DAU/MAU Ratio

A mobile product has many people who open it once a month but few return on an ordinary day, and DAU/MAU ratio reveals the size of the daily audience relative to the monthly audience. It prompts questions about repeat value, though the right frequency depends on the product's purpose.

Amplitude describes DAU divided by MAU as one possible stickiness measure and cautions that stickiness has more than one calculation, while Mixpanel discusses how a ratio can be interpreted differently depending on activity and measurement choices. A payroll app used once a month should not be judged like a daily messaging app, and high frequency is not always desirable, since a tax-filing product may serve users well with rare visits.

Define an active user by a meaningful event, because an accidental app open may not represent useful engagement, and use unique users, not total sessions, so that one person visiting five times on a day counts once in DAU. MAU is the number of unique users active during a chosen month or trailing 30-day window, so state which version is used, and check the monthly window's end date because a numerator outside the denominator window produces an incoherent ratio.

DAU is the number active on a particular day, and for trends a daily average over the month can reduce single-day noise. Suppose 2,000 unique users are active today and 10,000 during the trailing 30 days: the day's ratio is 20%, and a ratio based on average DAU of 2,000 and the same MAU also equals 20%, but the numerator convention is different.

A weekday may differ from a weekend, so compare like days and seasons where usage patterns are strong, and keep the identity method stable, since a switch from device IDs to signed-in accounts can change unique-user counts. Bots, internal testers and duplicate accounts can inflate activity, so exclude them consistently, and respect privacy choices and consent rather than using identifiers without permission.

A growing MAU can temporarily lower the ratio when many new users have not yet formed a habit, and a rising ratio can also reflect a shrinking monthly audience while a loyal core remains, so check absolute DAU and MAU. Retention cohorts ask whether the same people return over time, while DAU/MAU is a cross-sectional ratio and does not replace cohort retention.

Segment new and established users, and segment customer groups if the product serves both daily operators and occasional administrators or has paying and free users with different patterns. A product change that makes the home screen open automatically may boost DAU without adding value, so combine frequency with task completion, user feedback, conversion and customer support signals, and track useful actions and outcomes because repeated error retries are activity but not success.

No universal good ratio exists, since it depends on expected use, category and customer workflow, and for an experiment you should compare similarly defined groups over enough time because a one-day jump is weak evidence. Document event definitions and changes in analytics instrumentation, as a new tracking tag can mimic behavioural improvement, and when interpreting a trend ask whether a person came back to do something valuable; the metric is a compact starting point for a deeper engagement review, not a verdict on product quality.