What does a DAU/MAU stickiness ratio of 0.2 tell you about how a product is used?
answer
- measures frequency, not audience size
- denominator is unique users in the window
- sum of daily counts is total user-days
- multiply the ratio by days in the window
- a daily-visiting core hides inside the mean
basics
~10 sA DAU/MAU of 0.2 means the typical monthly active user opens the product on about 6 days out of 30. It measures return frequency, not audience size, and averages over very different user types.
solid answer
~50 sDAU/MAU divides average daily active users by monthly active users, where MAU counts unique users active at least once in the 30-day window. Because the sum of daily active counts over the window equals total active user-days, the ratio is exactly average active days per monthly active user divided by 30 — so 0.2 means roughly 6 active days per user per month. Its floor is about 1/30 (everyone visits once) and its ceiling is 1 (everyone visits every day). The big caveat is that it is a mean over a very skewed distribution: a small core visiting daily plus a long tail visiting once can produce the same 0.2 as everyone visiting on 6 days. It is also only interpretable against a product's natural cadence — 0.2 is weak for a messaging app and strong for a tax-filing tool.
go deeper
Be ready to state the formula, convert 0.2 into about 6 active days per month, and say plainly that it measures return frequency rather than audience size or loyalty.
Explain why the ratio equals average active days divided by window length, give its 1/30 to 1 bounds, and show how the activity definition or window choice changes the number without any behaviour changing.
Show you would read the distribution of active days per user behind the mean, segment by acquisition source before calling a movement real, and refuse comparisons across products with different natural cadences.
Own the question of whether stickiness deserves to be a company-level health number at all, what distribution you publish next to it, and how you stop teams from gaming it with notification volume.
## What the ratio is **DAU** (daily active users) is the count of distinct users who performed a qualifying action on a given day. **MAU** (monthly active users) is the count of distinct users who performed that action at least once over a 30-day window. **Stickiness** is usually reported as mean DAU over the window divided by MAU for the same window. Both counts use the same activity definition and the same product; only the window length differs. That is what makes the ratio interpretable. ## Why 0.2 means about 6 days a month Sum the daily active counts across the 30 days of the window. Each user contributes one to that sum for every day they were active, so the sum equals **total active user-days**. Dividing by 30 gives mean DAU. So: ``` mean DAU / MAU = (total active user-days / 30) / MAU = (average active days per monthly active user) / 30 ``` With a ratio of 0.2, the average monthly active user was active on `0.2 * 30 = 6` days. This identity is exact given the definitions above; it is not an approximation or a modelling assumption. The bounds follow immediately. Every monthly active user contributes at least one active day, so mean DAU is at least `MAU / 30` and the ratio cannot fall below about `1/30 = 0.033`. If every monthly active user were active on all 30 days, mean DAU would equal MAU and the ratio would be 1. A ratio outside `[1/30, 1]` signals a definitional mismatch — for example DAU and MAU counting different events, or MAU being computed over a calendar month while DAU is averaged over a different span. ## What it does and does not tell you It tells you **frequency of return within a month**. It is a shape statistic, not a size statistic: doubling acquisition with the same usage habits leaves the ratio unchanged, which is exactly why teams like it as a health measure independent of marketing spend. It does **not** tell you: - **Who is engaged.** The ratio is a mean over a heavy-tailed distribution of per-user active days. A product where 20% of users come every day and 80% come once has roughly the same ratio as one where every user comes on 6 days, and those are completely different products with completely different risk profiles. Read the histogram of active days per user, not just its mean. - **Whether users stay.** A cohort can be highly frequent for a month and then vanish. Frequency within a window and survival across windows are different properties, and a product can be strong on one and weak on the other. - **Whether the number is good.** Interpretation depends on the product's natural cadence. Communication and feed products expect people back daily, so 0.2 is a warning sign there. Tools people need occasionally — filing taxes, booking travel, paying a bill — have a low intrinsic ceiling, and pushing their ratio up may just mean sending notifications that annoy people. ## Practical traps **Activity definition drives the number.** If a background sync, a push-notification receipt, or an automated job counts as "active", the ratio inflates without any human behaviour changing. Two teams quoting stickiness for the same product routinely disagree because they qualify different events. **Window length is not neutral.** DAU/WAU is a stricter statistic than DAU/MAU on the same product, and the two are not comparable. Weekly windows also absorb weekday/weekend rhythm that daily counts expose. Always state the pair being divided. **Mix effects.** A surge of one-time visitors from a marketing campaign raises MAU immediately and raises mean DAU only slightly, so the ratio falls even though the core user base is unchanged. Segment by acquisition source before reading a movement as an engagement change. **It moves slowly and is easy to over-read.** Because MAU is a 30-day rolling count, both numerator and denominator smooth over shocks, so small week-to-week wiggles are mostly noise. ## How to use it well Quote it alongside the distribution it summarises: the share of monthly actives who were active on 1 day, 2-5 days, 6-15 days, and 16+ days. Track that distribution over time. A stable 0.2 that hides a shrinking daily core and a growing one-visit tail is a deteriorating product with a flat headline number — and that is the failure mode the single ratio is worst at exposing.
- What is the theoretical range of DAU/MAU over a 30-day window, and what does a value outside it mean?The floor is about 1/30, reached when every monthly active user visits on exactly one day, because each of them contributes at least one active day to the sum. The ceiling is 1, reached when every monthly active is active all 30 days. A reported value outside that range means the numerator and denominator are not measuring the same population or the same event — for example DAU counting sessions while MAU counts users.
- Would you rather see stickiness rise from 0.20 to 0.25 with flat MAU, or MAU grow 30% with stickiness falling to 0.16?It depends what the growth is made of. Flat MAU with rising stickiness means existing users found more reason to return, which usually compounds. A 30% MAU jump with stickiness down to 0.16 can be healthy if the new users are simply young and have not built a habit yet, or hollow if they are one-visit traffic from a campaign. Split the ratio by tenure and acquisition source before judging.
- Why might DAU/WAU be a more honest engagement measure than DAU/MAU for some products?A 7-day window is far less forgiving: a user has to return within the week to stay in the denominator, so the ratio reacts faster to habit forming or breaking. DAU/MAU can look stable for weeks because MAU keeps counting people whose last visit was 29 days ago. For products with an expected weekly rhythm, DAU/WAU exposes real changes sooner. The two are on different scales and must never be compared to each other.
It is like judging a gym by average visits per member per month. Six visits could be everyone going weekly, or a tenth of members going daily while the rest went once in January.
saying these in an interview costs you the question
- Says 0.2 means 20% of users open the app every day
- Treats stickiness as a retention or churn measure
- Compares DAU/WAU with DAU/MAU as if the same scale
- Ignores that a heavy-tailed usage distribution hides in the mean
- Judges 0.2 as bad without asking the product's natural cadence