How many purchases does a Meta ad need before you can trust its ROAS?

There is no universal number. You can find your own, though.

I took one Meta account, $1.2 million in spend over 22 weeks, and paired every ad with itself the week after: 722 pairs of ad weeks. Then I grouped the pairs by how many purchases the ad had in the first week and checked how well that week’s ROAS predicted the next.

One Meta account, $1.2M spend, 22 weeks, 722 pairs of ad weeks. Correlation measures how much a week carries into the next. It promises nothing about any single ad.

The group with 50 or more purchases came out on top. Its correlation between this week’s ROAS and next week’s was 0.44, the strongest of all the groups, and it had the smallest average swing from one week to the next.

Below 20 purchases the numbers do not climb in a neat line: 0.20 for 1 to 4 purchases, 0.05 for 5 to 9, 0.09 for 10 to 19. That is roughly what noise looks like. A handful of purchases can make any ROAS look like a trend.

0.44 is still moderate

Even with 50 purchases or more, a winning week does not guarantee another winning week. It makes the next week more likely to resemble this one, and that is all. Treat a strong week as one piece of evidence and wait for the second.

Find the threshold in your own account

Fifty is this account’s number. Yours depends on the business model, how much purchase values vary, the attribution setup and the spend level. It might be 30. It might be 100.

The analysis is simple to repeat:

Export ad results by week: spend, purchases and ROAS.
Keep the ads that ran in two consecutive weeks and pair each week with the one after it.
Group the pairs by purchases in the first week.
Inside each group, correlate this week’s ROAS with next week’s.

The point where the correlation jumps and the swings shrink is where ROAS starts to deserve trust in your account.

Do not copy the number. Copy the analysis.