Question

I'm testing two different ad creatives and heard that if you place both in the same ad set, Meta will automatically allocate more budget to the better-performing one.

Is that a bad thing when it comes to testing?

Would it be better to put each ad in its own ad set so I can get a more balanced comparison of results?

Or should I just trust that Meta is smart enough to pick the winner and optimize the budget accordingly?

Answer 1

Yes, Meta will auto-optimize budget within an ad set and favor the better-performing creative quickly, but that can skew results if you're trying to run a fair A/B test.

If both ads are in the same ad set, Meta may start pushing spend to one within hours, based on early signals (even if not statistically significant).

This can lead to a false winner, especially if you’re judging based on conversions and not just CTR.

Best Practice for Fair Testing:

• Use separate ad sets for each creative with equal budgets.
• This ensures both get equal opportunity to perform.
• After a few days (3–5), compare metrics like CPC, CTR, ROAS, or conversion rate.

When to Trust Meta's Optimization:

• If your goal is performance, not learning, Meta’s auto-optimization works well.
• If you’re testing just to find a winner and scale, letting Meta handle it inside one ad set is fine.