We already tracking events from last 1+ years in mixpanel. We need one one report in mixpanel. The report should show how many days users have actively used our web app, grouped into usage ranges. For example: • Out of 400 total users: 100 users used the app for 1 day 100 users used the app for 2–3 days 120 users used the app for 4–10 days 80 users used the app for more than 10 days Can anyone help me understand how we can create this kind of report in Mixpanel, as mentioned above?
Hi Anjali N., Please stand by as I search the Mixpanel knowledge base for an AI-powered answer. I normally take about 15 seconds or so. In case AI cannot answer your question, feel free to reach out to Mixpanel support at mixpanel.com/get-support. If you have follow-up questions for this thread, please tag me again with your question: @kapa.ai <follow-up question> 👋
Feel free to ask a follow-up question once I've answered by mentioning me with @kapa.ai
You can get very close to this using the Retention → Frequency View plus Custom Retention Brackets.
1. Decide what “active use” means
Pick the core event that represents “used the app” (e.g. Any Event, Session Start, Page Viewed, etc.). This will be your retention behavior’s “B” event. [Retention quick start]
2. Open a Frequency Retention report
1. Go to Reports → Retention. [Track retention] 2. In the Retention report, switch the toggle from Retention to Frequency. [Frequency view]Frequency view “groups unique users in time‑incremented buckets when they first complete an action, and then groups those same users in subsequent buckets when they return and perform the same or different actions,” and the buckets measure how many unique time buckets a user was active (unique days in a week / month, etc.). [Frequency view]
3. Configure the query
1. Retention behavior
- A event: when users “start” (e.g. Sign Up Completed or first Any Event)
- B event: your “active use” event (e.g. Any Event)2. Time unit & window
- Choose Day as the time unit.
- Choose a long enough date range (e.g. last 12 months) so you capture all usage you care about.3. Frequency criteria
- Use Non‑cumulative Frequency if you want “exactly X days” buckets.
- Use Cumulative Frequency if you want “at least X days” buckets. [Frequency criteria]The table will now show, for each cohort (e.g. users who did A in a given week), what % / count of users did B on 1, 2, 3, … unique days in that period. [Frequency view]
4. Create your “usage ranges” with custom brackets
To group “number of days active” into ranges like 1, 2–3, 4–10, > 10, use Custom Retention Brackets: 1. In the Retention query, open the time unit dropdown and select Custom. [Custom brackets]
2. Define brackets that correspond to your ranges (e.g. 1‑day, 2‑3‑day, 4‑10‑day, 11+‑day intervals). Each interval becomes its own bucket, and “if the user does the ‘come back and did B’ event anytime inside that bucket they are counted as retained.” [Custom brackets]This will give you, per cohort, how many users fall into each “number of active days” range.
5. Read the numbers
In the Frequency table: - Each column (or custom bracket) = a range of “unique days active”. - Each cell = % of users (or absolute users, if you switch the measurement) in that cohort who were active in that many unique days. [Frequency view]You can: - Switch the measurement to Unique Users to see counts instead of percentages. [Retention measurements] - Average across cohorts (or focus on a specific signup period) to approximate a single summary like your 400‑user example.---Limitations: the docs describe Frequency view and custom brackets per cohort and per time window; they don’t show a single, all‑time, one‑row summary exactly like your mock table, so you may need to interpret or export the cohort rows to get a single global distribution. The knowledge sources do not explicitly document a one‑click “lifetime days active per user” histogram.
kapa.ai is still learning and improving, please let me know how I did by reacting below.
ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.
is there any video tutorials ?
