The customer retention rate formula
The standard version is: retention rate = (customers at end of period − new customers acquired during the period) ÷ customers at start of period × 100%.
New customers are subtracted on purpose. If you leave them in, a successful ad campaign can hide the fact that existing buyers are leaving, because the total looks healthy. The metric answers one question only: of the people you already had, how many stayed?
- Start of period: all active customers on day one
- End of period: all active customers on the last day
- New customers: people who bought for the first time inside the period
- Churn rate = 100% − retention rate
A worked example
The numbers below are made up, so replace them with your own. Take an online store selling through Instagram and Uzum. At the start of a quarter it has 500 active customers, wins 150 new ones during the quarter, and ends with 560 active customers.
Retention = (560 − 150) ÷ 500 × 100% = 410 ÷ 500 × 100% = 82%. Of the original 500, 410 stayed and 90 left, so churn is 18%. The customer base grew, yet almost one in five earlier customers is gone. Without this calculation you would only see the growth.
Who counts as an active customer
Settle this before calculating anything. For a subscription it is obvious: a paid, running subscription means an active customer. For shops, cafés and service businesses you define a window yourself, for example at least one purchase in the last N months.
The window should match how often people naturally buy. A coffee shop thinks in weeks, a grocery store in a month, a beauty salon in a couple of months, a dental clinic in six to twelve months, a furniture or appliance seller in years. A window shorter than the real buying cycle makes retention look artificially low; a longer one makes it look too good. Pick the definition once and keep it, otherwise you cannot compare periods.
Cohort analysis: see where customers drop off
Overall retention blends long-standing customers with recent ones. A cohort analysis separates them. You group customers by the month of their first purchase, then track what share of each group is still buying one month later, two months later, and so on.
The table is easy to build: rows are first-purchase months, columns are months since that purchase, and each cell shows the share of that cohort who bought in that month. You can make it in Excel or Google Sheets from a CRM or point-of-sale export. It shows when people tend to leave (often right after the first order), whether newer cohorts hold up better than older ones, and whether a change in service made a difference.
How to put the number to work
A single figure says little; comparison is where the value is. Compare retention across periods, customer segments and acquisition channels. Buyers who came from Instagram, a Telegram channel or OLX may behave very differently.
Tie the metric to specific actions. If you started reminding customers who have not ordered for a while, look at the cohorts those messages reached. If you changed a manager or booking rules, check customers who arrived afterwards. That turns retention from a reporting number into a way to test decisions. It is also an input for customer lifetime value and for deciding how much you can afford to spend on acquisition.
Common mistakes
Most wrong results come from the same handful of errors.
- Not subtracting new customers from the closing count
- Using an activity window that does not fit the buying cycle
- Looking only at the overall figure and skipping cohorts
- Comparing businesses with different purchase rhythms, such as a café and a furniture showroom
- Measuring over a short, seasonal period like the weeks before a holiday
- Failing to identify customers: with cash payments and no loyalty card, retention is almost impossible to measure
Where to start
Define what an active customer means for your business. Export the last twelve months of purchases from your CRM or POS, link them to a phone number or loyalty card, and calculate retention by quarter. Then build a simple cohort table. Even the first pass usually shows the stage where you lose the most people.
If your data is scattered across the till, Click and Payme, a Telegram bot and ad accounts, end-to-end analytics can connect each order to its source and customer. markdev sets up this kind of reporting for businesses in Uzbekistan, so you can talk through your situation and see what data you already have.
FAQ
What is a good customer retention rate?
There is no universal benchmark because it depends on your niche and buying rhythm. Compare yourself with your own earlier periods and between cohorts rather than with other businesses.
What is the difference between retention and churn?
They are two sides of the same number: churn equals 100% minus retention. Choose one for your reports and use it consistently.
How do I measure retention if customers pay in cash?
Without a customer identifier it is nearly impossible. A loyalty card, phone-number booking or orders through a Telegram bot make customers traceable.
How often should I recalculate retention?
Match the period to your buying cycle: monthly suits food delivery, while a salon or clinic is better measured quarterly or half-yearly.
Let’s look at your task for free
Leave your contact — in a 15-minute call we’ll show where leads get lost and what to fix first.