← Docs

Creative & Testing

Meta Ad Library for research

How to use the Meta Ad Library to study competitor and category ads: what it shows, what it hides, and how to turn what you find into test ideas.

YieldBI TeamGrowth ResearchUpdated Oct 2026

The Meta Ad Library is a public, searchable database of ads running across Meta’s apps. Anyone can open it without an account. Advertisers use it to see which messages, formats and offers other brands in their category are running right now.

What the Ad Library shows

For most commercial ads, the library shows the creative, the ad copy, the Page that ran it, the platforms it appears on, the start date and whether it is active. If an advertiser runs one creative across several ad sets, the library groups those copies and tells you how many versions exist.

Ads about social issues, elections or politics carry extra detail, such as spend ranges and audience reach, and Meta keeps them visible for longer. Meta’s current guidance describes what is retained and for how long, so check it before relying on historical data.

What it does not show

The library does not show results. You cannot see spend, conversions, cost per result or targeting for ordinary commercial ads. An ad that has run for months suggests the advertiser kept paying for it, which is weak evidence that it performs. It is not proof.

You can search by advertiser name, Page, or keyword, and filter by country, platform, media type and active status. Start with three to five direct competitors, then widen to adjacent categories that sell to the same buyer. Searching a problem phrase your customers use often surfaces advertisers you had not thought of.

What to look for

Read the ads as a set, not one at a time. Useful patterns include:

  • Which angles and messages repeat across many ads from one brand.
  • Formats: short video, static image, carousel, user-made style footage.
  • Offers and calls to action, such as free shipping or a quiz.
  • Ads that have been live a long time versus ones that vanished within days.

Long-running ads and large clusters of near-identical versions hint at what an advertiser has already tested and kept. Treat those as hypotheses to check in your own account.

Turning research into tests

Write down the concept behind each ad you save, not the surface details. “Founder explains the problem in the first three seconds” is a concept you can adapt. Copying a competitor’s exact wording is not, and it ties your brand to their message.

Then feed those concepts into your creative testing framework and test each as a distinct idea, not a minor variation. Your own data decides which ones survive.

Limits to keep in mind

You see only what is running or retained. Ads can be targeted narrowly, so a competitor’s best-performing ad may reach people who never see it in your feed. Check the library on a regular schedule, since active ads change quickly.

How YieldBI helps

Once you have concepts worth testing, YieldBI’s AI creative generation helps you produce variations quickly. Its ad-level signal analysis then shows which of your own ads are winning, so your decisions rest on your results instead of what a competitor appears to be running.