“Our repeat rate is 25%” sounds useful, but it hides the most important question: is it getting better or worse? Cohort analysis answers that, and it’s one of the first things investors ask to see.
What a cohort is
A cohort is a group of customers who made their first purchase in the same month. Tracking each group separately shows how customer behaviour changes over time, instead of blending everyone together.
A worked example
Numbers in this example are illustrative.
This table shows what percentage of each month’s new customers ordered again within 1, 3, and 6 months:
| First purchase in | New customers | Repeat by month 1 | By month 3 | By month 6 |
|---|---|---|---|---|
| January | 1,000 | 12% | 22% | 30% |
| February | 1,200 | 10% | 19% | 26% |
| March | 1,500 | 8% | 15% | Not yet |
How to read it
New customer numbers are growing every month, which looks great on a sales dashboard. But each new cohort is repeating less than the one before. Something about how customers are being acquired has changed.
Common causes include:
- Heavier discounting that attracts one-time bargain hunters.
- New ad audiences or channels that bring in less-suited customers.
- Product or delivery issues hurting the first experience.
- A shift toward marketplaces, where you can’t easily re-market to buyers.
What good looks like
Healthy cohorts either hold steady or improve over time. Newer cohorts matching or beating older ones suggests your product, experience, and targeting are getting better.
How to set up cohort tracking
- Export orders with customer IDs and dates from your store.
- Tag each customer with the month of their first order.
- Count repeat purchases at 30, 90, and 180 days.
- Update monthly and compare new cohorts with old ones.
- Split by channel or first product once the basics work.
Cohort data also feeds directly into your lifetime value calculation. See our guide to CAC, contribution margin and LTV.
FAQs
Can I do cohort analysis in a spreadsheet?
Yes. A monthly export and a pivot table are enough for most early-stage brands.
Should I track revenue cohorts or customer cohorts?
Start with customer repeat rates. Revenue cohorts, showing how much each group spends over time, are a useful next step.
How many months of data do I need?
At least six months gives you a meaningful trend, but you can start learning from three.
Download The D2C Founder’s Cash Flow Checklist to check your margin health alongside your repeat rate.
Get the free checklist →Want to understand your repeat customers properly? The Financial Arc builds cohort analysis into financial models and investor reporting for D2C brands.
Get in touch