Reading Shopify’s customer retention report on a small store

Shopify’s Customer cohort analysis report is the nearest thing to a retention table in the admin. This guide, which calls it the retention report, explains what it groups and measures, how to read the table, and what a store with few customers or little history can and can’t take from it.

What the report groups

Shopify’s Help Center says customers are grouped “based on the date that they placed their first order” (Customers reports). Each row is a group of customers who made their first purchase in the same period. The first column names the group by the month of that first purchase. The other columns show your chosen metric by week, month or quarter since the first order. The period 0 column holds orders made in the same period as the first order.

Where to find it and what it measures

From your Shopify admin, go to Analytics, then Reports. The Help page also gives a direct link to the report in your admin. It doesn’t say which Shopify plans include the report or how many customers a store needs, so check that your admin shows it.

The Metric menu offers number of customers, customer retention rate, gross sales, net sales and average order value. Choosing Amount spent per customer enables projections. The page doesn’t define how customer retention rate is calculated, so write down what you read it to mean before you compare it with your own sums. Clicking a cell opens the details for that group and period, including total sales, average order value, orders per customer, amount spent per customer, new and returning customers, top channels and the predicted spend tier overview.

Reading a retention table

Read across a row to follow one group of customers. Read down a column to compare groups at the same age. A group that returns more than its neighbours is worth a question: what was different about that month? Offers, products and sales channels are the places to look.

Newer rows are shorter, because less time has passed. Compare a recent group’s first months with the same months of an older group, not with its full history.

Limits for small stores

Small groups give noisy figures. In a group of 20 customers, one more repeat buyer moves the retention rate by 5 percentage points. Worked example (made up) A swing of a few points in a small group can be one or two customers.

The forecasts have their own minimums. For projections, Shopify says that if 24 months of data aren’t available, the Show projections toggle won’t display, and that “Projections aren’t a guarantee of future sales.” Predicted spend tier needs a store with more than a hundred sales (Predicted spend tier). Klaviyo sets its own minimums for lifetime value predictions, at least 500 customers who have ordered and 180 days of history (Predictive analytics). Checked October 24, 2026.

On a small store, treat the table as a trend over many months, and treat any single cell as a prompt for a question.

What duplicates do to these groups

Shopify counts profiles. A shopper who first bought under one email and returns under another gets a second profile, and that profile’s first order puts it in a later group as a new customer. Our reasoning, not a Shopify statement: this makes retention look lower than it is, because the returning shopper appears as a new customer in the later group and as a one-time buyer in the earlier one.

The size of the effect depends on how many people hold more than one profile, and no public figure measures that for Shopify stores. Count your own. Customer lifetime value for a small Shopify store shows how to count profiles and people. The guest checkout guide explains how one shopper becomes several profiles.

A monthly routine

Once a month, open the report with the same metric and date range. Write down the size of the newest group, its first-period figure and anything that changed that month. After a few months you have your own record, which tells you more than any single reading. If your numbers jump after a customer-list cleanup, note it, because merging profiles changes how customers are counted. Merges are permanent, so record what you merged and when.