Our SaaS has more active users. Why are fewer paying teams coming back?
By Trayan Hristov · Founder, Thrivelia · 19 November 2026
One enthusiastic customer can invite 50 teammates and make the active-user chart look terrific. Meanwhile ten small customers may have quietly stopped using the product. For a founder selling by workspace, I’d put the paying organisation on the first line of the retention report.
Give activation a product-specific definition
Choose a meaningful outcome that can recur. For a collaborative tool it might be one teammate creating work and a second teammate using or approving it. For a single-user workflow it might be running the core task again on a later day. A login, background sync or automated notification is rarely enough. Exclude internal accounts and check how impersonation or support activity is recorded.
Store workspace ID, event time, actor, role and account status. Keep user-level adoption underneath. A user belonging to two organisations can act in each, but the event needs the correct workspace context. Keep invited but unactivated workspaces in the acquisition-cohort denominator when analysing onboarding.
Read retention at a fixed interval
Suppose 100 paid workspaces activate in April. In week four, 60 complete the defined core event. Week-four workspace retention is 60%. If 50 of those 60 use the product again in week five, that does not tell you total week-five retention. Some of the other 40 may return. Re-evaluate all 100 against the same event each week.
Choose whether the chart means “returned during week four” or “returned in week four or any later week”. Both are legitimate and can yield very different answers. Show only cohorts old enough to observe the chosen interval and label activity retention separately from renewal. Amplitude’s retention calculation guide.
Route the intervention by role
If owners activate but teammates never accept invitations, test invitation timing and the in-product handoff. If teams collaborate once and never repeat the core task, investigate whether the use case recurs or whether setup was too costly. If activity is healthy but billing fails, product nudges won’t repair payment collection.
Run onboarding tests at workspace level so teammates see the same experience. Compare meaningful actions, paid retention and customer-support load across assigned workspaces. Viktor’s role-aware onboarding case is an example of separating owner and teammate journeys.
Ask ChatGPT or Claude about your own numbers
Calculate fixed-week workspace retention from these activated accounts and core-event counts. Separate paid renewal from activity and invited users from active teams. What changes if retention means returned on or after that week?

.png)



Comments