Meta Ads Library vs. LinkedIn Ad Library: which is more useful for B2B research?

If you sell software at $20K+ ACVs, the LinkedIn Ad Library is your primary competitive research tool and the Meta Ads Library is your early-warning system. LinkedIn is likely where your competitors put the campaigns they back most strongly. Treat Meta as the place to look for hooks they may test first.

For ordinary commercial ads, neither library shows spend, CTR, or conversions. LinkedIn's useful targeting data is limited to ads targeted to the EU, while Meta provides additional targeting and demographic transparency for ads that deliver in the EU (and, for Meta, the UK).

After the paid specialist spots the ad, outbound may never hear about it, and the finding dies in a tab.

What the Meta Ad Library shows a B2B researcher

Search a competitor's Facebook Page name and Meta returns every active ad under that Page: creative, primary text, headline, CTA button, and start date. Filters cover image and video formats, with carousel also available. Click the CTA and you get the destination URL.

For a US-only commercial advertiser, the ads do not appear in the Meta Ads Library at all, so there is no dataset shown.

Commercial ads are visible only while active if they have no EU or UK impression. When a US competitor pauses a campaign, the ads disappear from the library as soon as they stop running. Only ads delivered to the EU or UK stay archived for a year after the last impression.

The Ad Library API is worse for our purposes. It returns nothing for non-EU/UK commercial ads, so you cannot pull a US competitor's ads programmatically at all.

EU-served ads add estimated total reach and beneficiary and payor legal names, with an age, gender, and country breakdown behind the "View details" button. Those categories are the ceiling.

Interest audiences, lookalikes, and retargeting lists never appear for any commercial ad, EU or not.

What the LinkedIn Ad Library shows

LinkedIn's library covers ads shown since June 1, 2023 and keeps each ad for one year after its last impression. That retention applies globally, which already beats Meta for tracking a US competitor over time.

Every record shows ad preview, ad format, advertiser name, payer name, and restriction status. Formats include single image, carousel, video, event, document, and Thought Leader Ads. Filters cover company name, payer name, keyword, country, and date range.

For ads targeted at EU members, LinkedIn discloses:

• First and last impression dates
• An estimated impression range
• An impression split by country
• Full targeting parameters: job function, seniority, title, skills, company size, revenue, industry, and named companies, each flagged as included or excluded

Meta gives you age and gender. LinkedIn gives you something closer to the actual persona brief.

Restricted ads hide the preview, advertiser, and payer. Do not assume you will see Message and Conversation Ads in the library.

Why the EU rows are richer: DSA Article 39

Both platforms expose run dates, reach, and targeting for EU ads because the EU's Digital Services Act makes them. Article 39 requires very large online platforms to publish a searchable repository with the ad content, the payer, the period presented, and the main targeting parameters, including exclusions.

Aggregate reach by member state is also required. Each entry must stay available for one year after the last presentation.

Outside the EU (and the UK for Meta), neither company volunteers any of it.

Run every LinkedIn search twice: once in your home market, once with an EU country selected. Compare the fields shown in each pass. Selecting Germany or Ireland can expose run dates, impression bands, and targeting for ads served there, even when the competitor is a US company.

If your competitors never serve ads in Europe, you are stuck with the thin US-only view on both libraries. That is a hard ceiling, and no third-party tool fixes it.

Where LinkedIn wins for $20K+ ACV research

For $20K+ ACV research, prioritize LinkedIn. It tends to reach mid-market and enterprise buyers, while Meta provides more context on SMB-oriented messaging.

Thought Leader Ads are the first thing to hunt for. Because every record distinguishes the advertiser from the payer, search by payer as well as company name.

Treat format as a funnel hypothesis. Document Ads may point to asset-led lead capture; Thought Leader Ads may point to top-of-funnel trust. "Book a demo" and "Download the guide" ask the same person for different levels of commitment, and a shift from one to the other across a competitor's ads usually means a go-to-market change.

When every competitor you track pushes "Book a demo," test an ungated asset such as a benchmarks report to fill the obvious offer gap.

Understory co-founder Alex Fine searched by category keywords and the names of well-funded competitors, filtered for the longest continuous run dates, and broke down the long-running ads in the library. He found four recurring moves: named reports, dollar-amount hooks, time-bound prompts, and pain-point copy that gave buyers something concrete to evaluate.

He split each long-running ad into three parts: hook, proof, and CTA, with each part in its own column in the log, so one competitor's dollar-amount hook sits next to every other hook in the category.

Once several competitors are side by side, the proof and offer columns show which type nobody is using. He looked for a test in that empty slot.

Where Meta wins: it often runs ahead of LinkedIn

Use Meta to form a hypothesis about a competitor's next LinkedIn campaign. Meta's CPM can differ from LinkedIn's, which makes it a practical place to pressure-test hooks before running them on LinkedIn.

The Meta library also exposes destination and variant details LinkedIn's library does not. Click through and you see whether a competitor sends traffic to an Instant Form or an external landing page, which can suggest whether the campaign prioritizes volume or qualification.

In our reviews, destination URLs sometimes carry a parameter like utm_campaign=mof-retargeting, which we treat as a possible funnel-stage hint. We also compare visible Reels against founder-voice content in Thought Leader Ads.

Variant count carries more weight on Meta than you would expect. Treat multiple versions as a hypothesis that the advertiser may be actively testing.

Meta's asset-mixing feature can inflate those counts: it lets Meta mix multiple images or videos, text options, headlines, and descriptions, so many apparent variants are auto-generated combinations rather than distinct concepts. Dedupe by headline plus first line of body copy before you count anything.

The blind spots both libraries share

LinkedIn's ad library doesn't show clicks, conversions, A/B lineage, retargeting lists, or ABM account lists. Meta's Ad Library only shows limited spend and reach data for certain political and social-issue ads, not general campaign performance metrics like clicks or conversions.

Start from the assumption that you do not know whether a competitor's ad is working. Even LinkedIn's EU targeting block shows the message an advertiser chose, not the full audience: a headline aimed at CFOs could be running against finance managers, founders, or a matched account list.

Longevity is the only performance proxy we use, and we treat thresholds anywhere from 30-90+ days as hypotheses rather than proof. An ad running unchanged for 90+ days either has a big budget behind it or nobody is optimizing it, and from the outside you cannot tell which.

Running both without drowning in tabs

Pick five advertisers: three direct competitors and two adjacent companies that sell to your buyer. Then:

1. Search each in the LinkedIn library twice: once in your home market, once with an EU country selected.
2. Search the same five in Meta.
3. Log only what you can see: hook, format, offer, CTA, destination page, payer, dates.
4. Write inferences on a separate line labeled as hypotheses.

After that first pass, run a weekly 15-minute sweep across LinkedIn, Meta, and Google's Ads Transparency Center, ending each sweep with one testable campaign hypothesis rather than a swipe file.

If you are a technical founder without bandwidth for a weekly sweep, the minimum version is the twice-per-competitor LinkedIn search once a month, plus a Meta check whenever a competitor launches something.

A finding usually dies in a tab when nobody shares it with the teams that can use it. At Understory, we pair GTM engineering with paid media research so one team can turn a finding into an ad test and a trigger-based outbound sequence in the same week.

For one anonymized SaaS client, we searched the LinkedIn library by product-category keywords, filtered for the longest-running ads in the category, and used those long-running ads to shape our LinkedIn creative.

We paired those ads with outbound triggered by a recent CRO hire or a fresh funding round, then synced the work into the client's CRM through OutboundSync. Competitor research was one input among several in that engagement, alongside the outbound trigger itself.

Turn ad library research into pipeline with Understory

Understory runs LinkedIn ad campaigns and signal-based outbound on Instantly and HeyReach, with creative on the same allbound team, so competitor research becomes live tests instead of a slide nobody reads. Schedule a demo to pair Clay-powered outbound with ad library research, the coordination behind Rivial Security's scale to $70K spend.