1. Check the feed before checking the customer
If several customers all become silent at the same moment, investigate the integration first. A failed webhook, expired connection, or changed event name can look like a sudden retention problem.
Distinguish a failed payment, which is a direct billing observation, from missing activity, which may be a collection problem. Do not attach a churn explanation to data you cannot verify.
2. Match the event to the right business
For SaaS sold to barbershops, the shop is the customer account and each staff member is a user inside it. Send a stable account_id for the shop and a separate user_id for the staff member. An accepted request alone does not prove that either was matched correctly.
3. Ask what normal usage looks like
A week without activity means something different for a daily sales tool and a monthly reporting tool. Start with the expected frequency, then compare enough history to understand the customer's own routine. Account for weekends, closures, and seasonal use.
A last-activity date can support a recency check. It cannot, by itself, establish a usage trend, first value, or a probability of cancellation.
4. Know who belongs in the staff count
Observed users are not the same as assigned seats. A roster can include staff who never became active and exclude people who left the business. Never describe a person as a drop-off simply because they were invited and did not start.
Fictional example: a shop has 15 assigned staff, 12 have used the product, and 6 were active this week. Four previously active staff recently stopped. That is worth reviewing, but the owner still needs to confirm why participation changed.
5. Send an observation, not a diagnosis
Once the data is reliable, keep the check-in specific and neutral: 'I noticed fewer staff have recorded sales recently. Did staffing or your process change, or is something getting in the way?'
How to review the result next Monday
Record whether the founder found the warning useful, what action they took, and what happened afterward. Keep a delivery failure separate from a customer not replying. If activity improves after an email, report the sequence honestly rather than claiming the email caused the improvement.
Frequently asked questions
Does no activity mean a customer will churn?
No. Silence can reflect normal usage, a closure, a staffing change, or missing tracking data. Compare verified meaningful activity with the expected routine before treating it as risk.
Can a customer list prove first value?
Not from signup and last-activity dates alone. First value needs a defined meaningful outcome and evidence that the customer completed it.