Swipe file & analysis · Creative Testing

Ad creative testing tools and Meta ads creative testing, sourced from live competitor ads

Most ad creative testing fails at the start, not the end. You launch four variants pulled out of a brainstorm, they all underperform, and two weeks of budget tells you nothing you could not have guessed. The bottleneck is never the test framework, it is the quality of the ideas going into it.

AdEye fixes the input. Every hook, angle and format in your test queue can start from a creative a competitor already ran for months, which means their spend paid for the first round of validation. Search any brand, filter to the ads with real run length behind them, and pull those patterns into your own testing calendar.

Ad Intelligence Feed

4 libraries live

Enter a brand or niche

Try:

Ad libraries

Public ad-library data only. Nothing behind a login.

ads found · showing top creatives

See it run

4 ad libraries, 1 feed · AI hook and angle tags · Public data only

META GOOGLE TIKTOK YOUTUBE

4 libraries one feed

AI-tagged hooks & angles

Why it works

What ad creative testing looks like with AdEye

Test ideas with evidence behind them

Filter to competitor ads that have been live for months. Those concepts already survived somebody else's testing budget, which makes them far better starting hypotheses than a blank whiteboard.

Isolate one variable at a time

AI tags separate hook, angle and format, so you can hold the offer steady and test only the opening three seconds, or keep the hook and swap the format from static to UGC video.

Read a competitor's test results for free

When a brand runs twelve variations of one concept and quietly kills the other four, that is a published test result. Variation count and run length tell you which way their tests landed.

What it surfaces

Every ad card carries the hook, angle and a working signal

You type a brand or niche. AdEye pulls the live public ads into one feed and AI-tags each creative, so you read the strategy and spot the proven winners before you save a single one.

Build a test queue from proven hooks
Tag every variant by hook, angle and format
See which concepts rivals kept scaling
Spot format gaps nobody is testing
Public ad-library data only

META Top performer

HOOK Before/after ANGLE Results UGC video

RUNNING 210d · 31 VARIATIONS

Why AdEye

One feed, the AI read, the working signal in view

Not four browser tabs and a stale doc. One query across every ad library, the strategy behind each creative, and a living swipe file with alerts. Public ad-library data only.

Every platform at once

Meta, Google, TikTok and YouTube searched in one feed for any brand or niche, so competitor ad research is one query instead of four manual hunts.

The AI reads the why

Every creative is tagged with its hook, angle and format, and a working signal flags the long-runners and scalers, so you see why an ad works, not just what it ran.

A swipe file that stays current

Save winners to boards and get alerted when a tracked competitor launches or scales a new ad, so your research never goes stale in a Google Doc.

Good questions

Questions about creative testing

Creative testing in Meta ads is running multiple ad variants against the same audience and budget to find which hook, angle or format performs best. Meta's delivery system rewards fresh, high-performing creative, so most accounts test continuously rather than once. The usual structure is one campaign, a broad audience, and several distinct creative concepts competing for spend.

Three to five genuinely different concepts per test is the practical range for most budgets. Fewer than three and you learn little; more than five and each variant gets too few impressions to reach a reliable read. Test distinct concepts first, then produce variations of whichever one wins.

Long enough to clear the learning phase and gather meaningful conversion volume, which for most accounts means at least five to seven days and roughly fifty conversions per variant. Calling a winner after two days of noisy data is the single most common creative testing mistake.

Not directly, because no public ad library reports spend or performance for commercial ads. What it can show is the honest proxy: how long each ad has been live and how many variations of it exist. A concept running four months with a dozen variations passed the advertiser's own testing; a one-week ad with no variations probably did not.

AdEye is the research half. It does not run your tests or report your results inside Meta Ads Manager. It supplies the input: a searchable feed of live public competitor ads, tagged by hook, angle and format, so the concepts you push into your testing framework start from evidence instead of guesswork.

Read every question we get

Explore more