What share of my customers did I keep?
Last updated
Pick a period, a month, quarter or year, and one definition of an active customer. Use both for every count.
Active customers or live subscribers on the first day.
The end count includes the newcomers. The calculator takes them back out.
The same count on the last day, same definition.
First-time buyers or new subscribers who joined after the start date.
Enter how many customers you had at the start of a period, how many at the end, and how many of the end figure were new during it. There is no retention rate until all three are in.
- Customers at the start of a periodNeeded
- Customers at the end of the same periodNeeded
- New customers who joined during itNeeded
This is a head count, not a money figure. To put a value on the customers you keep, the customer lifetime value calculator works it out from order value, repeat purchases and margin, and the CAC calculator prices the new customers who replace the ones you lost.
No benchmark is used: the rate comes only from the counts you enter. Nothing you type is sent anywhere, and there is no account, no install and no store connection.
Start count, end count and the newcomers in between
Customer retention rate is the share of the customers you had at the start of a period who are still customers at the end of it, leaving out anyone who arrived in between, so a coffee subscription that began a quarter with 1,200 subscribers and ended with 1,310, of whom 260 were new, kept 1,050 of its original 1,200, a retention rate of 87.5 percent and churn of 12.5 percent.
No benchmark is used: the rate comes only from your own counts. It measures people, not money — the customer lifetime value calculator is where a kept customer gets a value.
Three head counts from the same period. The hard part is not the arithmetic but agreeing what counts as a customer, and holding that definition steady from start to end.
- Customers at the start
- Everyone you counted as an active customer on the first day of the period. For a subscription that is live subscribers; for a shop it is usually anyone who ordered within a window you choose, such as the last twelve months.
- Customers at the end
- The same count on the last day, using exactly the same definition. It includes the new customers who arrived during the period, which is why they have to be taken back out before retention means anything.
- New customers in the period
- First-time buyers or new subscribers who joined after the start date. They are growth, not retention, and leaving them in would let a strong acquisition month hide a leaking customer base.
- Retention rate
- The share of your starting customers still counted at the end. It only ever looks back at the people you already had, so it answers whether the base is holding, not whether it is growing.
- Churn rate
- The share of starting customers you lost over the period, the other side of retention. The two always add up to the whole of your starting count, so a 90 percent retention rate is a 10 percent churn rate.
Nudgesmart is available on the Shopify App Store. Browse the template library or see what it costs.
Retention and churn for six stores
Invented head counts for six shops. The table is built by the code behind the calculator above, the last row included, where the three counts contradict each other and the page refuses to print a rate.
| Scenario | Start | End | New | Kept | Retention | Churn |
|---|---|---|---|---|---|---|
| Coffee subscription, one quarter | 1,200 | 1,310 | 260 | 1,050 | 87.5% | 12.5% |
| Apparel store, busy season | 900 | 1,150 | 550 | 600 | 66.7% | 33.3% |
| Pet supplies, repeat buyers, one year | 2,400 | 2,380 | 610 | 1,770 | 73.8% | 26.2% |
| Skincare brand, one month | 540 | 548 | 36 | 512 | 94.8% | 5.2% |
| Furniture shop, one year | 300 | 410 | 390 | 20 | 6.7% | 93.3% |
| Counts that do not fit together | 500 | 480 | 520 | These three do not fit together | ||
When a retention rate misleads
The definition of active decides the answer
A shop that calls anyone who ordered in the last year a customer will report far higher retention over a month than one that uses the last 90 days. Pick one definition, write it down and use it for every period you compare.
Period length changes the rate
Monthly and yearly retention are not comparable, because a year gives every customer more chances to drift away. Only compare a rate with another rate measured over a period of the same length.
One-time products churn by design
A mattress or a wedding dress is not bought again next quarter, so low retention there is the product, not a failure. For those shops, referrals and reviews matter more than a repeat order.
Reactivated customers can hide in either count
A lapsed buyer who comes back mid-period may be counted as new by one report and as retained by another. Whichever way your store counts them, keep it the same from period to period.
A kept customer is not a valuable one
This is a head count. Two stores with the same retention rate can earn very different amounts from the customers they kept. What those customers are worth is a separate calculation.
Retention and churn questions
How do you calculate customer retention rate?
Enter the customers you had at the start of a period, the customers at the end, and how many were new during it, and the calculator returns the share of the original customers you kept. 1,200 at the start, 1,310 at the end and 260 new comes out at 87.5 percent retention.
What is the difference between retention rate and churn rate?
They describe the same customers from opposite sides. Retention is the share you kept and churn is the share you lost, so together they always make up all of your starting customers. The calculator shows both.
Why does the calculator reject my numbers?
Because the three cannot all be true in one period. More new customers than the end count, or more kept customers than you started with, usually means the counts come from different date ranges or that returning buyers were counted as new. Fix the source and the rate appears.
Can my customer count grow while retention is poor?
Yes, and it is common in a busy season. In the second worked example the store ends with 250 more customers than it began with, yet it lost 300 of its original 900. A headline customer count mixes the two stories together; retention separates them.
Does this work out customer lifetime value?
No. Retention here is a share of people, with no money in it. To put a figure on what a retained customer is worth, use the separate customer lifetime value calculator, which works from order value, repeat purchases and margin.
Do I need to connect my store?
No. Copy the three counts out of your own reports. They stay in this page, and there is no login.
Keeping a customer starts before they have gone.
Returning visitor is one of the 147 targeting conditions in Nudgesmart, so a regular and a first-time visitor can be shown different popups.
