The check-in email you're about to send won't help. It signals you noticed but didn't investigate. And in payroll and compliance SaaS — where accounts run on strict schedules and power users are often single points of failure — the wrong response to the wrong cause can accelerate the churn it's supposed to prevent.
Low activity is a symptom. There are five causes. Each one requires a different response. Treating all five the same way is how you lose an account you could have kept.
Usage dropped 40% last month. The instinct is to check in. The check-in goes unanswered. Two more weeks pass. The renewal is in four months.
The account went quiet in mid-cycle. It looks like disengagement. It's actually seasonal — their payroll cycle runs quarterly and they're between runs. The check-in creates urgency that didn't exist and trust erodes for no reason.
Three users went dark on the same day. The power user who championed the implementation left the company. Her replacement has never seen the product. This is not an adoption problem. It's a continuity problem — and it needs the Executive Continuity Risk protocol, not a usage nudge.
None of these situations are the same. None of them get better with the same response. This template builds the diagnosis infrastructure that changes that.
A root cause classification protocol — the process for identifying which of the five causes is driving low activity before any outreach is made. The diagnosis determines the response. The response determines whether the account stays or goes.
| Cause | What it looks like | What it is not | Correct first response |
|---|---|---|---|
| Seasonal / Cyclical | Activity dips on a recurring pattern — quarterly payroll cycle, fiscal year-end, seasonal field ops slowdown. Matches prior-year pattern. | Not churn risk. Not a relationship problem. | Monitor only. No outreach. |
| Configuration Gap | Activity peaked at go-live then gradually declined. Users found workarounds or stopped using specific features. | Not a relationship problem. A product-fit problem. | Working session — not a QBR. |
| Champion Departure | Activity drop correlates with a specific user going inactive. The power user left or changed roles. Drop date matches departure date. | Not an adoption problem. A continuity problem. | Activate Executive Continuity Risk protocol immediately. |
| Business Change | Customer downsized, reorganized, or paused operations. Usage reflects the business reality, not disengagement. | Not a post-sale failure. May be a contract scope issue. | Get on the phone. Understand the change. Assess contract fit. |
| Disengagement | Activity dipped, no obvious external cause, renewal within 6 months. No response to recent touchpoints. | This is the dangerous one. It is also the most recoverable if caught early. | Senior outreach. Not a check-in email. |
🚦 The gate
- No outreach is made before the diagnosis is complete — not a check-in email, not a call, nothing.
- The cause is classified using data, not intuition — baseline, drop date, user-level breakdown.
- The response protocol is selected based on the classified cause — one cause, one protocol.
- Health score is updated with the diagnosis rationale, not just a color change.
🚦 The diagnosis protocol
- Characterize the drop — baseline, depth, timing, which users.
- Check for external causes — seasonal pattern, user departure, business change, system issues.
- Classify the cause — one of the five.
- Select the response — based on the cause, not the symptom.