You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi all! Right now we're using record_sessions_percent to roll out session replay to a small percentage of our users. However, due to the way that Mixpanel uses Math.random to calculate a page view-level eligibility for session replay, it's hard to compare replays in a funnel or between different sessions by the same user.
Instead, Mixpanel should hash the user's distinct_id to calculate a stable identifier that can be used for determining whether the user is in the "record sessions" cohort, so that replays from the same user can be analyzed together.
The text was updated successfully, but these errors were encountered:
This is largely a product decision (whether you want user-based sampling or not), but we are indeed planning to offer some more complete controls over sampling in future releases. In the meantime, if you want user-based sampling, the recommended workaround is to write your own sampling logic and then initialize with record_sessions_percent set to either 0 or 100 based on your determination of whether the current user falls in sample or not.
Hi all! Right now we're using record_sessions_percent to roll out session replay to a small percentage of our users. However, due to the way that Mixpanel uses
Math.random
to calculate a page view-level eligibility for session replay, it's hard to compare replays in a funnel or between different sessions by the same user.Instead, Mixpanel should hash the user's distinct_id to calculate a stable identifier that can be used for determining whether the user is in the "record sessions" cohort, so that replays from the same user can be analyzed together.
The text was updated successfully, but these errors were encountered: