Customer retention software that flags at-risk accounts and names why they leave

Customer retention rate = ((customers at the end of a period, minus new customers acquired during it) divided by customers at the start) times 100. If you began the quarter with 500 customers, added 80 and finished with 540, the calculation is ((540 - 80) / 500) x 100, which is 92%. The corresponding churn rate is 8%.

Two refinements matter. Measure by revenue as well as by logo, because losing five small accounts and losing your largest one look identical in a customer count. And measure by cohort, tracking each month's signups over their own lifetime, since a blended rate hides whether recent cohorts are getting better or worse. Those two cuts usually reveal the real story faster than any single headline number.

Across most B2B products, the recurring causes are a short list: the customer never reached the value they signed up for, a persistent friction or reliability problem wore them down, the champion who bought it left, price stopped matching perceived value, or a competitor solved the job better. Sudden, single-event churn is rarer than teams assume; most cancellations are the end of a slow decline that was visible in usage and tone months earlier.

That is what makes early signals so valuable. Falling login frequency, a narrowing set of features used, rising ticket volume with sharper language and a dropping satisfaction score usually appear well before the cancellation. UserInsight watches those signals together and ranks the recurring causes by the revenue attached, so a retention team can act on the pattern rather than react to one lost account at a time.

It varies sharply by model. Established B2B SaaS companies typically target annual logo retention in the 85% to 95% range, with net revenue retention above 100% once expansion is counted, while self-serve consumer subscriptions run far lower and are considered healthy at levels that would alarm an enterprise vendor. Contract length, price point and how essential the product is to daily work all shift what good looks like.

Because of that spread, the more useful benchmark is your own trend by cohort and segment. A rate holding steady in aggregate while your newest cohorts retain worse each month is a warning that a headline number will not show for another year. Track the direction, segment it, and pair every movement with the reasons behind it before drawing conclusions.