The most important starting point is to decide whether it’s too soon to compare the R% for a cohort with earlier cohorts.

The R% for a cohort measures how many people in a cohort have bought at least twice. In the graphic below, for Nov ‘21, we can see 407 people bought from the rband for the first time. The R% is 34.64%, meaning 141 out of those 407 have bought at least twice since their first purchase in Nov ‘21. So this cohort has had 6 months to buy again (assuming today is May ‘22).

However when we look at a newer cohort, say March ‘22, we see the R% drop to 14.09%, meaning only 73 out of the 518 customers who first bought from the brand in April have bought again.

14.09% for March vs 34.64% for Nov is a big difference, right?

The main issue when comparing Nov ‘21 cohort with the March ‘22 cohort is that the Nov ‘21 cohort has had 6 months to make a repeat order whereas the March’ 22 cohort only has had 2 months. So we expect the March ‘22 cohort to have a lower R% simply because they have not had enough time to make a repeat purchase!

We would recommend looking at the Purchase Intervals chart to understand when most people buy again. In the chart below, we can see that 75% of people who buy again do so in the next 120 days (4 months) after their first purchase. So we would recommend only comparing the March R% to November’s R% at the end of July ‘22 (i.e. 4 months after March ‘22).

If after 4 months March ‘22’s R% is still substantially lower that earlier cohorts, things to check would be whether you increased the price after March, you decreased any re-marketing ad spend, you had any out of stock issues, or whether a competitor may be targeting your listing with ads.