Enter customers at the start of a period, at the end, and how many you acquired in between. You get customer retention rate, the matching churn rate, and the expected customer lifetime it implies.
Subtracting new customers is the step that makes the figure honest. Without it, a business acquiring quickly can report retention above 100 percent while losing a large share of the people it already had.
In the worked example the customer base grew from 2,400 to 2,580. Reported as net change, that is 7.5 percent growth and everything looks fine.
But 540 new customers were acquired to get there. Take them out and only 2,040 of the original 2,400 remained. The business lost 360 customers in a quarter, which is fifteen percent of its base, and replaced them with acquisition spend.
This is the pattern that makes retention the metric most worth watching in a business that is advertising heavily. Acquisition can outrun losses for a long time. When spend is cut, the growth stops immediately and the churn does not, and the base starts shrinking in a single month.
If you know your acquisition cost, those 360 lost customers have a price. At an acquisition cost of just $24 each, replacing them would consume more than the quarter’s entire operating profit in the site-wide example.
Quarterly retention of 85 percent is monthly retention of about 94.7 percent, because the three monthly rates compound. The calculator converts it for you.
| Monthly retention | Left after 12 months, from 1,000 | Expected lifetime |
|---|---|---|
| 90% | 282 | 10 months |
| 94.7% | 521 | 19 months |
| 97% | 694 | 33 months |
| 98% | 785 | 50 months |
| 99% | 886 | 100 months |
Moving from 98 to 99 percent monthly retention looks like a one point improvement. It doubles expected lifetime from fifty months to a hundred, and it doubles lifetime value with it. That is why the LTV calculator is so sensitive to this single input.
Customer lifetime value is roughly revenue per period multiplied by expected lifetime, and expected lifetime is one divided by churn. So retention sets the ceiling on what you can afford to pay for a customer.
At 94.7 percent monthly retention, a customer stays about nineteen months. At 97 percent, thirty three. The second business can pay nearly twice as much to acquire the same customer and still earn the same return, which means it can outbid the first in every auction.
This is why a retention improvement is usually worth more than an equivalent acquisition improvement. Cutting CAC by ten percent saves ten percent. Improving retention by a few points raises the value of every customer you have already bought and every one you will buy. Check the effect with the CAC calculator and payback period calculator together.
Growth hides losses. A base can rise while a sixth of it walks out the door.
The first trap is leaving new customers in the numerator. It produces figures that can exceed 100 percent and it is the single most common error in retention reporting.
The second is defining a customer inconsistently. For a subscription the definition is clear. For repeat purchase businesses you must decide how long someone can go without buying before they count as lost, and the answer changes the figure dramatically. Pick a window, write it down, and never change it quietly.
The third is blending cohorts. Customers acquired last month churn far faster than customers who have stayed two years. A single blended rate rises when acquisition slows, simply because the base gets older, and it falls when acquisition accelerates. Neither movement means the product changed.
The last is comparing periods of different length. Quarterly and monthly retention are different numbers for the same business. Convert to a monthly rate before comparing, which the calculator does automatically.
Subtract new customers acquired during the period from the customers you had at the end, then divide by the customers you had at the start. Leaving new customers in produces a figure that can exceed 100 percent.
They are the same measurement from opposite sides. Retention is the share that stayed, churn is the share that left, and together they sum to 100 percent for a given period.
It depends on the business model and the period. Compare monthly rates rather than quarterly or annual ones, and compare against your own history, because the definition of a lost customer varies too much between businesses to benchmark reliably.
Because retention compounds and expected lifetime is one divided by churn. Moving monthly retention from 98 to 99 percent doubles expected lifetime and roughly doubles what each customer is worth.
We use Google Analytics to count visits, only if you allow it. Nothing you type into a calculator is ever sent. Privacy policy