How to Use the Meta Ad Library to Find Winning Ads (2026 Guide)
The Meta Ad Library is free, public and underused. Here is a step-by-step method for finding the ads in your niche that actually keep selling, and turning them into creative briefs.
Most brands open the Meta Ad Library, scroll through a competitor's ads for ten minutes, screenshot a few that look nice, and close the tab. That is browsing, not research. The library holds far more signal than that, if you know which filters to use and what to look for.
This guide walks through the Meta Ad Library step by step: what it shows, how to search it, how to separate winning ads from tests, and how to turn what you find into briefs your team (or your agency) can execute.
What the Meta Ad Library actually shows
The Meta Ad Library lives at facebook.com/ads/library. It is a public, searchable archive of every ad currently running across Meta's properties: Facebook, Instagram, Messenger and the Audience Network. You do not need an ad account or a login to use it.
For each active ad you can typically see:
- The advertiser's page name and profile image
- The ad creative (image, video or carousel) and its primary text, headline and call to action
- The platforms it runs on
- A line reading "Started running on" followed by a date
- Whether multiple versions of the ad exist (Meta groups near-identical ads)
What you generally cannot see for ordinary commercial ads is spend, impressions, clicks or conversions. The exceptions are ads about social issues, elections or politics, which show extra transparency data including spend ranges, and ads delivered in the European Union, which show additional reach and targeting information.
That missing performance data is why most people underuse the library. The trick is that one of the visible fields, the start date, is a strong proxy for performance.
Step 1: Set up your search correctly
Open the library and set three things before you type anything:
- Country. Pick the market you sell into. An ad may run in the US but not the UK, and the results change accordingly. Choose "All" only when you want the widest view.
- Ad category. Leave this on "All ads" for commercial research. The "Issues, elections or politics" category is a separate archive.
- Search type. You can search by keyword or by advertiser. Keyword search looks across ad text; advertiser search pulls up a specific page.
Start with keyword searches for the problem your product solves, not your product category name. "Can't sleep" or "back pain at desk" surfaces ads written in your buyer's language. "Melatonin gummies" mostly surfaces brands that describe themselves the same way you do.
Step 2: Use the filters that matter
Once results load, open the filters panel. The ones worth using:
- Active status. Set this to active ads only. Inactive ads are history; you want what is still being paid for today.
- Platform. Filter to Instagram if your audience skews younger or visual, or Facebook if it skews older. Most ecommerce brands run both.
- Media type. Separate images from videos. Static and video ads have different jobs, and you will brief them differently.
- Impressions by date. This lets you narrow to ads that started in a specific window, which is the most useful filter for finding long runners.
Sort mentally by the "Started running on" date. An ad that started three days ago is a test. An ad that started four months ago and is still active is a decision someone made, repeatedly, to keep paying for it.
Step 3: Separate winners from tests
This is the core skill. Brands that run Meta seriously launch many creatives and kill most of them within days. What survives is what pays. So the age of an active ad is your best free performance signal.
A rough framework many media buyers use:
| Ad age (still active) | What it usually means |
|---|---|
| Under 14 days | A test. No conclusion yet. |
| 14 to 30 days | Survived the first cut. Worth noting. |
| 30 to 60 days | Likely profitable or strategically important. |
| 60+ days | Strong signal the ad keeps selling. Study it closely. |
The 60-day threshold is not magic, but it filters out almost all the noise. Few brands will keep funding an unprofitable ad for two months, especially direct-to-consumer brands watching cash flow every week. We explain the reasoning in depth in why long-running ads are your best creative research.
Two more signals strengthen the case:
- Multiple versions. If the library shows "This ad has multiple versions" or the same concept appears in several sizes and variations, the brand is investing in it.
- Duplication across pages. Some brands run the same winning concept from several pages or with creator partners. That is a scaling move.
Step 4: Read the ad for its message, not its look
Once you have a shortlist of long runners, resist the urge to copy the visual. The visual is the least transferable part. What transfers is the message underneath. For each ad, write down three things:
- The promise. What outcome does the ad say you will get? "Fall asleep in 20 minutes" or "a desk setup that stops neck pain."
- The proof. Why should you believe it? Reviews, a demonstration, a founder story, an ingredient explanation, a guarantee.
- The offer. What does it ask you to do and what do you get for doing it? A bundle, free shipping threshold, quiz, trial size, first-order discount.
This promise, proof and offer breakdown is the part that keeps working across formats and even channels. A worksheet helps. Paste the ad text into the winning ad analyzer and it will pull those elements apart for you.
Step 5: Look for patterns across advertisers
One long-running ad tells you about one brand. Five long-running ads from five brands that all lead with the same promise tell you about the market.
Build a simple spreadsheet with a row per long-running ad and these columns: advertiser, start date, format, hook (first line or first three seconds), promise, proof type, offer, landing page type. After 20 to 30 rows, patterns jump out. You might find that every surviving ad in your niche uses a demonstration rather than testimonials, or that the bundles outlast the percentage discounts.
Those patterns are your creative strategy. They tell you which angles the market has already validated with real money.
Step 6: Check the landing page
Click through the ad's link (from the library preview, the CTA usually links out) and look at where it sends people. Note whether it is a product page, a collection, a quiz, a listicle or a long-form advertorial. Long-running ads often point to dedicated landing pages rather than the homepage, and the page tells you a lot about how the brand thinks its buyers decide.
Pay attention to message match: does the headline on the page repeat the ad's promise? Usually, on the winners, it does.
Step 7: Turn findings into briefs
Research is only useful once it becomes new creative. For each pattern you found, write a short brief:
- Angle: the promise you will lead with
- Hook options: three opening lines or visual openers
- Proof: what you will show to make the promise believable
- Offer: what you will put in front of the click
- Format: static, UGC video, demo, carousel
- What we will not do: copy the competitor's wording, imagery or brand assets
That last line matters. You are borrowing validated ideas, not someone else's creative. Copying ads word for word or reusing images is a fast route to IP complaints and, more practically, makes you look like a knockoff. The creative brief generator can turn your notes into a clean brief in a few minutes.
Common mistakes when using the Meta Ad Library
- Only searching direct competitors. Some of the best messages in your niche come from brands two steps away. A sleep supplement brand can learn plenty from mattress and weighted-blanket advertisers.
- Ignoring inactive status. Researching inactive ads shows you what failed or what ran seasonally.
- Treating new ads as winners. A brand's newest ad is just its latest guess.
- Copying the visual. The visual is specific to their brand. The message is what transfers.
- Doing it once. Markets shift. A quarterly pass keeps your research current, and a monthly glance at your top three competitors catches new long runners early.
If you want to go deeper on the legal and ethical side of researching specific competitors, read how to see your competitors' Facebook ads legally.
Using Meta Ad Library research beyond Meta
Here is what most people miss: the messages you validate in the Meta Ad Library are not Meta-specific. A promise that keeps selling on Instagram is a promise your buyers care about, period. That makes the library a research input for other channels too, including email subject lines, landing page headlines and, increasingly, ChatGPT ads.
ChatGPT ads show up below answers to questions people ask while they are deciding what to buy. If the long-running Meta ads in your niche all promise "fall asleep faster without grogginess," there is a good chance buyers are also asking ChatGPT how to fall asleep faster without feeling groggy the next day. The message translates directly into an answer-shaped ad.
How SecondWin uses the Meta Ad Library for you
SecondWin automates exactly this research. You enter your URL, and it reads your site to understand your product, buyer and niche. It then mines the Meta Ad Library for the longest-running ads in that niche, extracts the proven promise, proof and offer, and maps each message to the questions buyers ask ChatGPT. From there it writes original ChatGPT ads (never copying competitor copy or images), checks them against OpenAI's ad policies, and runs them in your own OpenAI ad account.
The research step is free. Run a free analysis of your site to see which proven messages exist in your niche, and check SecondWin pricing if you want the campaigns managed for you.
FAQ
Is the Meta Ad Library free to use?
Yes. The Meta Ad Library at facebook.com/ads/library is free and public. You do not need a Facebook account, a business account or an ad account to search it. You can view any active ad running on Facebook, Instagram, Messenger and the Audience Network, filter by country, platform and media type, and see when each ad started running. Spend and performance data are not shown for ordinary commercial ads.
Can you see how much a competitor spends in the Meta Ad Library?
Not for ordinary commercial ads. Spend ranges are shown for ads about social issues, elections or politics, and ads delivered in the EU show extra reach and targeting transparency. For everything else, the best proxy is run time. An ad that has been active for 60 days or more is very likely profitable, because few brands keep funding losing creative for two months.
How do I know if an ad in the library is a winner?
Look at the "Started running on" date for active ads. Ads still running after 30 to 60 days have survived repeated decisions to keep spending. Stronger signals include multiple versions of the same concept, the same creative running from several pages, and a dedicated landing page. A brand-new ad is just a test, regardless of how polished it looks.
Is it legal to copy ads from the Meta Ad Library?
Researching ads is fine; copying them is not. The library exists for transparency, and studying angles, offers and formats is normal practice. Copying another brand's exact wording, images, video or brand assets can infringe copyright and trademarks and may get your ads rejected. Extract the underlying message, then write and produce your own original creative around it.