Ad creative testing on Meta and TikTok
Launching five ads and keeping whichever one looks best isn't a test. It's a guess. Creative testing tells you which part of an ad actually drives results. Here's what to test, how to set up a clean test on Meta and TikTok, which metrics to read and when to call a winner. If you need ads to test, we make motion graphics video ads in 24 business hours, starting with a 3 video trial for $97.
What creative testing is and why you test one variable
Creative testing means comparing versions of an ad under equal conditions so you learn something you can reuse. If two videos change the hook, the music and the offer all at once, you'll know which one won but not why.
That's why the core rule is one variable per test. TikTok builds it into the tool: a Split Test only allows one variable. Meta's A/B testing guidance says the same. Before you launch, write your hypothesis in one sentence, like "opening with the problem beats opening with the product."
What to test first
Start with what changes people's reaction the most and save the details for later.
| Variable | What you change | What stays the same |
|---|---|---|
| Hook | First frame and on-screen line | The rest of the video |
| Angle | The reason to buy: time, price, ease | Product, format and offer |
| Format | Demo, list, comparison, problem and solution | Message and offer |
| Offer | The incentive in the close | Everything up to the call to action |
Hooks are usually the cheapest test because you reuse the same body; see video ad hooks and video ad ideas. Swapping a color rarely teaches you much. TikTok advises clear differences between versions.
Torn between video and images? Read video ads vs static ads.
Not sure how many ads to put in each ad set? See how many ad creatives to test.
To see what changes when Meta builds the variations for you, read Advantage+ creative.
If you run a consumer app or a mobile game and want installs, see video ads for mobile apps and games.
How to set up the test on Meta and TikTok
On Meta. Ads Manager lets you build an A/B test from the toolbar or from Experiments, by duplicating an ad or picking two existing ones. Meta splits the audience so nobody sees both versions, recommends the same budget for each and suggests running at least 7 days (30 max). If you use Advantage+ creative, keep its automatic enhancements identical on both versions.
On TikTok. A Split Test divides the audience into two equal groups that don't compete. Creative variables include the video, the copy, the call to action and the initial hook. TikTok recommends at least 7 days, a budget with at least 80% estimated testing power (Ads Manager shows it) and no ad group changes once the test is live.
On both. On Meta, a significant edit can send an ad set back into the learning phase. Without a formal test, TikTok suggests 3 to 5 different creatives per ad group; the algorithm decides delivery there, so it's good for finding candidates, not isolating a variable. More in Meta video ad creatives and TikTok video ads.
Which metrics to read at each stage
- Hook rate. Not an official metric. It's the share of impressions that turn into an initial view, and it tells you whether the opening stops the scroll.
- ThruPlay or retention. On Meta, ThruPlay counts plays that reach the end or a late point. It tells you whether the body holds attention.
- CTR. Tells you whether the message earns the click.
- CPA or cost per result. The metric that settles it. A hook can win on hook rate and still lose on CPA.
Don't chase benchmarks you found online. They swing by industry, country and offer. Compare versions against each other and against your own account history.
When to call a winner and how to iterate
Call a winner when the test has run its planned length and the tool flags a result you can trust. TikTok only picks a winner when results are statistically significant, at 90% confidence. If versions come out even, that variable doesn't move your results much, so test another one.
- Scale it separately. Add it to your main campaign as a new ad instead of editing the one that's running.
- Test the next variable. If a hook won, lock it and test angles or formats.
- Log what you learned. What won, against what, on which metric.
Every winner wears out. Have the next round ready before ad creative fatigue hits.
How a video package keeps your tests running
The bottleneck in testing is having versions ready. The 3 video trial for $97 covers a first round, like three hooks on the same body. Starter gets you 10 videos a month for $297, enough to test a new variable every week.
Fill out a short online brief (product, offer, audience, link and photos or clips) and tell us what you want to test. We animate your own material, no filming, and you get 9:16 vertical MP4s in English or Spanish through a download link, no calls. If we miss the 24 business hour window, you get an extra video free; if a video doesn't match your brief, we redo it. We don't promise sales. Start with the 3 video trial.
Ad ideas to start with
Still building this report by hand?
Illustrative question hook for a fictional reporting app. Everything after the opening stays the same in all four versions.
Monday's report, done before your coffee
Illustrative benefit hook that shows the finished report from the very first frame.
Spreadsheet vs. one click
Illustrative comparison hook with a split screen of the old process next to the app.
3 reports you can stop making
Illustrative list hook. The four versions show which opening stops more people and which one ends in more signups.
Illustrative examples of concepts we produce. You choose the result you want and we write the script.
How it works
Pay online
Pick the trial or a plan. No calls, no meetings.
Tell us the result
What you sell, to whom and what you want to achieve. Colors, logo and a reference if you have one.
Get them in 24 business hours
Script, animation and music, ready to upload to Meta or TikTok.
Pricing in USD
Delivered in 24 business hours or you get a free extra video. If a video doesn't match your brief, we redo it at no cost.
Video ads by industry
Video ads for ecommerce
Motion graphics video ads for online stores: script, animation and music, ready for Meta and TikTok Ads in 24 business hours. Try 3 videos for $97.
Video ads for SaaS and apps
Motion graphics video ads for software and apps: your UI animated, script and music, ready for Meta and TikTok Ads in 24 business hours. Try 3 ads for $97.
White label video production for agencies
White label video production for agencies: motion graphics ads under your brand, up to 100 a month, delivered in 24 business hours. Test it with 3 for $97.
FAQ
How many creatives do I need to start testing?
A formal A/B test only needs two versions that differ in one variable. If you'd rather let the algorithm split delivery, TikTok suggests 3 to 5 different creatives per ad group.
How long should a creative test run?
Meta and TikTok both recommend at least 7 days, and a Meta A/B test can run up to 30. Don't shut it off early because one day looks good.
What should I test first: hook, angle, format or offer?
Usually the hook. It's the cheapest thing to change and it decides whether people watch the rest. Then test angles and formats on the winning hook.
Can I edit an ad while it's being tested?
Better not. On Meta, a significant edit can restart the learning phase, and TikTok asks you not to change the ad group during a Split Test. Launch the change as a new version.
Can you make the same video with several hooks to test?
Yes. Say so in your brief and we'll make versions with the same body and different openings. Each version counts as one video in your package.
How much do the videos for testing cost?
The 3 video trial is a one-time $97 per business. Monthly plans: Starter is 10 videos for $297, Growth is 20 for $497, Scale is 40 for $797 with white label, and Agency is 100 for $1,797. Trial add-ons: revisions $29 per video, 1:1 and 4:5 formats $19, and 3 statics $47.
Try it with 3 ads
Pay $97 once, tell us what you want to achieve and get 3 ads ready to run in 24 business hours.
