What counter metrics track health of weekly active users?

What's really being asked

This tests whether you know that a North Star metric like weekly active users is a lagging indicator that can be inflated by low-quality growth tactics such as paid acquisition spam, notification abuse, or feature bloat. The interviewer wants to see if you instinctively build guardrails around primary metrics rather than celebrating raw numbers in isolation. It also checks your ability to translate abstract UX health into concrete, measurable events in tracking code.

The full answer

A good answer hits four things in order. First, it names three to five specific counter metrics that are measurable in product telemetry, such as 7-day retention rate, sessions per user per week, core action completion rate, time to first core action, and client-side error rate. Second, it maps each counter to a distinct risk created by over-optimizing WAU, for example retention catches churn from drive-by traffic, sessions per user catches shallow engagement from notification spam, core action completion catches feature bloat confusion, and error rate catches release instability from rapid shipping. Third, it explains why these must be tracked in the same code path or experiment as the North Star, so product teams see trade-offs in real time rather than in a quarterly business review. Fourth, it distinguishes counter metrics from secondary goals by emphasizing that counters are sanity checks that should hold steady or improve, not necessarily move up and to the right.

The mistakes people make

Common wrong answers include naming revenue or lifetime value as a counter metric, since those are lagging business outcomes rather than real-time UX health signals. Another red flag is suggesting aggregate page views or total signups, which can rise while actual user experience deteriorates. Candidates also stumble by proposing survey-based metrics like NPS without explaining how they tie to the same user sessions that generate the WAU event, since the interviewer explicitly asked for tracking code implementation.

What usually comes next

Interviewers often push deeper by asking how you would set guardrail thresholds, such as requiring that a 10 percent lift in WAU never accompanies more than a 2 percent drop in 7-day retention. They may also ask how you would instrument these in an A/B testing framework, or what you would do if WAU and retention moved in opposite directions during an experiment.

A concrete example

Suppose a growth team adds an aggressive weekly email re-engagement campaign that pushes WAU up by 15 percent. Without counter metrics, the team celebrates. With counter metrics in the tracking code, they notice that sessions per user drops from 3.2 to 1.8 and the core action completion rate falls by 12 percent, revealing that users are opening the app just to dismiss the email prompt and then bouncing. The team kills the campaign before it trains users to ignore future touchpoints.