Facebook Ad Library Product Research: A Founder's Guide
The Facebook Ad Library (now the Meta Ad Library) is a free, public search of the ads running on Facebook and Instagram. For product research, search the words a buyer would use, filter to active ads in your country, and look at the start date of each ad. An ad that has run for 30 days or more means a company has kept paying to sell something close to your idea. It isn’t proof of profit, but it is a strong public hint that selling works, and you can check it in minutes.
Listings show that someone wants to sell. Long-running ads suggest that selling works well enough for someone to keep paying. That difference is why the Ad Library belongs in every idea validation routine, next to marketplaces and Reddit.
Why is the Ad Library useful before you build?
Most early research measures talk: threads, videos, likes. Ad spend is different, because it costs the advertiser money every day. Most advertisers stop ads that don’t pay back, so an ad still running after weeks has usually passed that test. Usually, not always: some brands keep ads running for awareness, and some sellers simply forget to turn them off.
The library also shows you the words that sell. The first line and the headline of a long-running ad are often the result of someone else’s testing. You can read the promises your market already responds to before you write a single line of your own.
The risk of skipping it: you guess at demand from views and comments, then find out after launch that nobody in the category can make paid traffic work.
How do you search the Ad Library for a product idea?
- Open the Meta Ad Library and pick the “All ads” category.
- Choose your country. Ads differ a lot by market.
- Search a phrase a buyer would type (“adhd planner”, “notion template”, “meal planning app”), not a brand name.
- Filter to active ads.
- Open the ads and note the advertiser, the start date and the first line.
Repeat with two or three phrases. One phrase rarely catches the whole market.
What should you look at in each ad?
| Field | Where to find it | What it tells you |
|---|---|---|
| Start date | “Started running on” | How long the ad has survived. 30+ days is a strong signal |
| Advertiser | The page name | Whether many brands advertise or one owns the space |
| Number of versions | Multiple versions on one ad | The advertiser is testing, so the product matters to them |
| Headline and first line | The ad text | The promise that sells in this market |
| Link | The landing page | Price, offer, guarantee, bundle |
| Format | Image, video, carousel | What creative the category runs on |
Write these into the same sheet you use for finding competitors, so ads and listings sit side by side.
How do you tell real demand from a one-off test?
Look at duration and repetition, not volume alone.
- Many ads, most started this week: lots of testing, not much proof yet.
- A few ads running for months: a good signal. Someone has likely found an ad that pays back.
- Many advertisers with long runners: a proven, competitive market.
- One advertiser owns almost every ad: demand exists, but you would be fighting a big budget.
Here is one example check Vettoz ran on 2026-10-18, for an undated paper planner for adults with ADHD in the US. The Ad Library search for “adhd planner” returned 960 active ads, and 12 of the ads Vettoz read had been running for 30 days or more. Together with about 178,000 units sold on the top TikTok Shop listing, that was a strong sign people pay in that category. The same check also flagged that “room” was the weakest dimension: a busy ad market is a crowded one.
Compare that with a team hydration-reminder SaaS from the same day: only 7 active ads in the US, Reddit threads mostly about electrolyte drinks rather than software, and no priced competitors. That idea scored 46 out of 100.
What are the limits of Ad Library research?
Be honest about what you can’t see:
- No spend or sales numbers for most product ads. You infer from duration, not revenue.
- Not every seller advertises on Meta. Some live on search ads or marketplaces. Check TikTok Shop and Google too.
- Brand ads aren’t product ads. A big company may run awareness ads that say nothing about one product line.
- Countries differ. A crowded US market can be empty elsewhere, and the reverse.
Use the library as one source among several. On its own it tells you that selling works. With Reddit comments it tells you why buyers buy. See how to find pain points on Reddit.
How do you turn ads into positioning?
Copy every first line into a list and group them by promise. You will usually see three or four promises repeated by everyone (“stay organized”, “finally stick to a routine”, “made for ADHD brains”). Those are table stakes. The opening is the promise nobody makes that buyers ask for in comments and reviews. That is the start of brand positioning for a new product.
What should one Ad Library session leave you with?
After thirty to sixty minutes you should have a short sheet: the advertisers for your phrases, how many of their ads have run 30 days or more, the first line of each long runner, and the landing page price. From that you can answer three questions. Does paid traffic work in this category? Who spends the most? What promise do they all make? If you can’t answer the first one, look at organic channels before you plan an ad budget. If one advertiser dominates, plan a narrower angle than theirs.
FAQ
Is the Facebook Ad Library free?
It is. Anyone can search it without an account or an ad budget.
How long should an ad run before it counts as proof?
There is no official rule. Thirty days is a practical line, not a guarantee: most advertisers don’t keep paying for an ad that loses money for a month, but some do.
Can I see how much an advertiser spends?
Not for ordinary product ads. Spend ranges are shown for political and issue ads only, so infer from how long ads run and how many versions exist.
Does the Ad Library work for SaaS and apps?
Yes. Search the job (“invoice software”, “habit tracker”) and you will see software and app advertisers alongside physical products. Pair it with App Store research for apps.
Read the ads with Vettoz
Vettoz searches the Meta Ad Library for your idea’s main keyword in your market, keeps the longest-running ads with their start dates and promises, and counts how many are active and how many have run 30 days or more. He adds Reddit, TikTok, YouTube, Google and marketplaces, scores the idea with the evidence linked, and your AI writes the report inside Claude or ChatGPT. See how it works, read the setup guide, or check 2 ideas free.