Free ad tool
Most ad tests end too early, on too few results, with a winner picked by gut. Part one tells you how many days your budget needs to give each ad enough results. Part two takes clicks and conversions for ad A and ad B and tells you whether the gap is real or noise. You type in the numbers from Ads Manager.
The planner splits the daily budget evenly across the variants, then divides by your cost per result. With three ads, $60 a day and a $20 cost per result, each ad earns about one result a day, so 50 results each takes about 50 days. Those are example numbers, not a recommendation.
If the answer runs past three weeks, the tool says so. You can shorten it three ways: test fewer variants, raise the budget, or optimise for a cheaper event that happens more often, such as add to cart or a lead, before you test on purchases.
The same numbers give you the ad testing cost: variants x results per variant x cost per result. Two ads at 50 results each and $20 a result comes to $2,000. Know that number before you start, because stopping halfway wastes what you already spent.
Ad B can beat ad A by chance, especially on small numbers. The checker runs a two-sided two-proportion z-test on the conversion rates and reports the confidence that the gap is real. At 95% or more, keep the winner and move on. Between 80% and 95% it leans one way but isn't proven, so keep both running. Below 80%, there is no real difference yet.
Why 95%? At that level you will crown a false winner about one time in twenty. At 80% it is one time in five, which means a lot of budget moved on noise. Use the same rule for every test so results stay comparable.
Don't stop a test the first time it crosses 95%. Confidence swings up and down as results arrive, and checking every day until it looks good inflates the false winners. Decide the number of results in part one, wait for it, then read part two once.
Test one variable at a time, or you won't know what caused the difference. Start with the hook (the first line or first three seconds), because it decides whether anyone sees the rest. Keep the body, offer and landing page identical across variants. Draft options with the AI hook generator, run them through the ad hook checker, and put the best two or three into the test.
Once a hook wins, test the body copy behind it, then the offer, then the format. A full loop for doing this every week is in our ad creative testing system.
This online ad testing check runs in your browser. It does not connect to Ads Manager or Google Ads, read your account or run tests for you. It covers the maths half of a creative testing platform or ad testing software: you bring the numbers, and it tells you how long to wait and whether to believe the result. To dig into why an ad won, the ad analysis tool breaks a creative down by hook, angle and format.
Enough for the gap to clear 95% confidence. Small differences between ads need many more results than big ones. Set a target per variant in the planner, wait for it, then check significance.
Yes. Enter link clicks as visitors and purchases as conversions, or impressions and clicks to test CTR. Click-through tests finish faster, but a better CTR does not always mean more sales.
Yes, and it works as a meta ads testing tool for Instagram too, or for Google and any other platform. You type in the numbers each ad got. It does not connect to any ad account.
It is how likely it is that the gap between ad A and ad B is not chance. At 95%, a gap this size would appear by luck about one time in twenty.
Keep going: Facebook Lead Ads Testing Tool, Google Ad Mockup Generator, LinkedIn Ad Mockup Generator, all free ad tools, the ad hook bank, and 3,000+ free Canva ad templates.
Take the angles you spotted into the AI Ad-Making Kit, or start with the free hook and copy tools.