Competitor Ad Research With the Meta Ad Library: A Step-by-Step Guide

Meta ad library competitor research is the fastest free way to see what your competitors are running on Facebook and Instagram right now, how long each ad has been live, and which angles they keep investing in. It does not show spend or results for most ads, so the skill is in reading the signals it does give: run time, the number of variations, and what gets repeated. This guide walks through a five-step process you can finish in an afternoon for a set of competitors, then shows how to keep it running without redoing the work every month.

Quick answer: To research competitors in the Meta Ad Library: 1) list 8 to 12 direct and adjacent competitors, 2) search each advertiser and filter to active ads on your target platforms and country, 3) sort by start date and treat ads live for 30 days or more, or with many variations, as likely winners, 4) tag each one by hook, format, offer and awareness stage, 5) turn the patterns into 3 to 5 test hypotheses for your own account. Repeat monthly or automate the tracking.

What the Meta Ad Library Shows and What It Hides

The Meta Ad Library is a public, searchable database of ads running across Facebook, Instagram, Messenger and the Audience Network. For any advertiser you can see active ads, the creative, the primary text and headline, the platforms it runs on, the start date and in many cases whether multiple versions exist. For ads about social issues, elections or politics, and in some regions for all ads, it shows more: spend ranges, impressions and audience demographics. For most commercial ads it does not.

So the library tells you what competitors are willing to keep paying for, not how well it performs. That is still valuable. An advertiser does not run an ad for 90 days unless it is at least covering its cost. And an advertiser does not make 15 variations of one concept unless the concept is worth iterating. Treat run time and variation count as your proxies for performance, and treat the library as a map of what the market has already validated.

Step 1: Build Your Competitor List

Start wider than your direct competitors. A list of 8 to 12 advertisers works well: four or five direct competitors who sell the same thing to the same buyer, three or four adjacent brands who sell to the same buyer but a different product, and two or three aspirational brands in other categories whose creative you admire. Adjacent and aspirational brands are where the fresh angles come from; direct competitors mostly tell you what the category already says.

1. Direct: the brands your customers compare you to in reviews and support tickets.
2. Adjacent: brands your customers also buy (check your post-purchase surveys or shop-the-look data).
3. Aspirational: two or three brands whose ads you save; often larger spenders in a different vertical.
4. Market-specific: if you sell in several countries, add one or two local competitors per market.
5. Record each as a row in a sheet: brand, Facebook page name, category, why it is on the list.

Refresh this list quarterly. Brands enter and leave categories, and the one that matters most next quarter is often the one spending heavily on Meta for the first time, which the library will show you before you notice it anywhere else.

Step 2: Search and Filter the Meta Ad Library for Competitor Research

Open the Ad Library, pick the country you sell in, set the ad category to All ads, and search the advertiser's page name. Open the page's results rather than the keyword results, because keyword search mixes in unrelated pages. Then apply filters. The ones that matter most for commercial research are platform (so you see only Instagram, for example), media type (video versus image), active status, and the date range for when the ad started.

A filter recipe that works

• Country: your target market. Repeat per market if you sell in several.
• Active status: Active only on the first pass. Add Inactive later to see what they stopped running.
• Platform: Instagram and Facebook separately if placements matter to you.
• Media type: Video first, then Image. Compare the mix: a brand that is mostly video is telling you where its results come from.
• Impressions by date: not available for most commercial ads; skip it.
• Keyword search within an advertiser: use for offer words such as free, bundle, sale, guarantee.

If you want a walkthrough with the interface on screen, this tutorial covers the search, filters and reading of results in a practical way.

Step 3: Identify Winning Ads by Run Time and Variations

With a competitor's active ads in front of you, you want to separate the tests from the winners. Three signals do most of the work. Run time: an ad that started 30 or more days ago and is still live is probably profitable, and 90 or more days is a strong signal. Variations: the library shows when an ad has multiple versions, which usually means the advertiser is iterating a concept that works. Recurrence: the same hook or offer appearing across several ads, often with different footage, means the concept has survived more than one round of testing.

Save the strong and moderate ads to a swipe file with a screenshot, the library link, the start date and the date you saved it. Check back in 30 days: if it is still running, upgrade your confidence. If a brand's long-running ads suddenly disappear, they may have found something better, which is your cue to look at what is new.

Step 4: Break Down Hooks, Formats and Offers

Research only becomes useful when you tag it. For every saved ad, record a handful of attributes in your sheet. The point is not to copy the ad but to see the distribution: what share of validated competitor ads lead with a problem versus a claim, how many use a founder versus a customer, which offers recur. Patterns across 40 to 60 tagged ads from 10 advertisers are far more informative than any single ad.

Tags to record

• Hook type: problem, bold claim, question, social proof, demonstration, curiosity, comparison.
• Format: UGC talking head, founder, product demo, static with text, carousel, before and after, listicle.
• Offer: discount, bundle, free gift, free shipping, guarantee, no offer.
• Awareness stage: unaware, problem aware, solution aware, product aware.
• Proof: reviews, numbers, press, expert, none.
• First three seconds (video): what is on screen and what is said.
• Landing page type: product page, collection, advertorial, quiz, lead form.

When you finish, look for gaps as well as patterns. If every competitor leads with a discount and none leads with a guarantee, a guarantee hook is an uncrowded test. If nobody is running comparison ads, you can own that frame. The Ad Library shows you the consensus; your opportunity is often just outside it.

Step 5: Turn Research Into Your Own Test Plan

Convert the patterns into hypotheses, not copies. A hypothesis names the angle, the audience assumption behind it and the metric you expect to move. Then produce two or three creatives per hypothesis in your own brand voice and test them against your current control. Three to five hypotheses per research round is plenty; more than that and the testing budget gets too thin to read results.

Run the test for at least a week or until each variant has 20 to 30 results, whichever is longer. Then feed the results back into the sheet: which borrowed angles worked for you and which did not. Over a few rounds this becomes a record of what your market responds to, independent of any competitor.

How to Monitor Competitors Continuously

A one-off research sprint goes stale in a month. Competitors launch new angles weekly, and the most valuable signal, a new ad that then survives 30 days, only shows up if you look again. You have three options: a calendar reminder to re-run the five steps monthly for your top competitors, a scraping or tracking tool that saves new ads from the library automatically, or an AI layer that watches the advertisers you name and summarises what changed. For agencies with many clients, manual monthly checks do not scale; the second or third option is the practical choice.

• Monthly manual: 60 to 90 minutes per client for 8 to 12 competitors. Fine for one or two brands.
• Tracking tools: save new ads and their run time automatically. Check the vendor's current pricing and whether they respect Meta's terms.
• AI competitor tracking: watches named advertisers, flags new and long-running ads, and summarises angles into the creative brief.
• Whatever the method, review the output in the same meeting where you plan next week's creative, so the research actually changes what you produce.

How Loraloop Fits

Loraloop includes competitor monitoring as part of the ads workflow. You name the competitors you care about, and Angie, the ads agent, tracks their ads and content, flags what is new and what has been running for a long time, and uses those observations when she generates ad creative and copy for your brand, grounded in your Brand DNA so the output sounds like you rather than them. Lora includes notable competitor moves in the daily briefing. Everything Angie drafts waits for your approval before it runs. It does not reveal competitor spend or results, because the Ad Library does not expose them, so the same proxies of run time and recurrence apply.

Frequently Asked Questions