Your app can gain installs while quietly losing its audience. A weekly active users count helps expose that gap sooner, but only when everyone agrees on which action makes a user “active” and how the seven-day window works.
WAU means weekly active users: the number of unique people who complete a defined activity in your product during a seven-day period. A person is counted once per window, whether they use the app once or 20 times. WAU is most useful when your product has a weekly rhythm or daily numbers are too volatile to show the underlying engagement trend.
WAU meaning: what does weekly active users measure?
WAU measures the breadth of weekly engagement, not the total volume of activity. If 800 distinct users trigger your qualifying event this week, your WAU is 800—even if those users generate 12,000 sessions.
The qualifying event matters as much as the count. Adjust defines WAU around unique users who trigger a defined event within seven days. For a messaging app, that event might be sending a message. For a marketplace, it might be viewing a product, saving an item, or purchasing. A background sync, automatic token refresh, or notification delivery should not make somebody active unless that technical event genuinely represents product value.
A practical definition is:
> A weekly active user is a distinct person who completes at least one intentional, value-related action during the selected seven-day window.
Document the event, identity rule, timezone, and window type next to the metric. Otherwise two dashboards can both say “WAU” while reporting different populations.
How to calculate WAU meaningfully
The formula is simple:
WAU = COUNT(DISTINCT user_id) for users with at least one qualifying event during seven days
Suppose a fitness app defines workout_completed as active behavior. In one calendar week:
- User A completes four workouts.
- User B completes one workout.
- User C opens the app but completes none.
- User D completes two workouts on different devices while signed into one account.
The result is three weekly active users: A, B, and D. Repeat events do not add users, C never performs the qualifying action, and D should be deduplicated to one person.
Rolling versus fixed WAU
Choose one window and label it clearly:
| Window | How it works | Best for | Main caution |
|---|---|---|---|
| Rolling seven days | Recalculates each day using that day plus the previous six | Monitoring recent movement and experiments | Adjacent values overlap by six days |
| Fixed calendar week | Counts a defined week, such as Monday through Sunday | Weekly reporting and non-overlapping comparisons | A midweek change appears only after the week closes |
Spotify Confidence’s WAU guidance notes that rolling windows respond faster, while fixed windows are cleaner for analysis because they do not overlap. Neither method is universally better. The error is switching between them without relabeling the series.
How to track WAU meaning without corrupting the number
The query may take one line; trustworthy instrumentation takes more care.
1. Define activity around delivered value
Start with the behavior your product is designed to repeat. A game could use level_completed; a banking app might use transfer_completed; a B2B dashboard could use report_viewed. Logins and app opens are acceptable for broad reach, but they can hide shallow use.
Amplitude’s active-user definition counts users who log at least one active event in the selected interval and lets teams mark non-user actions as inactive. Apply the same principle in any analytics stack: exclude noisy system events, version the definition, and record when it changes.
2. Deduplicate people, not sessions
Use a stable internal user ID after authentication and a clearly governed anonymous ID beforehand. Amplitude documents how device, platform, and user IDs can be reconciled so the same person is not counted once per device. Shared devices and late sign-ins still require explicit identity rules.
Do not put email addresses or other directly identifying data into analytics IDs. Firebase’s User-ID guidance recommends an organization-assigned identifier that outside parties cannot use to identify the person and notes that setting it affects future, not past, events.
3. Lock the calendar rules
Choose a reporting timezone, define the week start, and decide how late events and backfills are handled. Store timestamps in UTC, then apply the reporting timezone consistently. Annotate releases, outages, holidays, and tracking changes so analysts do not mistake instrumentation movement for user behavior.
4. Segment the total before acting
Break WAU down by new versus returning users, platform, app version, country, acquisition channel, and campaign. A rising total may come from a large acquisition push while existing-user engagement falls. Pair the aggregate with cohort retention to see whether new users return after their first week.
When you need to explain WAU by acquisition source, Deeplinkly can carry campaign context through deep and deferred deep links, then connect clicks, installs, and downstream events. That lets mobile teams segment weekly active users by channel or campaign instead of relying only on a blended total.
WAU meaning compared with DAU and MAU
DAU, WAU, and MAU apply the same unique-user idea to different time windows:
| Metric | Window | Use it when |
|---|---|---|
| DAU | 24 hours | The core behavior should happen daily, as in chat, news, or habit apps |
| WAU | 7 days | Usage follows a weekly cadence or daily data is noisy |
| MAU | 30 days | The product is used less often or you need a broader reach measure |
The right primary metric follows the product’s expected frequency. A tax app may be healthy with occasional monthly use; a team chat app with the same pattern likely is not.
Ratios add context. DAU/WAU estimates how much of the weekly audience is active on a typical day, while WAU/MAU shows how much of the monthly audience returns within a week. Google Analytics describes these as user-stickiness ratios, but a higher ratio is only desirable when frequent use reflects value rather than forced steps or notification fatigue.
WAU is also not retention. It mixes new and returning people in one count. Retention asks whether a defined acquisition cohort comes back later; WAU asks how many distinct people were active in the window. Use both.
How to interpret and improve WAU
Read the trend as a diagnostic signal, then investigate the cause:
- WAU up, retention stable: acquisition or reactivation may be expanding the audience.
- WAU up, returning-user share down: new users may be masking churn.
- WAU flat, DAU up: existing weekly users may be visiting more often without broader reach.
- WAU down after a release: check crashes, event delivery, login changes, and the core workflow before launching a campaign.
- WAU differs across dashboards: audit date windows, timezones, event filters, and identity stitching.
To improve genuine WAU, remove friction from the action that defines “active.” Fix onboarding drop-offs, make the next valuable action obvious, and use in-app marketing to guide users when context is available. Push notifications, email, and deep links can bring people back, but respect opt-ins and route each message to a relevant destination; a reopened app that delivers no value may lift a shallow event count without improving retention.
Review WAU with a small scorecard: total WAU, returning WAU, new WAU, WAU/MAU, qualifying actions per WAU, and cohort retention. That combination shows reach, frequency, depth, and durability without asking one metric to explain the whole product.
Frequently asked questions
What does WAU stand for?
WAU stands for weekly active users. It is the number of unique users who complete a defined activity in a product during a seven-day window.
How is WAU calculated?
Count distinct user IDs that triggered at least one qualifying event during the chosen seven days. Deduplicate repeat activity and multiple devices, and keep the event definition, timezone, and rolling or fixed window consistent.
What is a good WAU number?
There is no universal good WAU because healthy usage frequency varies by product and audience. Compare your WAU trend with your own historical baseline, eligible user base, expected cadence, and retention cohorts rather than using an unrelated industry total.
What is the difference between WAU and MAU?
WAU counts unique active users over seven days; MAU counts them over 30 days. WAU reacts faster to short-term changes, while MAU provides a broader and smoother view of active reach.
Can one person count more than once in WAU?
Not within the same seven-day window. If your identity setup treats the same person on two devices as two users, the dashboard may overcount, which is why stable user IDs and identity stitching matter.
Turn the WAU definition into a decision
Use WAU when a week matches the natural rhythm of your app, then define one meaningful activity, one identity rule, and one window policy. Track the result alongside retention and acquisition segments. If the number changes, you will know whether to fix measurement, improve the product loop, or shift campaign spend—not merely celebrate or panic over a line chart.