Dot Plots: How to Actually See What Your Users Are Doing

Retention curves show whether groups of users keep returning, but they do not explain how people use a product. DAU and MAU combine everyone’s activity, so they can rise even when users are not getting value. Dave argues that tracking individual behavior helps reveal usage patterns and pacing.

A dot plot places each user on a row and each time period, often a day, in a column. A dot marks an event that represents product value, such as listening to a song, sharing a photo, or processing an invoice. A ring can identify a user’s first day, and different symbols can track different actions.

The chart can reveal weekday and weekend usage patterns, as well as users who try a product once and never return. Dave describes sorting users by attributes such as platform or first-use date, and encoding their state with symbols or color. He also recalls PayPal using visual transaction patterns to help people spot possible fraud for further investigation.

In Dave’s Spotify example, a DAU chart shows daily counts such as two, three, and two, but little about who is active or why. A dot plot can show which users return and whether they use features such as search or public playlists. Dave suggests treating feature patterns as clues to investigate, rather than proof that a feature causes retention.

A company bought 10 seats on an $80,000-a-year contract, but only three people activated, and their usage was sporadic. The dot plot could have signaled risk after the internal champion left. Dave advises choosing events that represent real user value and using daily or finer time periods; he recommends combining dot plots with cohort retention curves.