All guides

    What is sessions per user?

    Sessions per user is how many times the average person opened your product during the period.

    Take every visit by an identified user in the last 30 days, divide by the number of active users in the same 30 days. 300 visits from 100 people gives you 3.

    Both halves count identified people only, so anonymous traffic can't inflate it.

    What it tells you

    Frequency. Whether your product is part of someone's week or something they remember occasionally.

    Repeat rate answers a yes-or-no question: did they come back at all? Sessions per user answers how often. A product can have 60% repeat rate and 2.1 sessions per user, which means most people come back exactly once and then stop.

    The gap between the two is the interesting part. A high repeat rate with a low average tells you people are willing to return but nothing is pulling them in regularly.

    What good looks like

    It depends on how often your product is supposed to be used, so there is no target. What matters is whether the number matches the rhythm you designed for.

    On a 30-day window:

    • Around 1. People try it and leave. Whatever the intended rhythm, this isn't it.
    • 2 to 4. Roughly weekly. Fine for a tool tied to a weekly task.
    • Above 8. People are opening it most working days. That's a habit, and it's worth protecting before you chase anything else.

    Write down the number you expect before you look. A product built for monthly invoicing landing at 1.2 is working. A messaging tool landing at 1.2 is not.

    How to raise it

    Give the product a trigger it owns. Products people open often usually have something arriving: a summary, a mention, a change someone else made. Without a trigger, you rely on memory, and memory loses.

    Shorten the loop. If the useful output takes twenty minutes to produce, people will batch it. If it takes two, they'll do it as it comes up.

    Check the sessions are real before you optimise. A single visit that gets split by a timeout will double this number without anyone visiting twice, so confirm the shape of your tracking first.

    Read it next to engagement. More visits that each do less is usually a symptom, not progress.

    What counts as a session?+

    One continuous visit. A user who comes back the next day starts a new one. A user who leaves a tab open over lunch and returns to it does not.

    How is this different from repeat rate?+

    Repeat rate is the share of people who came back at all. This is how many times the average person came. Repeat rate can look strong while this sits just above 1, which means almost everyone returns exactly once.

    Are anonymous visits included?+

    No. Both the visit count and the user count are restricted to identified people, so the ratio stays honest. Anonymous visits do count towards engagement.

    Why does mine show a dash?+

    No identified users were active in the period, so there's no one to average across.