Guide

Growth and monetization

Engagement metrics

Retention asks whether they come back. Engagement asks how deeply they use it. DAU/MAU, L7 and power-user curves only work if you match the product's natural frequency.

83 of 128 · 2 min read

Retention answers "do they come back?" Engagement answers "how deeply do they use it when they do?" Early PMF work usually prioritizes flattening the retention curve (breadth). Later engagement work deepens intensity for users who already retained. Deepening engagement for people who never activated wastes effort — fix activation first.

Trade-off

Optimize for breadth (more users returning at all) vs depth (power users doing more). Breadth first. Depth later, and only for people who already found the job.

The big idea

  • DAU / MAU (stickiness): daily active users ÷ monthly active users. A product used most days by its monthly base sits near 50–60%+; a monthly utility might sit at 10–20% and still be healthy. High DAU/MAU with shrinking MAU is a shrinking cult, not growth.
  • L7 / L28: days active out of the last 7 or 28. Distributions beat averages — a mean of 3 days/week can hide half the users at 1 and a power cohort at 7.
  • Power user curves (PUC): histogram of users by days active in a period. Healthy consumer products often show a "smile": many light users, a dip, then a spike of power users. A single pile at 1 day is a warning.

Match the metric to natural frequency. Daily tool (chat, mobile social): D1 / D7 / D30 plus DAU/MAU. Weekly tool: W1 / W4 / W12. Monthly / episodic (invoicing, compliance): M1 / M3 / M6, and be careful with DAU vanity. Using D30 for a monthly product makes healthy customers look churned. Using MAU-only for a daily messenger hides a collapse in daily habit.

Write the natural frequency into the metric doc so new hires do not "improve" retention by changing the definition.

Walking through it

A B2B project tool used once a week by healthy teams sits at DAU/MAU ≈ 0.2. A growth lead copies a consumer social target of 50%+ and launches daily push. Open rate rises; opt-out rises; week-4 retention does not. The stickiness target was from the wrong category.

A consumer social app at DAU/MAU 55% with MAU falling 10% month on month is not "highly engaged growth." It is a shrinking core. Pair stickiness with the stock (MAU or retained cohort size) or you will celebrate a cult.

Common mistakes

  • Copying DAU/MAU targets across categories. Weekly B2B is not daily social.
  • Averages without the histogram. Mean L7 of 3 can hide a pile at 1.
  • High stickiness, shrinking MAU. Depth of a dying base is not growth.

Check yourself

  1. DAU/MAU is 55% and MAU is down 10%. Healthy?

No. High stickiness on a shrinking base is a cult, not growth. Pair the ratio with the stock.

  1. Your invoicing product is judged on D30. What goes wrong?

Healthy monthly customers look churned. Use M1 / M3 / M6 for an episodic job.

Key takeaways

  • Breadth (flatten the curve) before depth (power users).
  • DAU/MAU, L7 and PUCs only work at the product's natural frequency.
  • Distributions beat averages; pair stickiness with the size of the base.

Further learning

16 Startup Metrics (a16z)We have the privilege of meeting with thousands of entrepreneurs every year, and in the course of those discussions are presented with all kinds of numbers, measures, and metrics that illustrate the promise and health…Andreessen Horowitz

Next: what actually moves the curve in retention levers.

Retention and engagement

Retention curves, cohort analysis, DAU/MAU and why retention is the foundation of growth. Start here for the map.

Retention curves

Three shapes matter — decay to zero, flatten above zero, smile. Flattening is the product-market-fit signal. Compare cohorts, not blended actives.

Retention levers

Habit loops, careful notifications, resurrection, critical-path quality and — for paid products — splitting voluntary vs involuntary churn. Diagnose the curve before you add an engagement feature.