Monitoring Product Health

• Founders frequently rely on aggregate DAUs or MAUs which can rise during growth phases even while individual users find no persistent value in the product.0:09
• A dot plot visualizes activity as a row-by-column grid where each row is an individual user and each column represents a day, using symbols to mark high-value events rather than trivial ones like app opens.1:42
• This granularity exposes critical patterns, such as the difference between weekday and weekend users or the "one-day-churn" signature common in failed onboarding cycles.3:45

Preventing Enterprise Churn

• B2B products often appear healthy based on contract metrics, but seat-level data can reveal existential risk before renewal arrives.10:38
• In one instance, a $80,000 annual contract for 10 seats churned because only three seats actually activated, and usage was limited to sparse, twice-weekly activity.
• The account’s collapse was finalized when its internal champion departed, leaving a new stakeholder to question the software’s necessity—a vulnerability that high-level utilization charts failed to flag.11:32

Scaling and Interpretation

• While aggregate cohort retention curves measure overall stickiness, dot plots function as the essential complement for identifying the behavioral "why" behind retention drops.13:05
• The method remains scalable for massive products by sampling user segments; Google Photos historically used printed dot-plot pages for segmented groups to spot anomalies across billion-user datasets.9:53
• Human pattern recognition is often faster than standard alerting, just as early PayPal teams visually spotted fraud by scanning dense transaction patterns that automated systems missed.6:41