To audit repeat purchase rate, define exactly which customers and orders you count, group buyers by their first purchase, and give each group the same amount of time to return. Then check whether repeat orders produce useful revenue after refunds and costs. A single percentage can hide a lot of different customer behavior.
This walkthrough helps an ecommerce team build a review it can repeat. You do not need a new analytics subscription to start. You need reliable order history, a stable customer identifier, and a written definition everyone on the team can follow.
Start by choosing the question
Two useful questions often get mixed together. One asks what share of the customers buying during a reporting period had purchased before. The other asks what share of a particular group of first-time buyers placed a second order within a defined window.
Shopify defines returning customer rate as returning customers divided by customers who placed orders. That period-based measure helps describe the mix of buyers in a report. It is different from following a first-purchase cohort to see how many people return within 60 or 90 days.
Source: Shopify analytics fields reference
For an audit of the first-to-second purchase, use a cohort question: “Of the customers who first bought in this month, how many placed another qualifying order within 90 days of their own first order?” Choose a window that fits the product. Ninety days is an illustrative review window, not a universal target for every store.
A consumable and a durable piece of furniture give customers different reasons and timelines to buy again. Decide what a useful return looks like for your catalog before copying someone else's benchmark.
Write the counting rules
Create a short definition sheet before exporting or calculating anything:
- Unit: unique customers, identified by a stable store customer ID.
- First purchase: the earliest qualifying order in available history.
- Repeat event: another qualifying order within the chosen window.
- Window: measured from each customer's first order timestamp.
- Exclusions: labeled test orders, canceled orders, and other noncommercial records.
- Refund treatment: explicit rules for fully and partially refunded purchases.
- Observation cutoff: the date through which order data is complete.
There is no benefit in making a rate look better by changing the rules between months. If you revise a definition, calculate both periods again and label the change.
Check the order history before calculating
Start with customer ID, order ID, order date, order status, and refund information. Add first-order product, discounts, and acquisition source only where the data is dependable. Keep personally identifying details out of the working summary when they are unnecessary.
Look for duplicated order rows, missing customer IDs, changed identifiers, and imported history that begins after a customer's true first purchase. If you cannot identify the earlier purchase, label that limitation instead of treating every imported customer as newly acquired.
Reconcile a small sample against actual order records. Include a customer with one order, one with several, one with a refund, and one whose repeat falls just outside the window. These checks can reveal a counting error before it affects every row.
Give cohorts equal time to mature
A cohort acquired last week has had less opportunity to return than one acquired four months ago. Comparing their raw repeat rates creates a misleading result.
For a 90-day review, count only customers whose first order is at least 90 days before the observation cutoff. Within a monthly acquisition cohort, later buyers may not yet qualify even when earlier buyers do. Either wait until the entire month has matured or show the eligible subset and label it clearly.
Shopify's customer reports include a cohort analysis based on first purchase, with periods since acquisition and selectable measures. Use the report's definitions and available settings to understand what it shows. An exported custom 90-day calculation can differ from a calendar-month report, so do not assume the figures should match exactly.
Source: Shopify customer reports documentation
Calculate the first-to-second purchase rate
For your documented cohort definition:
Repeat purchase rate within the window = eligible customers with a qualifying second order within the window / eligible first-time customers in the cohort.
Illustrative example: 400 eligible customers made their first purchase in a cohort. Of those customers, 84 placed a qualifying second order within 90 days. The result is 84 / 400, or 21%.
That is a made-up example to explain the calculation. It is not a D2C Champion customer result or an industry benchmark. Multiple repeat orders from one customer still count as one returning customer in this rate. Order frequency is a separate measure.
If 30 customers have not completed their observation window, leave them out of the eligible denominator and disclose that the cohort is incomplete. Do not count them as nonreturners just because the calendar advanced.
Investigate the reasons behind a change
Compare mature cohorts using the same definition. Then split them by a small number of meaningful factors, such as first product, first-order discount, subscription versus one-time purchase, or acquisition source where attribution is reliable.
Display counts beside percentages. A segment with three repeat buyers out of ten customers deserves a different level of confidence from one with 300 out of 1,000. Treat small samples as leads for investigation.
Check product availability, delivery issues, customer service themes, and the buying experience before assuming the email program caused the change. A customer may want to reorder and find the product unavailable. Another may have bought a gift with no immediate reason to return.
For the broader strategy, read our retention and lifetime value guide. Use this audit to identify the problem before choosing a retention tactic.
Review the economics of repeat orders
Keep repeat rate, revenue, and profitability separate. A discount can encourage another purchase while leaving little contribution after product cost, fulfillment, refunds, and marketing expense.
Report the value of qualifying repeat orders alongside the customer count. Document which costs are included in a contribution calculation and which remain unavailable. Do not call gross revenue profit or present a short observation window as complete lifetime value.
If the apparent improvement depends on a promotion, compare the relevant costs and customer behavior after the promotion. The audit should help your team make a business decision, rather than celebrate a percentage in isolation.
Turn the audit into one next action
Save the definition, cutoff date, source report, eligible customer count, repeat customer count, and important exclusions. Add the finding you want to investigate, its owner, and the next review date.
If customers cannot find an appropriate second product, improve the post-purchase recommendation. If the first product runs out naturally, investigate an appropriate replenishment reminder with the required marketing permissions. If returns or service problems dominate, resolve those issues before pushing another promotion.
Choose one change and define the outcome before making it. Recheck comparable mature cohorts and acknowledge other changes that could influence the result. A before-and-after movement alone does not prove causation.
Connect the result to your ecommerce scorecard, then use D2C Champion to organize the growth questions you want to investigate. The audit itself must come from your actual store records. Start by writing the metric definition today, before changing a campaign or purchasing another tool.