Meta ad library

Meta renamed Facebook the company to Meta Platforms in October 2021. Most of the consumer-facing surfaces - the Facebook app, the Instagram app, Messenger - kept their existing brands; what changed was the parent company name and a series of product-page rebrands across the developer and advertiser ecosystem. The 'Facebook Ad Library' became the 'Meta Ad Library' in advertiser-facing copy roughly around the same window, but the underlying URL, data model, and access controls did not move.

This matters because keyword research and SERP rankings split: people still search 'facebook ad library' (the original name, deeply indexed) and 'meta ad library' (the new name, less indexed but rising). Google treats them as separate queries despite identical intent. If you're optimizing for organic discoverability on this topic, you write for both - which is why we keep a separate guide for 'facebook ads library' covering the same product from the legacy-keyword angle. Same tool, two queries, two pages.

In day-to-day use, call it whatever your team is used to calling it. The product is the same. Meta's own documentation flips between the two names depending on the page; there's no internal consistency to be respected.

The single most-important structural feature of the Meta Ad Library is that it discloses different data for different ad categories. Political and social-issue ads - the original regulatory target - disclose spend ranges, impression ranges, audience demographic breakdowns, and funding-entity attribution. Commercial ads - the vast majority - disclose creative and platform delivery only.

This split exists because the original library was built to satisfy regulators (the US Honest Ads Act framework, the EU's transparency requirements, similar laws in the UK, Brazil, and elsewhere). Commercial-ad disclosure was never legally required; Meta chose to surface the creative voluntarily as a transparency goodwill gesture. They have no obligation to widen the commercial disclosures and structurally have no incentive to either - more disclosure helps competitors more than it helps Meta.

Practical implication for marketers: don't expect commercial-ad disclosure to deepen. Plan your competitive workflow around the data Meta does disclose (creative, format, start date, placement), and source the data Meta doesn't disclose (spend, performance, audience) from other places - earnings reports, third-party spend estimators, your own attribution stack.

Most teams open the Meta Ad Library, scroll through 30 ads of a competitor, screenshot what looks interesting, and call it research. That's not an audit; it's a vibes check. A real audit produces a tagged, searchable corpus that compounds over time. Three rules separate the two:

Tag every saved ad against the same schema. Hook archetype (problem-solution, day-in-the-life, founder-to-camera, transformation, demo, testimonial, etc.), format (UGC, studio, motion, screen-record), and angle (price, status, ease, identity, comparison). After 100-200 tagged ads from your competitive set the meta-pattern emerges - which archetypes dominate your category, where the saturation is, where the whitespace is.

Capture the start date alongside the file. A 200-day-old ad and a 5-day-old ad sitting in the same folder are different signals; they need to be separable later. Build the folder structure as competitor / YYYY-MM-DD-started / hook-archetype.mp4.

Rerun the audit weekly, not as a one-off. Compounding intelligence comes from the second pass, not the first. The brands that pull ahead on creative are the ones whose ops team does 15 minutes per competitor per week for 12 months straight.

Two notable library updates shipped quietly between Q4 2025 and Q1 2026. The first: variant grouping - creative variants of the same ad now collapse under a single entry by default, instead of cluttering the result set with near-duplicates. Useful for high-velocity advertisers whose result pages were previously dominated by 10-card variant explosions of the same concept.

The second: start-date sort - you can now sort results by start date directly rather than scrolling chronologically. This sounds trivial; it's not. The fastest way to find a brand's evergreen winners is to sort by 'started oldest' and skim the top 5-10 results. Pre-2026 you had to scroll to the bottom of the unsorted list; post-2026 it's a single click.

Beyond these two, nothing structural has changed. The commercial-vs-political bifurcation persists. The API access scope persists (EU + Brazil for commercial, global for political). The disclosure gaps persist. Plan accordingly.