Blog · Board and management reporting · Software
SaaS onboarding KPIs connect checklist completion, activation time, and later usage. Compare signup cohorts with an account-level example and checks that keep unfinished cohorts out of the results.
SaaS onboarding KPIs measure whether new customer accounts finish setup, reach a useful first outcome, and keep using the product afterward. Compare those measures by signup cohort, with a fixed activation deadline and the same later retention window. A checklist completion rate alone cannot show whether onboarding delivered value.
Separate setup from the first business outcome. For a US workflow software company, setup might mean creating the workspace and assigning account roles. Activation might mean completing the first approved customer workflow. Opening a welcome email is neither milestone.
Write the qualifying action, account identifier, deadline, and owner beside each KPI. A cohort needs both an entry condition and a later return condition; this distinction is explicit in the cohort inclusion and return criteria documentation. Here, entry is account signup and return means completing another qualifying workflow during days 31 through 60 after signup.
Keep the activation definition stable across the cohorts being compared. If the workflow changes, retain a definition version and explain the break. Otherwise the report can announce improvement because the qualifying action became easier.
Start with an account register containing account ID, signup date, customer segment, and onboarding owner. Add the checklist completion timestamp, first qualifying workflow timestamp, and later qualifying workflow dates from the files your business already exports. Resolve duplicate account IDs before joining the files.
Produce one reporting row per account. Repeated workflows must not multiply the customer count. Retain the source timestamps behind the summary flags so a reviewer can inspect why an account passed a milestone. The broader customer base KPIs for SaaS sales teams cover the customer population outside this onboarding view.
Use elapsed days from signup, rather than the calendar month in which activity happened. An account signing up late in a month deserves the same observation period as one signing up early.
An illustrative US software company reviews these accounts on May 1, 2026. January accounts signed up January 1; February accounts signed up February 1. Every account has completed its later observation window. Activation day is elapsed days after signup, and the activation deadline is day 14.
| Account | Signup cohort | Checklist complete by day 14 | Activation day | Qualifying activity in days 31–60 |
|---|---|---|---|---|
| A | January | Yes | 2 | Yes |
| B | January | Yes | 4 | Yes |
| C | January | Yes | 8 | No |
| D | January | Yes | — | No |
| E | February | Yes | 1 | Yes |
| F | February | Yes | 2 | Yes |
| G | February | Yes | 3 | Yes |
| H | February | Yes | 6 | No |
The dash means no qualifying activation by the deadline, not immediate activation.
Completion rate = Accounts completing the checklist by day 14 / Eligible signup accounts
Activation rate = Accounts activating by day 14 / Eligible signup accounts
Later usage retention = Signup accounts active in days 31–60 / Eligible signup accounts
| Measure | January calculation | January result | February calculation | February result |
|---|---|---|---|---|
| Checklist completion | 4 / 4 | 100.0% | 4 / 4 | 100.0% |
| Activation | 3 / 4 | 75.0% | 4 / 4 | 100.0% |
| Median activation days | Middle of 2, 4, 8 | 4.0 | (2 + 3) / 2 | 2.5 |
| Later usage retention | 2 / 4 | 50.0% | 3 / 4 | 75.0% |
Percentages and durations are rounded to one decimal place. The median uses the middle value, or the average of the middle pair, following the NIST definition of measures of location.
Completion is unchanged, while activation and later usage are stronger in February. The median is shorter by 4.0 − 2.5 = 1.5 days. These small cohorts describe the accounts shown; they do not establish a reliable company benchmark.
Across both cohorts, the register has eight accounts. Seven activated and five returned, giving activation of 7 / 8 = 87.5% and later retention of 5 / 8 = 62.5%. The activation reconciliation is seven activated plus one not activated equals eight signups.
For time to activation, sort the seven observed durations: 1, 2, 2, 3, 4, 6, 8. The overall median is 3.0 days. Compute it from account durations; combining cohort medians does not recover that result reliably.
If the question is whether activated accounts returned, change the denominator explicitly. January gives 2 / 3 = 66.7%; February gives 3 / 4 = 75.0%. Those are retention rates among activated accounts, rather than retention of all signups. The separate time to value and retention article examines that relationship in more detail.
Customer success owns checklist exceptions and the account follow-up. Product leadership owns the activation definition and investigates friction in the qualifying workflow. The analyst owns deduplication, cohort eligibility, and reconciliation back to the exports.
Start with account D: setup finished, but value was not reached by the deadline. Then examine C and H, which activated but did not return. Ask whether their intended workflow is monthly, whether the account needed help, and whether the supplied export covers the expected activity. Compare accounts with similar use cases before changing the onboarding process.
Bring a sample of your own account register and milestone or usage export to Covirage; it maps the columns, and its tools compute cohort rates, activation durations, and checks against the underlying rows. The external AI model explains the comparisons with every figure traced to its rows. Create a free Covirage account, then set up your business and bring your file.
See Covirage for SaaS sales teams for the supporting industry context.
Completion means the customer finished a defined setup checklist. Activation means the customer achieved an outcome that demonstrates product value. An account can complete administrative steps without doing useful work, so report both measures rather than assuming they describe the same event.
Use accounts when the onboarding process and commercial relationship belong to the customer organization. Use users for individual training or seat activation. Label each measure and keep those denominators separate, because several users may belong to one customer account.
Leave their activation duration blank and exclude them from the median of completed activations. Keep them in the activation-rate denominator once their observation window has closed. Otherwise a faster median can hide a growing group of customers who never reach value.
No. Customers that activate quickly may differ in size, readiness, contract, or support. Compare equivalent cohorts and inspect account rows before drawing conclusions. A relationship between activation time and later usage can identify a question for investigation without proving causation.