The signal
Login activity looked like a straightforward way to assess engagement. In practice, the interpretation was not straightforward. An account-team member accessing a customer tenant could make it look active. A customer using automated emailed reports could receive value without logging in frequently.
The dig
Those were two different ways for the same proxy to mislead us. Before treating activity as evidence of health, we needed to understand who generated it and how the customer actually consumed the product’s output. A counter was an observation, not a conclusion.
The response
I co-built a renewal-risk approach with two GTM partners. My contribution included the practitioner context behind useful and misleading signals, as well as the review cadence, coaching, and account-specific follow-through with my team.
The work was not simply to color an account red or green. It was to ask what the available evidence meant, what remained uncertain, and what someone needed to do next.
What changed
The team had a structured way to examine risk in context and connect the discussion to action. The operating approach brought the conversation back to the evidence: what the signal actually meant, what else we needed to know, and who would follow through.
What this case teaches
Before you trust a signal, understand how it is produced. Corroborate it with the customer’s workflow and the outcome they care about.