Repeat Purchase Rate Analyzer

Customer and order counts

All 45 orders fall inside the range and every order has a customer ID (checked by script).

Cohort table

Repeat purchase rate here is my working definition: customers with 2 or more orders in the range divided by customers with at least 1 order. Shopify's Returning customers report uses a similar idea, customers with "two or more orders", and Shopify's cohort analysis groups customers by "when they made their first purchase", which is how I grouped them. Small group rule: a cohort with fewer than 10 customers, or a figure resting on fewer than 5 repeat customers, is labeled in the table.

Rule: rate = repeat customers / customers. Recent cohorts have had less time to repeat: 2026-09 had 29 days observed against 121 for 2026-06, so its rate is not comparable yet.

Days to second order

Tests

Hypotheses to test, not predictions: 1. All cohorts, median 46 days to second order, range 9 to 65 (9 repeat customers): test a reminder email a little before the median, for example day 39, against no reminder, and compare the share with a second order by day 65. The day is my suggestion. 2. Cohort 2026-08, rate 20.0% (1 of 5 customers), the lowest of the cohorts with at least 60 days observed: check what those first orders had in common before testing a change. Small group (n = 5 customers); small group (n = 1 repeat customer), so treat it as a lead, not a finding. 3. Cohort 2026-06, rate 55.6% (5 of 9 customers): ask what was different about that month (offers, products) before copying it. Small group (n = 9 customers), so treat it as a lead, not a finding.

Questions for you

1. Do you want week cohorts instead of month? With 36 customers, month cohorts already hold 5 to 13 customers each.
2. Do you have an earlier period of your own to compare these rates with? My sources hold no benchmark for other stores.