What Counts as an Active User in GA4, Mixpanel and Amplitude

What counts as an active user?

In analytics tools, an active user is a person who performed at least one qualifying action inside a time window, and each tool chooses its own qualifying action, so the same traffic produces a different active-user count in every product that measures it. GA4 requires signs of engagement. Mixpanel counts whoever performed the event you pick. Amplitude counts anyone who logged an event you have not marked inactive. None of them matches the others.

An active-user number means nothing until it states which action qualified and which window it covered. "12,000 monthly active users" could mean 12,000 people who loaded a page in a rolling 30 days, or 12,000 who completed the core action in a calendar month.

How does GA4 define an active user?

GA4 defines an active user as a user with an engaged session or one of a short list of engagement signals, and it is the only major tool on this page that ties the word "active" to engagement by default. The help center's user metrics article says:

An active user is any user who has an engaged session or when Analytics collects:

• the first_visit event or engagement_time_msec parameter from a website
• the first_open event or engagement_time_msec parameter from an Android app
• the first_open or user_engagement event from an iOS app

GA4 defines the engaged session in the same article: "a session that lasted 10 seconds or longer, or had 1 or more conversion events or 2 or more page or screen views." GA4's own example is a returning user who "stays on the site for less than 10 seconds without being active on the page." That user "doesn't meet any of the criteria for being active, and will only be counted in the total user count, but not in the active users count." The same article warns that "not all new users are active, so new users may exceed active users."

GA4 also contradicts itself on this definition. The Data API schema describes the activeUsers metric as "The number of distinct users who visited your site or app", with no mention of engagement. The help center uses the engaged definition above. Both pages are live. The wider split between users, sessions and pageviews is covered in pageviews vs sessions vs users, and the engagement side in engagement rate.

How do Mixpanel, Amplitude and PostHog define one?

Mixpanel, Amplitude and PostHog all define an active user as someone who performed an event, and they differ on which events count and who picks them.

Mixpanel makes the analyst choose the event. Its Insights documentation defines DAU as "The number of users who performed [event] within the last 24 hours", WAU as "within the last 7 days" and MAU as "within the last 30 days." The window is rolling: for a monthly interval, Mixpanel looks "at the last date in the time frame, and count[s] 30 days back from that", so the December MAU covers December 2 to December 31.

Amplitude counts everything by default and lets you subtract. Its glossary says "By default, an active user has logged at least one event during your selected interval: real-time (five minutes), hourly, daily, weekly, or monthly. You can edit this definition by marking certain events as inactive." Inactive events are typically "events that aren't tied to a user action but still occur within the app or website. One example is a push notification or email sent to the end user." The catch is in the next definition: a new user "has logged an event for the first time, including inactive events." A user who only received a push notification is new but never active. Amplitude's billing unit, the monthly tracked user, counts "A unique user who triggers one or more events in a project within a calendar month", with no active filter at all.

PostHog follows Mixpanel's approach. Its trends aggregation table defines WAU as "The total number of unique users who have performed this event within the last 7 days. This is a rolling count for each period, counting the previous 7 days", and MAU the same way over 30 days. The table has no DAU row. A daily count comes from the "Unique users" aggregation grouped by day.

Tool	Qualifying action by default	Who picks it	Monthly window
GA4	Engaged session, first_visit, first_open, engagement_time_msec or user_engagement	Fixed by GA4	30 days in DAU/MAU, 28 days in the API
Mixpanel	The event in the query	Analyst, per report	Rolling 30 days
Amplitude	Any event not marked inactive (Any Active Event)	Admin marks events inactive; analyst can pick a specific event	"Monthly" interval
PostHog	The event in the query	Analyst, per report	Rolling 30 days

Is MAU measured over 28 days or 30 days?

MAU is measured over 30 days in Mixpanel and PostHog, and over both 28 and 30 days inside GA4, depending on which page you read. The GA4 Data API exposes active1DayUsers, active7DayUsers and active28DayUsers, the last defined as "The number of distinct active users on your site or app within a 28 day period." The API has no 30-day active users metric. Yet the same schema defines dauPerMau as "The rolling percent of 30-day active users who are also 1-day active users."

So the denominator of GA4's DAU/MAU ratio is a 30-day count that the API does not let you pull on its own. Anyone computing the ratio by hand from active1DayUsers / active28DayUsers gets a different number from GA4's dauPerMau.

The help center adds another split. Its metrics reference describes DAU/MAU as users "who engaged for the calendar day out of the users who engaged in the last 30 days", but DAU/WAU as users "who engaged in the last 24 hours." A calendar day and the last 24 hours are different windows, sitting one row apart in the same table. The same table labels WAU/MAU as "Weekly Active Users (DAU) / Monthly Active Users (MAU)", a copy slip that leaves the reader guessing. The ratio itself, and where its benchmarks came from, is covered in the DAU/MAU ratio.

How do public companies define active users?

Public companies that report active users write their own definitions into their filings. Meta's 10-K for 2025, filed with the SEC on 29 January 2026, reports daily active people rather than users:

We define a daily active person as a registered and logged-in user of Facebook, Instagram, Messenger, and/or WhatsApp (collectively, our "Family" of products) who visited at least one of these Family products through a mobile device application or using a web or mobile browser on a given day.

Meta reports Family DAP of 3.58 billion on average for December 2025, and it describes how the number is produced: "Our calculations of DAP rely upon complex techniques, algorithms, and machine learning models that seek to estimate the underlying number of unique people." The same filing estimates that "less than 5% of our worldwide DAP consisted solely of violating accounts", meaning bots and spam. So even an active-user figure in an SEC filing is a modelled estimate.

How should you define an active user for your own product?

Define an active user by the action that shows your product did its job for that person, then write the definition down where every report can point to it. Four choices come up, each stricter than the last:

1. Any event. Easiest to count, and it includes people who opened an email or received a notification.
2. Any user-initiated event. Amplitude's Any Active Event, with system events excluded.
3. An engaged visit. GA4's approach, which filters out short single-page visits.
4. The core action. Sent a message, published a post, ran a report.

The fourth is the only one that tracks value. A count built on the first can rise while the product gets worse, which is how active users turn into a vanity metric. Put the chosen event, its trigger and the window in your tracking plan, so the definition survives the next dashboard someone builds.

Writing your own definition

Pick the core action

Name the event

Note its trigger

Set the window

Add it to the tracking plan

Frequently asked questions

What is the difference between total users and active users in GA4?

Total users counts every unique user who triggered any event in the date range. Active users counts only those with an engaged session or one of GA4's listed engagement signals, so a returning visitor who stays under 10 seconds without being active on the page appears in total users but not active users.

Is DAU in Mixpanel a calendar day or the last 24 hours?

Mixpanel defines DAU as users who performed the chosen event "within the last 24 hours." When the date range includes today, Mixpanel says "the query will take the end of the current day as the end of the query's time segment (even though it's in the future)."

Can a new user not be an active user?

Yes, in both GA4 and Amplitude. GA4 warns that new users may exceed active users. Amplitude counts inactive events, such as a push notification sent, toward new users but not toward active users.

What is an inactive event in Amplitude?

An inactive event is an event you mark as inactive in Amplitude Data so it no longer makes a user active. Amplitude gives push notifications and emails sent to the user as typical examples, since the user did nothing to cause them.

Why do GA4 and Mixpanel report different active users for the same site?

GA4 requires engagement signals and fixes the definition, while Mixpanel counts users who performed whatever event the query names. The two tools also identify users differently, so the counts differ even when both track the same pages.

How does Meta define a daily active person?

Meta defines a daily active person as a registered and logged-in user of Facebook, Instagram, Messenger or WhatsApp who visited at least one of them on a given day. Meta estimates the count with models that match multiple accounts to one person.

What is an engaged session in GA4?

GA4 defines an engaged session as one that lasted 10 seconds or longer, or had 1 or more conversion events or 2 or more page or screen views. A user with an engaged session counts as active. A returning visitor who stays under 10 seconds without being active on the page does not.

How does Mixpanel calculate MAU?

Mixpanel counts users who performed the chosen event within the last 30 days. The window is rolling, so Mixpanel looks at the last date in the time frame and counts 30 days back from that. The December MAU, for example, covers December 2 to December 31.

Does PostHog have a DAU metric?

PostHog's trends aggregation table lists WAU and MAU but has no DAU row. You get a daily count from the "Unique users" aggregation grouped by day. WAU and MAU are rolling counts over 7 and 30 days.

Can I calculate GA4's DAU/MAU myself from the API?

Not exactly. GA4's dauPerMau divides by 30-day active users, and the API only exposes active1DayUsers, active7DayUsers and active28DayUsers. Dividing active1DayUsers by active28DayUsers gives a different number from dauPerMau.