Testing is how good advertisers get better results month after month. The key is testing one thing at a time and giving each test enough data.
What to test, in order of impact
- Offer: free consultation vs discount vs free trial.
- Creative concept: testimonial vs demo vs founder video.
- Hook: the first line or first 3 seconds.
- Format: video vs image vs carousel.
- Audience: broad vs interest vs lookalike.
Ways to test
- Meta’s A/B test tool: splits audiences so tests do not overlap.
- Several ads in one ad set: quick, but Meta will push spend to early favourites.
- Separate ad sets with equal budgets: fair for audience tests.
Give tests enough data
Decisions made on a handful of results are often wrong. Check whether a difference is meaningful with our A/B test calculator, and plan sample sizes with our sample size calculator.
Judge on the right metric
Cost per lead or purchase matters more than CTR. A cheap click that never converts is not a winner.
Keep a test log
Write down each test, result and lesson. Over time, it becomes your playbook.
Read how many ads to run.
Want the full picture? This article is part of Meta ads: the complete guide, our in-depth guide with everything in one place.
