Quick answer
An active user is a unique user who performed at least one qualifying action in a product during a set window, such as a day, week or month. Who counts as a user, what counts as an action and how long the window is all change the number, so the same people can be 1,900 or 5,600 monthly active users.
Formula: Active users = unique users with ≥1 qualifying action in the window; stickiness = DAU ÷ MAU
What is an active user?
An active user is a unique user who performed at least one qualifying action in your product during a set time window. Three choices are packed into that sentence, and each one changes the number:
- Who counts as one user. A visitor, a device or a signed-in person.
- What counts as a qualifying action. Opening the app, viewing a page, finishing a session of some length, or using a core feature.
- How long the window is. A day, a week or a month.
There is no single standard. Google Analytics 4, for example, defines active users as the unique users who had an engaged session or triggered a user engagement event, and an engaged session as one that lasts 10 seconds or more, includes a conversion event or has two or more page or screen views. Meta's annual report says its daily and monthly active user metrics for Facebook have been based on activity on Facebook and Messenger only, leaving out users of its other products unless they qualify through Facebook. Two reasonable definitions, two different counts for the same people. That is why a number labelled "active users" means nothing until you know the user, the action and the window behind it. The related monthly active users and DAU/MAU pages cover those two windows in detail.
How do you calculate active users?
Active users in a window = number of unique users with at least one qualifying action in that window
The ratios built from it are what teams usually watch:
- DAU, WAU, MAU: active users in a day, a week (7 days) and a month (usually 30 days).
- Stickiness = DAU ÷ MAU. The share of a month's active users who show up on an average day.
- Active rate = active users ÷ registered users for the same period.
Sample data: 10,000 registered users and one month of activity, counted under three definitions of "active".
| Definition of active | MAU | Average DAU | Stickiness (DAU ÷ MAU) | Share of registered users |
|---|---|---|---|---|
| A. Any event, including passive ones | 5,600 | 1,120 | 20.0% | 56.0% |
| B. An engaged session | 4,300 | 1,075 | 25.0% | 43.0% |
| C. A core action, such as creating a report | 1,900 | 570 | 30.0% | 19.0% |
The arithmetic is 1,120 ÷ 5,600 = 20.0%, 1,075 ÷ 4,300 = 25.0% and 570 ÷ 1,900 = 30.0%. The same user base is 5,600, 4,300 or 1,900 monthly active users depending on one definition, a threefold range, and stickiness moves in the opposite direction from the user count: the stricter the definition, the fewer users and the higher the stickiness. A report that celebrates the 5,600 and one that celebrates the 30.0% stickiness can both be true on the same data. Pick one definition, keep it fixed and compare it only with itself over time.
How does the length of the window change the count?
For a fixed definition, a longer window always counts at least as many users, because anyone active in a shorter window is also active in the longer one that contains it.
Sample data: definition B (an engaged session), same 10,000 registered users.
| Window | Active users | Share of registered users |
|---|---|---|
| 1 day (average) | 1,075 | 10.8% |
| 7 days | 2,580 | 25.8% |
| 28 days | 4,150 | 41.5% |
| 30 days | 4,300 | 43.0% |
The 28-day and 30-day rows are the reminder: a "monthly" figure from a 28-day window and one from a 30-day window are not interchangeable, and comparing a 28-day figure with last month's 30-day one makes growth look smaller than it is. Choose a rolling window rather than a calendar month if you want periods that are the same length, and write the length next to the figure.
Window length also decides who is "lost". A user who was active 29 days ago is a monthly active user today and not a weekly one, so a falling weekly figure with a flat monthly figure often means people are visiting less often, not that they have left.
What counts as a qualifying action?
The action should match how the product creates value. Four common choices, from loosest to strictest:
| Action rule | Counts as active when… | Good for | Watch out for |
|---|---|---|---|
| Any activity | The user did anything, including a passive event | Reach and traffic | Counts people who opened and left, and automatic events |
| Session started | The user opened the product | Mobile apps and daily habits | Inflated by notifications that bring people in for a second |
| Engaged session | A session of a minimum length, with a conversion or several views | Websites and content | The length and views thresholds are your choice and must be written down |
| Core action | The user did the thing the product exists for | Products with a clear core loop | Misses people who read and learn without ever "doing" |
Two rules keep the choice honest. First, exclude events the system triggers rather than the user, such as renewal charges, admin approvals or automated emails, because they make every user look active once a month and inflate retention. Second, if your product has two kinds of use, such as reading and creating, report both rather than blending them into one number.
What does "active" miss, and how do teams misuse it?
- Mixing definitions. Daily, weekly and monthly active users must share one definition and one user rule. If DAU counts any event and MAU counts engaged sessions, the ratio between them is not stickiness.
- Counting devices as people. If a person uses a phone and a laptop and you count by device, one person is two active users. Count by signed-in person where you can.
- Letting anonymous traffic dominate. Visitors who never sign in have no profile, so they cannot be grouped or followed, and crawlers and bots can inflate the count if they are not separated from real people.
- Treating active as healthy. A user who opens the app and does nothing is active under a start-based rule. A growing active count can hide a shrinking core of people who actually get value, which is why you should read it next to a core-action count.
- Reading a range total as a daily figure. Adding each day's active users over a month counts the same person many times. A daily figure should be an average of daily values, and a monthly figure should be a deduplicated count over the whole window.
How do you work with active users in TapCub?
TapCub lets you set the action rule instead of assuming one, and it keeps the identity choice visible.
- Decide which events count. In All-platform analytics › Data › Events & properties, the Counts as active column decides whether an event means the user "showed up". Auto (the default) counts events sent from a browser, app or mini program and does not count server-side events; Counts makes a server-side event count, which suits actions the user takes that your backend reports, such as sign-up, login or an order; Doesn't count excludes events the system triggers, such as renewal charges. Active users, DAU, WAU and MAU, new users, retention and lifecycle all follow this column, and changing it recalculates the last 90 days.
- Choose the identity. Analyses default to counting people: the identified ID where there is one, otherwise the device, otherwise the visitor. An unidentified person on a new device looks like a new user, so turn on identified mode and report logins if you want one row per person.
- Read the app numbers as defined. In App analytics › Users › Users, active users are people with a start or a screen view in the range. Weekly is the seven days ending on the last day of the range, deduplicated, and monthly is the thirty days ending that day. The main daily card is the daily average, because a range total would add each day's people without deduplicating them.
- Build a core-action count. Under All-platform analytics › Behavior › Metrics & SQL, define a metric as an event plus a unique-people aggregation, such as the people who created a report, and put it next to the any-activity figure.
- Check stickiness. The lifecycle view in All-platform analytics › Behavior › Behavior analysis reads active days per week or month, with an optional activity event, so you can see how many people are active on one day of the week versus five.
Sources
- Google Analytics Help, [[GA4] Dimensions and metrics](https://support.google.com/analytics/answer/12253918) — definitions of active users, total users, new users and engaged session (10 seconds or more, a conversion event or at least two page or screen views). Checked 2026-11-21.
- Meta Platforms, Form 10-K for fiscal 2023 — notes that its Facebook DAU and MAU metrics have been based on activity on Facebook and Messenger only. Checked 2026-11-21.
Was this explanation clear?
We use your answers to fix and expand this page.
Couldn’t send. Please try again in a minute.
Thanks for the feedback. We’ll use it to improve this page.You’ve already sent feedback on this page. Thanks.

Technical Engineer
Focuses on tracking code, SDKs, and API integrations, with practical guidance on implementation checks and troubleshooting.