Smart recommendation engine for preventing churn and providing prioritized insights
A recommendation engine can provide recommendations with respect to an application and can provide insights to a user of a computing device. The recommendation engine can receive a prediction based on user engagement with the application during an initial period of time (e.g., a trial period) as to whether the user will convert use of the application to a paid basis (e.g., a subscription or license to the application). An action can be recommended based on the prediction. The recommendation engine can provide insights to a user based on a score associated with the insight. The score can be determined by measuring previous user interactions with the insight over a period of time.
1. A method for preventing application churn, the method comprising:
receiving one or more indicators of user activity within an application during a time period;
determining, based at least in part, on the one or more indicators of user activity, a likelihood that the user will continue to use the application on a paid basis;
providing an actions library comprising a first set and a second set of recommended actions;
in response to determining that the likelihood that the user will continue to use the application on a paid basis is above a predetermined or configurable threshold, determining a recommended action from the first set of recommended actions; and
in response to determining that the likelihood that the user will continue to use the application on a paid basis is below a predetermined or configurable threshold, determining a recommended action from the second set of recommended actions.
2. The method of claim 1 , wherein said receiving the one or more indicators of user activity comprises receiving an indication of at least one of opening the application, clicking on a user interface element of the application, opening a file by the application, and application operation statistics.
3. The method of claim 1 , further comprising determining the likelihood based, at least in part, on information external to the application.
4. The method of claim 3 , wherein the information external to the application comprises at least one of date, time, day of week, demographic information about a user, pay date, or weather information.
5. The method of claim 1 , wherein said determining the recommended action from the first set of recommended actions comprises providing a recommendation for an upgrade to the application.
6. The method of claim 1 , wherein said determining the recommended action from the second set of recommended actions comprises at least one of providing a discount for the application, providing a recommendation of a second application, providing a feedback prompt to a user, or providing an advertisement to the user.
7. A non-transitory machine-readable storage medium having stored thereon computer-executable instructions for observing device events, the computer-executable instructions to cause one or more processors to perform operations comprising:
receive one or more indicators of user activity within an application during a time period;
determine, based at least in part, on the one or more indicators of user activity, a likelihood that the user will continue to use the application on a paid basis;
provide an actions library comprising a first set and a second set of recommended actions;
in response to a determination that the likelihood that the user will continue to use the application on a paid basis is above a predetermined or configurable threshold, determine a recommended action from the first set of recommended actions; and
in response to a determination that the likelihood that the user will continue to use the application on a paid basis is below the predetermined or configurable threshold, determine a recommended action from the second set of recommended actions.
8. The non-transitory machine-readable storage medium of claim 7 , wherein the operations to receive the one or more indicators of user activity comprise operations to receive an indication of at least one of opening the application, clicking on a user interface element of the application, opening a file by the application, and application operation statistics.
9. The non-transitory machine-readable storage medium of claim 7 , wherein the operations further comprise operations to determine the likelihood based, at least in part, on information external to the application.
10. The non-transitory machine-readable storage medium of claim 9 , wherein the information external to the application comprises at least one of date, time, day of week, demographic information about a user, pay date, or weather information.
11. The non-transitory machine-readable storage medium of claim 7 , wherein the operations to determine the recommended action from the first set of recommended actions comprise operations to provide a recommendation for an upgrade to the application.
12. The non-transitory machine-readable storage medium of claim 7 , wherein the operations to determine the recommended action from the second set of recommended actions comprise operations to perform at least one of provide a discount for the application, provide a recommendation of a second application, provide a feedback prompt to a user, or provide an advertisement to the user.
13. A system for preventing application churn, the system comprising:
one or more processors; and
a non-transitory machine-readable medium having stored thereon computer-executable instructions to cause the one or more processors to:
receive one or more indicators of user activity within an application during a time period;
determine, based at least in part, on the one or more indicators of user activity, a likelihood that the user will continue to use the application on a paid basis;
provide an actions library comprising a first set and a second set of recommend actions;
in response to a determination that the likelihood that the user will continue to use the application on a paid basis is above a predetermined or configurable threshold, determine a recommended action from the first set of recommended actions; and
in response to a determination that the likelihood that the user will continue to use the application on a paid basis is below the predetermined or configurable threshold, determine a recommended action from the second set of recommended actions.
14. The system of claim 13 , wherein the computer-executable instructions to receive the one or more indicators of user activity comprise instructions to receive an indication of at least one of opening the application, clicking on a user interface element of the application, opening a file by the application, and application operation statistics.
15. The system of claim 13 , wherein the computer-executable instructions further comprise instructions to determine the likelihood based, at least in part, on information external to the application.
16. The system of claim 15 , wherein the information external to the application comprises at least one of date, time, day of week, demographic information about a user, pay date, or weather information.
17. The system of claim 13 , wherein the computer-executable instructions to determine the recommended action from the first set of recommended actions comprise instructions to provide a recommendation for an upgrade to the application.
18. The system of claim 13 , wherein the computer-executable instructions to determine the recommended action from the second set of recommended actions comprise instructions to perform at least one of provide a discount for the application, provide a recommendation of a second application, provide a feedback prompt to a user, or provide an advertisement to the user.