In one sentence
Cohort analysis is tracking groups of customers who started in the same period, like the month of their first order, to see which groups come back, spend more and stay loyal over time.
How cohort analysis works for online stores
A cohort is a group of customers who started at the same time, usually the month of their first order. Cohort analysis follows each group forward and tracks the same numbers month by month: how many came back, how often they ordered, how much they spent.
Laid out as a grid, each row is a cohort and each column is a month since the first order. Read across a row to see how one group behaves over time. Read down a column to see whether newer groups do better or worse than older ones at the same age.
Cohort analysis vs. customer segmentation: what's the difference
Why cohort analysis matters for ecommerce brands
Store-wide averages hide what's really happening. A spike in new customers can make revenue look healthy while each new group comes back less often than the last.
Cohorts show you that early. They tell you whether a sale brought in loyal customers or one-time bargain hunters, and which acquisition channels bring customers worth paying for. They're also how you check customer lifetime value (LTV) against customer acquisition cost (CAC) with real data instead of a guess.
How to run a cohort analysis on your store
- Pick the starting point: first order month works for most stores.
- Pick one metric to follow, like repeat purchase rate or revenue per customer.
- Build the grid for your last 6 to 12 cohorts.
- Compare rows at the same age, then split by acquisition channel to see which sources bring customers who come back.
Cohort analysis FAQs
How do I do a cohort analysis in Shopify?
Start with the cohort report in Shopify's Analytics, which groups customers by first order date. For splits by channel or product, export your orders and build the grid in a spreadsheet, or use a retention app.
Should I group cohorts by week or by month?
Monthly suits most stores. Weekly cohorts are small and noisy unless you have a lot of orders. If customers buy on a long cycle, like furniture, quarterly cohorts can read more clearly.
What's a good retention rate for an ecommerce cohort?
It depends on what you sell. A coffee subscription and a mattress brand will never look alike. Compare your own cohorts against each other and watch the trend.
Can I track customers who found me through AI as their own cohort?
Yes. Tag the first visit's source with UTM parameters or referrer data, then group first-time customers by it. That shows whether AI referral traffic brings customers who come back, not just first orders.
About the author
Jenn Starr, Co-founder, Selix
Jenn has spent 20+ years helping SaaS startups build their marketing and implementation playbooks. She has bootstrapped three marketing agencies and is a 500 Startups alum. At Selix, she leads the work of getting ecommerce brands named in AI answers.
Related terms