How to Measure Repeat Order Rate
Repeat-order rate is only meaningful when the customer identity, eligible population and observation period are stated. A percentage without those definitions can make a healthy cohort look weak or a heavily discounte…
Repeat-order rate is only meaningful when the customer identity, eligible population and observation period are stated. A percentage without those definitions can make a healthy cohort look weak or a heavily discounted campaign look successful.
Choose the question
Common questions include:
- What share of first-time direct customers place a second completed order within 60 days?
- How many active customers order again in the following calendar month?
- What proportion of promotion users return at full price?
- How does repeat behaviour differ by channel, location or fulfilment method?
Do not combine these into one “repeat rate”.
Use a defined formula
For a first-to-second-order cohort:
Repeat-order rate = customers from the cohort with at least one later completed order ÷ eligible customers in the original cohort.
Exclude test, cancelled and fraudulent orders consistently. State whether refunded orders count. Give recent cohorts enough time to complete the observation window.
Handle identity uncertainty
Guest checkout, shared households, marketplace accounts and changed contact details can create duplicate or merged customers. Report the matching rule and test a sample manually. Where uncertainty is material, present a range.
Segment without hiding the denominator
| View | Useful interpretation | Caution |
|---|---|---|
| Acquisition source | Quality of new demand | Attribution may overlap |
| Offer vs no offer | Whether incentive users return | Groups may differ before the offer |
| Delivery vs collection | Impact of fulfilment experience | Customer occasion differs |
| Location | Operational consistency | Menu and opening pattern may vary |
| First-order month | Cohort trend | Seasonality and elapsed time matter |
Add contribution and experience
A repeat order is not automatically valuable. Report contribution, discount use, refund rate, late delivery and customer support alongside the rate. A programme can increase frequency while reducing margin or service quality.
Create a repeatable report
- Formula and eligible status definitions
- Identity matching rule
- Cohort and observation dates
- Channel and location filters
- Late-arriving data and refund adjustments
- Previous version comparison
Practical next step: calculate first-to-second-order rate for one completed monthly cohort using a 60-day window. Manually verify 20 matched customers before publishing the result.
Check edge cases before publishing
Decide how to handle a customer who places two orders on the same day, a household using one account, a refunded second order and an order that moves between locations. Document whether the second order must be on the same channel. These choices can materially change the result.
Interpret the rate with the business context
A declining repeat rate may reflect rapid acquisition of many new customers rather than worse loyalty. A rising rate may reflect reduced acquisition or a smaller remaining customer base. Report cohort size, acquisition volume and contribution beside the percentage.
Use confidence intervals or at least raw counts for small cohorts. A change from two repeat customers to three can look dramatic as a percentage but may not justify a business decision.
Keep the raw numerator and denominator beside the percentage in every report. This makes later audits possible and prevents a small change in customer-matching rules from appearing as a sudden retention improvement.
Related guides
- Restaurant Marketing and Customer Retention — the main guide for the wider topic.
- Measuring Marketing Performance — the broader guide that frames this implementation.
- Customer Lifetime Value for Takeaways — a closely related operational control to review alongside this page.
- Marketing Attribution for Restaurants — a closely related operational control to review alongside this page.
Sources and date checked
Guidance checked: 24 July 2026. Recheck official guidance and supplier rules before launch.
Operational, legal and platform requirements can change. Recheck official guidance and supplier documentation before altering a live service.
Related practical guides
Restaurant Marketing and Customer RetentionHow to Prevent Discount AbuseDiscount abuse should be managed as a cost and control problem, not as a reason to treat every enthusiastic customer as dishonest. The objective …Read guide →
Restaurant Marketing and Customer RetentionHow to Build a Customer DatabaseA useful customer database is not a large spreadsheet of every name, address and phone number the takeaway has ever received. It is a controlled …Read guide →
Restaurant Marketing and Customer RetentionEmail Deliverability and ComplianceDeliverability and compliance support each other: wanted, well-governed mail is more likely to reach customers and less likely to produce complai…Read guide →Restaurant Marketing and Customer RetentionEmail Campaign Types for RestaurantsCampaign type should follow a customer need and an operationally valid reason to contact them. A weekly discount email is not a default requireme…Read guide →Restaurant Marketing and Customer RetentionDealing with Fake or Unfair ReviewsA review can be inaccurate, abusive or unrelated to a genuine order without being easy to remove. The correct response is to preserve evidence, u…Read guide →Restaurant Marketing and Customer RetentionCustomer Segmentation for TakeawaysCustomer segmentation is useful when it changes a real decision: which message to send, which service problem to fix or which offer to test. It b…Read guide →Connect customer ordering with operations and profit
Use the wider guide library to check the menu, kitchen, fulfilment, payment and financial implications of each decision.
Browse all practical guides