If you want to know how to find winning ads in the Facebook Ads Library, the short answer is: search by keyword and category instead of brand name, filter down to active ads, then look for the ones that have been running the longest or show multiple variations of the same idea. Those two signals — longevity and iteration — are the closest thing to a performance clue you'll get, since the library itself doesn't show spend, clicks or conversions. This guide walks through the exact process, the filters worth using, and what to do with what you find once you've saved it.
What "Winning" Actually Means in the Ads Library
The Facebook Ads Library is a transparency tool, not an analytics dashboard. It lists every ad an advertiser is currently running (and, for some regions, political and issue ads going back further), but it never shows impressions, spend, click-through rate or conversions. So when people talk about finding "winning" ads there, they really mean finding ads that show behavioral evidence of working — not ads with a scoreboard attached.
The two most reliable clues are how long an ad has stayed live and whether the advertiser is running several versions of the same core idea at once. Advertisers rarely keep paying to show an ad that isn't earning its spot, and they rarely build five variations of a hook that's already failing. Neither clue is proof, but together they're a reasonable filter for where to spend your research time.
How to Find Winning Ads in the Facebook Ads Library: A Step-by-Step Process
- Open the Ads Library and set your filters first. Choose the country you care about and set the ad category to "All ads" (not just "Issues, elections or politics") so you see standard commerce ads.
- Search by keyword instead of only brand name. Try the problem your product solves ("back pain relief," "desk organizer," "dog anxiety") rather than just competitor names — this surfaces advertisers you've never heard of who may be running the strongest creative in your niche.
- Switch to grid view and scan for pattern before detail. Look at thumbnails across many ads before you click into any single one; patterns in color, framing and opening text jump out faster this way than reading one ad at a time.
- Check how long each ad has been running. The library shows the start date for every active ad. An ad that's been live for several weeks while the same advertiser is also running five newer ads is worth a closer look.
- Open the advertiser's full page and count variations. If you see multiple ads with the same offer but different opening lines, product shots or captions, that advertiser is actively testing hooks — a strong sign they've found a format worth iterating on.
- Save candidates immediately, not later. Screenshot the ad, note the advertiser name, start date and what you noticed about the hook or offer. The library has no save or bookmark feature, so anything you don't capture now is gone the next time you refresh.
Reading the Start Date Correctly
A long run time is a decent signal, but context matters. A seasonal product (holiday decor, event tickets) will naturally have a short window no matter how well it performs, while an evergreen product (supplements, apparel, home goods) that's been running for months is a stronger indicator. Treat this as a starting filter to test your own instinct against, not a guarantee.
Search Tactics That Narrow Things Down Fast
Scrolling the library with no plan burns time. These tactics get you to relevant ads faster:
- Search by problem, not product name — "fix dry skin" surfaces more creative angles than "moisturizer."
- Search by format keyword — adding words like "before and after" or "unboxing" to your query can surface ads built around a specific proof format.
- Go direct to known competitors — if you already know a few players in your space, open their pages directly rather than relying on keyword search to find them.
- Browse adjacent categories — a skincare brand can learn a lot from how a supplement brand builds proof, even though the products differ.
- Repeat the search across a few countries — the same advertiser often runs different creative angles per market, which multiplies your sample size.
Signals Worth Trusting vs. Signals to Ignore
| Signal | What it suggests | How to verify |
|---|---|---|
| Ad has run for several weeks or more | Advertiser is likely still getting value from it | Check the start date shown on the ad and compare to how long similar products are typically promoted |
| Multiple versions of the same hook or offer | Advertiser is actively testing variations, suggesting the core idea works | Open the full advertiser page and count how many active ads share the same CTA or product |
| High production polish alone | Not a reliable signal by itself — big budgets don't guarantee performance | Weigh polish against longevity and variation count, not on its own |
| Ad appears on multiple placements (Feed, Reels, Stories) | Advertiser is confident enough to spread the same creative across formats | Check the placement preview shown for the ad, where available |
| Comment count or reactions on the ad post | Can indicate engagement, but says nothing about actual conversions | Use only as a tie-breaker between two otherwise similar candidates |
Building a Research Habit, Not a One-Time Search
The library refreshes constantly, and a single afternoon of browsing won't tell you much. Treat it like a recurring task: a short session once or twice a week, focused on one niche keyword at a time, is more useful than one long scroll. Keep a running file of what you find so you can spot patterns across sessions — the same hook structure showing up from three unrelated advertisers is a far stronger signal than any one ad on its own. If you want a system for organizing what you collect, see how to organize a swipe file of winning ads.
It also helps to study hooks specifically, since the opening seconds decide whether anyone keeps watching. For a deeper look at pulling hook ideas apart, read how to find winning ad hooks from competitors and why most video ads fail in the first three seconds.
Turning What You Find Into Your Own Ad
The point of this research is never to copy an ad word for word — that's both a legal risk and a creative dead end, since you're betting on an angle that's already saturated for someone else's audience. Instead, pull out the structure: what's the hook built around (a question, a bold claim, a visual surprise)? What problem does it name in the first few seconds? What proof does it lean on — a demo, a before/after, a testimonial? What's the actual offer and call to action?
Write those four or five structural beats down for each strong candidate you save. That list becomes your brief. From there, you can write your own script that fits your product, your voice and your actual offer, rather than a reskin of someone else's ad.
Common Mistakes When Mining the Ads Library
- Judging an ad by polish alone — a rough, low-budget looking ad that's been running for months is often a better signal than a slick one that just launched.
- Only searching exact competitor names — this misses the wider pool of advertisers solving the same problem differently.
- Not noting the start date before closing the tab — without it, you can't tell longevity apart from a brand-new test.
- Copying visuals or copy directly — this risks platform flags and still won't fit your product's actual proof points or offer.
- Treating one strong ad as the whole strategy — the stronger pattern usually shows up across several advertisers, not one lucky find.
How FrameNotion Fits In
Once you've pulled a structure from the Ads Library — the hook angle, the proof format, the kind of offer that keeps showing up — the next step is turning that into a finished ad for your own product. FrameNotion is built for that handoff: paste your product or website link, and FrameNotion AI reads the page and writes a custom 30-second vertical video ad from scratch, following the standard hook, problem, benefit, proof, offer, call-to-action structure you just researched. You can add your logo, up to six product images and notes about your offer, and the AI builds the script, voiceover, music and captions around it — a finished ad takes about 10 to 20 minutes.
It outputs in 9:16 at 1080×1920 for Meta ads, Reels and Shorts, plus 4:5, 1:1 and 16:9 versions of the same ad, so you can run the winning structure you found across every placement without rebuilding it. The first ad is free to try — see how it works or browse example ads before checking pricing.
Where to Go From Here
Finding strong ads in the library is only half the job. The other half is turning research into a script and a finished video before the opportunity window closes. If you want help writing that script from what you've learned, the AI script generator for Facebook ads guide covers how to brief an AI tool with your research so it writes something usable on the first pass.
Frequently asked questions
Is the Facebook Ads Library free to use?+
Yes. It's a public tool from Meta and doesn't require an account or payment to search and browse active ads.
Can I see how well an ad is performing in the Ads Library?+
No. The library shows creative, start date and sometimes placement, but it never shows spend, clicks or conversion data. Any read on performance has to come from indirect signals like how long the ad has run.
Does a long-running ad always mean it's profitable?+
Not always. Seasonal products naturally run shorter campaigns regardless of performance, and some advertisers keep underperforming ads live for brand reasons. Treat longevity as one useful signal to test against other clues, not a guarantee.
Can I download ads from the Ads Library to reuse them?+
You can view and screenshot ads for research, but reusing someone else's exact copy, footage or design risks platform flags and intellectual property issues. Use what you find to understand structure, then build your own creative.
How often should I check the Ads Library for new ideas?+
A short, focused session once or twice a week on a specific keyword tends to produce better patterns than one long occasional scroll, since you can compare what's new against what you saved before.