How to See Competitor Ads: Three Free Libraries and What They Hide

The short answer: three official libraries cover most of what you need, and all three are free. Meta Ad Library shows every ad a page is currently running. Google Ads Transparency Center shows what an advertiser is paying for in search, which tells you which queries they believe convert. TikTok Creative Center is the only one of the three that publishes actual performance numbers. None of them show spend or return, so everything you find is a hypothesis to test rather than a result to copy.

These are public registers, not loopholes

Meta, Google and TikTok publish ad data because transparency regulation requires it. You are reading a public record, and using it is ordinary competitive research.

That origin also explains the shape of the data. These libraries were built to disclose what was published and who paid for it. They were never built to report how any of it performed, which is why the most common mistake is reading a long-running ad as a proven winner.

Meta Ad Library

The library covers ads running across Facebook, Instagram, Messenger, WhatsApp, Threads and Audience Network. Search a page name and you get every ad that page currently has live, with the date each one started running.

The duration heuristic is sound: an ad still running after a month is usually working, because nobody keeps paying to serve a loser. Two limits change how you should use it.

There are no performance figures for commercial ads. No impressions, no clicks, no click-through rate, no spend. Duration is the only signal, and you are inferring from it rather than reading it.

Commercial ads disappear when they stop. Only ads about social issues, elections and politics stay in the library, and those are retained for seven years. A regular product ad vanishes from the record the moment the advertiser pauses it. You cannot reconstruct a competitor's history from the library. You can only take repeated snapshots, which means the workflow is a recurring check with your own archive, not a one-off browse.

The Europe exception is worth knowing. Under the Digital Services Act, ads served in the EU and UK carry much more disclosure: the targeting parameters the advertiser selected, a demographic breakdown of who actually saw the ad, the paying and benefiting entities, and one year of retention after the last impression. If a competitor runs anywhere in Europe, the European version of their campaign is far more revealing than the version visible elsewhere. This is worth checking even for a regionally focused brand, because most international advertisers run European campaigns alongside everything else.

Google Ads Transparency Center

Available at adstransparency.google.com, with two useful entry points: advertiser search, using a verified legal name or domain, and keyword search, which surfaces how different brands position around a term.

Meta shows you creative. Google shows you intent, and that is the more commercially loaded signal. Paid search keywords are a competitor's own statement about which queries are worth money to them. A brand bidding on the same term month after month has almost certainly measured that it converts, because paid search punishes sentiment faster than any other channel.

You can filter by region, ad format across text, image and video, and date range, including custom ranges.

The read that pays off: pull an advertiser's search ads and list the phrases that recur across headlines. Those are their tested value propositions, already filtered by their own money. Then compare that list against the terms you rank for organically. Anything they pay for and you already rank for is leverage you are not using.

TikTok Creative Center

Free at creative-center.tiktok.com, with no funded ad account required, which makes it the least gated of the three.

It is also the only one that hands you performance data. In the Top Ads dashboard you can sort by reach, click-through rate, two-second view rate and six-second view rate, and filter by industry across more than twenty verticals, plus region, campaign objective, language and format.

The two-second and six-second rates are the ones to watch, because they isolate the hook from everything downstream. Sort your category by two-second rate, watch the opening three seconds of the top results, and record the structural move rather than the content: a question, on-screen text before speech, a face in frame immediately, the problem shown before the product, the result shown first.

Copying the creative is a losing game, since you are always a month behind and working with someone else's product. What transfers between categories is the structure.

Where their traffic actually comes from

Creative is one layer. Acquisition mix is another, and reading the first without the second produces bad conclusions. A competitor growing mostly through organic search is playing a different game from one buying every impression, and their ad library will mislead you if you do not know which you are looking at.

Similarweb estimates the split across search, social, direct and referral. The absolute traffic numbers deserve suspicion, especially for smaller sites, but the proportions hold up well enough to tell you which channel a competitor actually depends on. We cover where that fits in the SEO intelligence stack.

The click-through method

The ad is the top of the funnel. Clicking through is how you see the rest of it, and this is the step most people skip.

What to record when you land:

• Whether the page matches the ad promise, or dumps traffic onto a generic homepage
• The offer: free trial, discount, demo, consultation, lead magnet
• Form length, and which fields are required
• What occupies the space above the fold
• The proof on the page: client logos, review counts, specific numbers
• Whether each ad has its own page, or every ad points to the same one

That last one is the most diagnostic. A brand running twelve ads into one homepage is not really testing anything. A brand running twelve ads into twelve pages is measuring properly, and is the more serious opponent.

For regional work, the landing page is also where you find out whether a competitor is running translated copy or copy actually written in Arabic, which is a gap worth knowing about and a subject in its own right.

Keep a record, because the platforms will not

Since Meta drops commercial ads the moment they stop, your archive is the only history that will exist. A simple sheet is enough:

Column	Why it matters
Date captured	The ad may be gone next week
Advertiser and platform	Grouping
First seen, still live	Your only duration proxy
Hook, first line verbatim	The transferable part
Offer	What they lead with
Landing page URL and screenshot	Pages get edited without notice
Format	Static, video, carousel

A weekly capture is enough for most categories. Monthly is enough for slow ones.

What none of this tells you

Spend is not published for commercial ads on any of the three platforms. Neither is return, and an ad can run for months while losing money if the brand is funding a launch or measuring the wrong outcome.

The deeper limit is survivorship. You are seeing what is running now, not the thirty variants killed in week one, and that discarded set is where most of the learning happened. You inherit the conclusion without the experiment.

None of it establishes that an ad worked for them, let alone that it will work for you against a different audience at a different price with a different offer.

The honest summary

These libraries narrow your guesses. They tell you which claims a competitor is willing to pay for, which hooks are earning attention in your category, and how seriously the funnel behind the ad has been built.

They do not replace your own testing, and the teams that get the most from them are the ones treating each finding as a hypothesis with a test attached rather than a formula to copy.

Sources

Platform behaviour was checked against each publisher's own documentation rather than taken from secondary write-ups. Ad libraries change; re-check the retention and disclosure rules before relying on them.

• Meta Ad Library. The library itself.
• Meta Ad Library tools, Transparency Center. Coverage, retention rules and research access.
• Google Ads Transparency Center. Advertiser and keyword search, region, format and date filters.
• TikTok Creative Center. Free access without a funded ad account.
• TikTok, How to use the Top Ads dashboard. The sort and filter options, including CTR and view-rate metrics.
• Similarweb. Channel-mix estimates used in the traffic section.