Your competitor's CTR does not exist
Every tool selling rival ad metrics is estimating. The one honest signal in the Ad Library is free: how long an ad has been allowed to live.
Researching a crowded FMCG category for a client, I wanted the obvious thing: how competitors’ ads actually perform. CTR, spend, impressions.
Here is what I confirmed the hard way: that data is not obtainable. Meta’s Ad Library API serves commercial ads only in the EU, where transparency law forces it; in India the API is political ads only. The public Ad Library website works without a login and shows exactly four things — the creative, the copy, the date it started running, and the variant count. No performance data of any kind.
So any tool or agency claiming to know a rival’s CTR is modelling, not measuring. Some models are reasonable. None of them are receipts.
The proxy that is honest
Longevity. Nobody pays to keep a losing ad alive. An advertiser who leaves an ad running is voting for it with money, every day, in public.
Rough calibration I now use:
- Two weeks — a test. Means nothing.
- Two months — validated. Someone checked the numbers and kept paying.
- Twelve months and counting — the strongest winner signal available without inside access. I found exactly one of these in the category, and it told me more than any scraped metric could.
Variant count is the secondary signal: multiple versions of the same concept means the concept earned iteration budget.
The method
Search the Ad Library by category terms, not brand names — the generic words a buyer would use. Brand-name searches show you rivals you already know; category searches surface the ones you don’t, which are usually the more instructive set. Note the start date and variant count for everything active, then look for agreement.
What agreement looks like
In this category, four advertisers with nothing in common — different sizes, different cities, different products — were all surviving with the same shape of ad: a problem stated plainly in vernacular language, an explicit offer, and a shop button. Constipation, bloating, quitting tobacco. Not taste, not tradition, not brand story.
When four advertisers who have never met independently converge on the same structure and keep paying for it, that is a finding, not a hypothesis. The one advertiser running pure brand-story ads was the category leader — who has brand money and can afford awareness.
My client’s live ads were selling heritage and taste on a ₹500-a-day budget: playing the leader’s game without the leader’s balance sheet. The category had already run the experiment and published the result; it just published it as dates rather than dashboards.
The general form
When the metric you want is unobtainable, don’t accept a fabricated version of it — find the decision that real money is making in public, and read that instead. Prices, ad longevity, hiring pages, repeated behaviour. An estimate answers faster. A revealed preference answers truthfully.