Firebase Predictions predefines four user segments that are dynamically generated and continuously updated based on your analytics data.
The churn segment contains users who have been active during the last 7 days but are predicted to stop using your app in the near future. Note that churn is different from uninstalling your app. Users may churn long before they uninstall an app. However, Predictions uses app uninstalls as one of the signals to train its model.
The not_churn segment contains users who have been active during the last 7 days and are predicted to continue to be active.
Note that membership to the not_churn prediction is different from not being a member of the churn prediction. Users who are not members of the churn prediction include both members of the not_churn group and users for whom no prediction could be made with confidence.
Consequently, you should not target the negative of a prediction. When you do so, you also target users the model is unable to confidently predict.
Instead, target users that have been affirmatively predicted not to perform an action, such as users in the not_churn prediction.
All predictions make some errors, and while highly unlikely, it's possible, for example, that the churn model erroneously categorizes a user as likely to churn, while at the same time, the not_churn model correctly classifies the user as a non-churner. In a situation like this, by choosing low risk tolerance, you further reduce the likelihood of this happening.
The spend prediction contains users who are likely to make in-app
or ecommerce purchases (that is, users expected to log the
event). These are users who, behaviorally, look like past spenders. You should take action to motivate this segment to convert.
The not_spend prediction contains users who are not likely not to spend in the near future. You can use this information to experiment with different monetization strategies, like for example, introduce rewarded ads for users in this prediction so they can still continue to enjoy your game.