest ad library alternative breakdown.

X is the cautionary tale. It shut down its original Ads Transparency Center years before the DSA, and its compliance response, an EU-only ads repository delivered substantially as bulk file downloads, has been criticized by researchers since launch as hard to query and incomplete. If X creative research matters to your roadmap, expect manual work. Our guide on analyzing X ads shows what's realistically extractable.

How the EU DSA Rewrote Ad Transparency Data

What the DSA actually requires explains nearly every row of the comparison table, so let's be precise about it.

Article 39 obliges very large online platforms (those with more than 45 million EU users) to maintain a public repository of every ad presented in the EU. Each record must include the ad content, the advertiser, who paid, the dates of presentation, whether it targeted specific groups and the main parameters used, and aggregate counts of recipients reached. Records stay up for one year after the ad's last impression, and the repository must support multi-criteria search plus API access.

Five of the seven libraries in this guide exist primarily because of that paragraph. The asymmetry it created is the single most important fact about ad transparency data in 2026: an analyst in Berlin can see targeting parameters and reach for a TikTok or LinkedIn campaign, while an analyst in New York researching the same advertiser sees nothing. There's no US equivalent (proposed legislation like the Honest Ads Act never passed), so the EU repositories became the de facto global research surface, with the obvious blind spot that they only contain ads which ran in the EU.

The API mandate deserves its own asterisk. The DSA says repositories must be accessible through application programming interfaces, but compliance quality ranges from Meta's documented (if restricted) API down to X's file dumps. Regulators have noticed: the European Commission opened proceedings against X in 2023 citing, among other things, ad-repository deficiencies. The legal floor exists. The engineering floor varies.

One more 2026 shift: the EU's separate political-advertising regulation (the TTPA) imposed requirements strict enough that Meta and Google both chose to stop selling political ads in the EU rather than comply. Net effect: EU commercial ad transparency data keeps improving while EU political ad transparency data is partly going dark, an inversion of how this entire field started.

What's Still Missing From Every Library

Put all seven libraries together and you still can't answer the questions that actually drive media decisions. The gaps in the available ad transparency data are consistent across platforms.

No performance signals. Not one library tells you whether an ad worked. No engagement counts, no clicks, no conversions. Impressions, where present, arrive as bucketed ranges. The strongest available proxy is runtime (advertisers kill losers fast), and you have to derive it yourself from repeated observation.

No spend, or spend in fog. Outside political tiers, no platform discloses spend. Estimating competitor budgets means combining reach signals with market CPMs. Our ad spend estimator and CPM calculator exist precisely because official ad transparency data won't do this for you, and we've documented the full spend estimation workflow for Meta.

No history. Thirty days on Google, active-only on Meta outside the EU, one year in the DSA repositories. Want to study how a competitor's creative strategy evolved over three years? No official source holds that. Continuous archiving is the only fix.

No cross-platform identity. The same brand is a page ID on Meta, an AR-prefixed advertiser ID on Google, a company ID on LinkedIn. No library links them. Cross-platform portfolio research starts with manual entity resolution every single time.

No creative intelligence. Libraries hand you raw creatives, never analysis. Extracting the creative strategy layer (hooks, offers, formats, angles) from ten thousand raw records is your problem.

These gaps define the market for paid ad intelligence tooling. Official libraries answer "what ran?" Everything a competitive intelligence workflow needs beyond that lives in third-party data.

Unified Access: One Key Instead of Seven Integrations

Suppose you want competitor ads flowing into your own stack: a dashboard, a weekly report, an AI agent. You have three realistic routes to programmatic ad transparency data.

Route one: build against official sources. Meta's API plus the Google political BigQuery dataset plus manual exports from the rest. Free, legitimate, and very limited: you'll cover political ads well, EU ads partially, and US commercial ads on five platforms not at all. Budget weeks for Meta's app review and the 60-day token refresh dance.

Route two: scraping. Against every platform's terms, brittle the moment markup changes, and increasingly defended against. Teams that go this way spend more time on maintenance than analysis.

Route three: a commercial ad library API. This is the category AdLibrary sits in, so the positioning matters: Meta's Ad Library API is free and ours is paid, and if political ads on Meta are all you need, use theirs and keep your money. The paid product exists for everything the free tier structurally can't do — commercial ads, multi-platform coverage (Facebook, Instagram, TikTok, YouTube, Google, LinkedIn, Twitter, Pinterest, Yahoo, Unity Ads, AdMob), and performance signals like estimated spend, a heat score, impressions, and runtime on every ad. One adl_ key, no app review, no token expiry.

One search across all eleven platforms looks like this:

curl "https://adlibrary.com/api/search" \
  -H "Authorization: Bearer adl_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "keyword": "protein powder",
    "appType": "3",
    "platform": ["facebook", "tiktok", "linkedin"],
    "geo": ["US", "DE"],
    "sortField": "-impression",
    "daysBack": 30
  }'

Each result carries the fields official ad transparency data withholds: spend estimate, heat score (a 0–1000 momentum signal), impressions, runtime, and geo, all as plain JSON. One credit per search, refunded automatically if the search fails.

The cross-platform identity problem gets its own endpoint, and it's free:

curl -G "https://adlibrary.com/api/advertisers/search" \
  -H "Authorization: Bearer adl_your_api_key" \
  --data-urlencode "q=gymshark" \
  --data-urlencode "country=US"

That resolves one brand name to its Meta page ID, Google advertiser ID, and LinkedIn company ID in a single call, and tells you when the platforms agree it's the same advertiser — the entity resolution you'd otherwise do by hand for every competitor. Save the brand once and a single curate call then pulls its recent ads across platforms, deduplicated.

We've benchmarked the paid field honestly in our ad spy API comparison and the broader free vs paid ad library API trade-off analysis. The short version: if your research is occasional and political, free official sources win. If it's recurring, commercial, and multi-platform, the build-vs-buy math tips fast.

A Working Setup for Cross-Platform Ad Research

Here's how a strategist-developer pair typically operationalizes ad transparency data, regardless of which sources they choose.

Step 1: define the competitor set as entities, not names. Resolve each brand to platform-level IDs first. Free official option: search each platform's library by hand and record the IDs. Faster option: the advertiser-resolution call shown above.

Step 2: archive continuously. Because retention windows are short, schedule pulls rather than searching retroactively. A nightly job against a ten-competitor list catches new creatives the day they launch. Our Python cookbook has working scripts for exactly this pattern, and the end-to-end monitoring guide covers alerting.

Step 3: dedupe to concepts. A brand running one creative in eight variants is one idea, not eight. Dedupe on a stable ad key before any analysis, or your "most active advertiser" charts measure variant count instead of strategy.

Step 4: derive the performance proxy. Track first-seen and last-seen dates per creative. Runtime is the most reliable public signal that an ad converts, since underperformers get killed within days. Sort by it.

Step 5: layer analysis on top. Whether that's a weekly dashboard, a swipe file, or an LLM that writes creative teardowns, this is where the value concentrates. The whole competitor ad research workflow and the broader cross-platform strategy use case both assume steps 1–4 are already automated.

Teams running this on official sources alone usually cover Meta well and accept blindness elsewhere. Teams that need all seven platforms in one searchable surface, or the eleven the commercial APIs reach, typically end up on a paid key. For context on the manual-tool side of that decision, our ad spy tools comparison covers the app-based options too.

Frequently Asked Questions

What is ad transparency data?

Ad transparency data is the information platforms publicly disclose about ads they serve: the creative, the advertiser's verified identity, run dates, regions, and (for political and EU-served ads) targeting parameters, reach figures, and sometimes spend. Each platform publishes it through its own ad library, with no shared standard for fields, formats, or retention.

Which platform exposes the most ad transparency data?

Meta, by a wide margin. Its Ad Library shows all active ads worldwide, adds reach and demographics for EU-served ads, retains political ads with spend and impression ranges for seven years, and offers the only mature public API. Google covers more surfaces (Search, YouTube, Display) but keeps far thinner records with roughly 30-day commercial retention.

Why do EU users see more ad transparency data than US users?

The EU's Digital Services Act requires very large platforms to publish every ad shown in the EU with targeting parameters and reach, retained for one year. No equivalent US law exists, so TikTok, LinkedIn, Snapchat, Pinterest, and X built EU-only repositories. The same advertiser's campaign is documented in Europe and invisible in the US.

Does ad transparency data show how well an ad is performing?

No. No official library exposes engagement, clicks, or conversions, and impressions appear only as bucketed ranges where they appear at all. The strongest public performance proxy is runtime, which you derive by tracking first-seen and last-seen dates yourself — advertisers keep winners running and kill losers fast. Paid ad intelligence APIs add spend estimates and momentum scores per ad.

Is there one API for every platform's ad transparency data?

Not an official one. Meta has the only mature first-party API, and it's limited to political and EU ads. Commercial providers fill the gap: the AdLibrary API, for example, searches eleven platforms (including Facebook, Instagram, TikTok, YouTube, Google, and LinkedIn) with one paid key on the Business plan, adding spend estimates and heat scores that official libraries don't publish.

Where Ad Transparency Data Goes From Here

The trajectory is uneven but legible. Regulation keeps deepening the EU tier, and the EU tier alone. Retention stays short everywhere it isn't mandated. Performance stays out of every official library, permanently, because no platform will publish numbers that let outsiders audit its auction. Ad transparency data will keep telling you what ran and keep refusing to tell you what worked.

For a strategist, the comparison table above is the map: know which platform can answer which question before you promise a competitive analysis. For a developer, the build decision is narrower than it looks. Official APIs cover one platform well, and everything beyond that is either fragile scraping or a paid key.

If the recurring, multi-platform, automated version is what you need, that's exactly the workload the AdLibrary API was built for: eleven platforms behind one unified search, spend estimates and heat scores on every record, and AI creative teardowns on demand. API access ships with the Business plan at €329/mo with 1000+ monthly credits and hands-on integration help — see pricing for the full breakdown, and start pulling the ad transparency data your competitors assume nobody is watching.