How to Calculate Repeat Purchase Rate (With Examples)

To calculate repeat purchase rate, divide the number of customers who bought at least twice in a period by the total number of customers who bought in that period: repeat purchase rate = customers with 2+ purchases ÷ all customers × 100. Count unique customers, not orders. For example, if 1,800 different customers bought from you in the last 12 months and 486 of them bought at least twice, your repeat purchase rate is 486 ÷ 1,800 = 27%.

The metric is also called repeat customer rate or repurchase rate. It is the most direct retention measure for businesses without subscriptions, such as shops, online stores, restaurants and many service firms, because it does not need a definition of "cancelled". This guide covers the formula, how to choose the time window, a cohort version that is better for tracking change, a worked example and the metrics that sit next to it. To see what a higher rate is worth in money, use the customer lifetime value calculator.

The repeat purchase rate formula

Repeat purchase rate = (customers with two or more purchases in the period ÷ customers with at least one purchase in the period) × 100

You need one list: every order in the period with a customer ID and an order date. Then:

1. Count how many orders each customer placed in the period.
2. Count the customers with one or more orders. That is your denominator.
3. Count the customers with two or more orders. That is your numerator.
4. Divide and multiply by 100.

In a spreadsheet with one row per order, a pivot table with customer ID as rows and "count of order ID" as values gives you orders per customer. Copy the result without the total row to a new sheet (IDs in column A, order counts in column B): =COUNT(B:B) then gives the number of customers and =COUNTIF(B:B,">=2") the number of repeat customers.

Two details decide whether the number means anything:

• One person, one ID. If the same customer appears under two email addresses, or as a guest checkout and an account, they count as two one-time buyers. Clean up duplicates first. Our guide on what data to store in a CRM covers the rules that prevent duplicates.
• Only real purchases. Exclude cancelled and fully refunded orders, test orders and internal accounts.

Choosing the time window

The window should be long enough that a normal customer has a fair chance to buy twice. If your customers typically reorder every four months, a three-month window will show almost nobody as a repeat buyer, even if they are perfectly loyal.

A practical rule: make the window at least two to three times the typical time between two purchases. Some starting points to adapt:

Buying pattern	Example businesses	Window to start with
Weekly or more often	Café, bakery, lunch spot	1–3 months
Monthly	Hair salon, pet food, consumables	6 months
A few times a year	Fashion, gifts, garden supplies	12 months
Once a year or less	Furniture, bikes, electronics	24 months

Two more choices to make once and then keep:

• Rolling or calendar window. A rolling 12-month window (always the last 12 months) gives you a fresh figure every month. A calendar year is easier to explain but only updates once a year.
• Same window every time. Changing from 12 to 6 months will move the rate far more than any change in customer behaviour. Write the definition next to every report, just as with churn rate.

Period rate vs cohort rate

The formula above is a period rate: it looks at everyone who bought in a window. It has one built-in bias. Customers who placed their first order near the end of the window had little time to buy again, so a month with many new customers pushes the rate down, even if nothing changed in how well you keep people.

That is why it helps to track a second version:

Second-purchase rate (cohort) = customers from a first-purchase month who bought again within X days ÷ all customers whose first purchase was in that month

For example: of all customers whose first order was in March, what share placed a second order within 90 days? Every cohort gets the same 90 days, so you can compare March with April, or the months before and after you changed something. This is the same idea as a cohort retention analysis, reduced to one question: did they come back a second time?

Use the period rate for a yearly overview and the cohort rate to judge whether a change works.

Worked example: an online shop

A small online shop exports all orders from the last 12 months (rolling). The figures are hypothetical.

Orders per customer	Customers	Orders
1	1,680	1,680
2	480	960
3 or more	240	960
Total	2,400	3,600

• Repeat customers = 480 + 240 = 720
• Repeat purchase rate = 720 ÷ 2,400 = 30%
• Purchase frequency = 3,600 orders ÷ 2,400 customers = 1.5 orders per customer per year
• Share of orders from repeat customers = (960 + 960) ÷ 3,600 = 53%

The three numbers tell different stories. Only 30% of customers came back, yet they placed more than half of all orders. Losing a few of those repeat buyers would hurt far more than the headline rate suggests.

Now the cohort view. In March, 200 customers ordered for the first time; 46 of them ordered again within 90 days. Second-purchase rate: 23%. In May the shop started sending a short how-to email two weeks after delivery. Of 210 first-time buyers in May, 59 reordered within 90 days: 28%. With cohorts this small, a five-point difference can be chance, so the shop keeps the email running and checks whether the following months stay at the higher level before calling it a success.

Related metrics and how they fit together

Metric	Formula	Question it answers
Repeat purchase rate	Customers with 2+ orders ÷ all customers	How many customers come back at all?
Second-purchase rate	Cohort customers with a 2nd order within X days ÷ cohort size	Is the first experience convincing more people to return?
Purchase frequency	Orders ÷ unique customers	How often does a typical customer buy?
Time between purchases	Median days from one order to the next	When is the natural moment to follow up?
Share of revenue from returning customers	Revenue from repeat customers ÷ total revenue	How dependent are you on existing customers?
Churn rate	Customers lost ÷ customers at start	How fast are active customers lapsing?

For time between purchases, use the median rather than the average: a few customers who return after three years would otherwise stretch the average far beyond what is typical.

Churn and repeat purchase rate complement each other. Repeat purchase rate shows how many customers you turn into returning ones; churn, calculated with a lapse window, shows how many of your active customers you lose. The churn rate calculator handles the second part.

What a higher rate is worth

Repeat purchases feed straight into customer lifetime value through purchase frequency. Continuing the example:

• Average order value: $60
• Gross margin: 40%, so $24 gross profit per order

Suppose the shop turns 5% of its customers (120 people) from one-time into two-time buyers. That adds 120 orders a year, raising purchase frequency from 1.50 to 1.55, and adds 120 × $24 = $2,880 gross profit per year, with no extra acquisition cost. If those customers also stay longer, the effect grows over the following years.

Enter your own average order value, purchase frequency and margin in the customer lifetime value calculator, then change the frequency to see what a better repeat rate would be worth. The guide on how to calculate customer lifetime value explains the formulas behind it.

How to raise your repeat purchase rate

Most of the gain sits between the first and the second purchase. A customer who has bought twice has already formed a habit with you; a one-time buyer has not.

1. Make the first order go right. Late deliveries, unclear instructions or a slow reply to a question are the most common reasons people do not return. Our customer onboarding checklist applies to shops as well as to service firms.
2. Follow up at the natural moment. Calculate the median time between first and second purchase. Send a useful reminder or tip shortly before that point, not on a fixed weekly schedule.
3. Give one simple reason to return. A stamp card, a small credit for the next order or early access to new products. Pick one from our loyalty program ideas and calculate its cost from your margin.
4. Ask one-time buyers why. A single question to customers who did not reorder within your window often reveals a fixable problem: price, quality, delivery, or simply that they forgot about you.
5. Segment before you act. Break the rate down by first product, sales channel and region. A product with a very low repeat rate may be attracting the wrong customers, or disappointing the right ones.

Common mistakes

• Counting orders instead of customers. "Repeat orders ÷ all orders" is a different metric (share of orders from returning customers). Both are useful, but label them clearly.
• A window shorter than the buying cycle. The rate then mostly measures your window, not your customers.
• Duplicate customer records. Guest checkouts and changed email addresses turn one loyal customer into several one-time buyers and push the rate down.
• Comparing periods with different mixes of new customers. A month with a big acquisition campaign lowers the period rate. Use the cohort rate to compare.
• Reacting to small changes. With a few hundred customers, a move of two or three points is often noise. Look for trends over several months.

Your next step

Export last year's orders with customer ID and date, remove refunds and test orders, and merge obvious duplicates. Count customers with one order and with two or more, and calculate your repeat purchase rate. Then take the first-time buyers of one recent month and check how many bought again within 90 days: that cohort figure is the one to watch while you work on the second purchase. Finally, put your purchase frequency into the customer lifetime value calculator to see what each extra returning customer is worth.