Reducing churn rate for a social network service
Systems and methods for reducing a churn rate associated with subscribers of social network services are described. In some example embodiments, the systems and methods may access activity information associated with a former subscriber of a social network service, compare the accessed activity information to activity information associated with subscribers of the social network service, identify one or more differences between the activity information associated with the former subscriber of the social network service and the activity information associated with the subscribers of the social network service, and perform an action based on the identified one or more differences.
1. A system, comprising:
a processor for executing instructions stored in a memory device;
at least one memory device storing instructions, said instructions comprising:
comparing first activity information of a plurality of current subscribers of a software service to second activity information of a particular current subscriber of the software service, the first activity information representing at least one type of activity of the software service performed by the plurality of current subscribers the second activity information representing at least one type of activity of the software service performed by the particular current subscriber;
determining, based on a first weight assigned to activities of the first activity information and a second weight assigned to activities of the second activity information, a probability that the particular current subscriber will terminate software service subscription, the probability based at least on a difference of a first performance level of the at least one type of activity in the first activity information and a second performance level of the at least one type of activity in the second activity information, the determining including calculating a first product of the first weight and the first performance level and a second product of the second weight and the second performance level and summing the first product and the second product; and
causing display of a prompt to the particular current subscriber, the prompt indicating a particular type of activity to be performed by the particular current subscriber to reduce the probability that the particular current subscriber will terminate software service subscription, the particular type of activity being an activity that was previously performed by the plurality of current subscribers but not by the particular current subscriber.
2. The system of claim 1 , wherein comparing first activity information of at least one former subscriber of a software service to second activity information of a current subscriber of the software service includes:
identifying a first type of activity represented in both the first and the second activity information.
3. The system of claim 2 , wherein identifying a first type of activity represented in both the first and the second activity information includes:
identifying at least one respective instance of the first type of activity in both the first and the second activity information, each respective instance of the first type of activity performed during an initial subscription time period.
4. The system of claim 2 , wherein determining a probability that the current subscriber will terminate software service subscription further comprises:
determining the probability based in part on a type of job title present in profile data of the particular current subscriber.
5. The system of claim 4 , wherein the type of job title represent a senior level job title.
6. The system of claim 2 , wherein the first type of activity comprises an average number of searches in the software searches.
7. The system of claim 2 , wherein the first type of activity comprises an average number of software service profile views.
8. The system of claim 2 , wherein the first type of activity comprises an average number of messages sent within the software service.
9. A computer-implemented method comprising:
comparing first activity information of a plurality of current subscribers of a software service to second activity information of a particular current subscriber of the software service, the first activity information representing at least one type of activity of the software service performed by the plurality of current subscribers the second activity information representing at least one type of activity of the software service performed by the particular current subscriber;
determining, based on a first weight assigned to activities of the first activity information and a second weight assigned to activities of the second activity information, a probability that the particular current subscriber will terminate software service subscription, the probability based at least on a difference of a first performance level of the at least one type of activity in the first activity information and a second performance level of the at least one type of activity in the second activity information, the determining including calculating a first product of the first weight and the first performance level and a second product of the second weight and the second performance level and summing the first product and the second product; and
causing display of a prompt to the particular current subscriber, the prompt indicating a particular type of activity to be performed by the particular current subscriber to reduce the probability that the particular current subscriber will terminate software service subscription, the particular type of activity being an activity that was previously performed by the plurality of current subscribers but not by the particular current subscriber.
10. The computer-implemented method of claim 9 , wherein comparing first activity information of at least one former subscriber of a software service to second activity information of a current subscriber of the software service includes:
identifying a first type of activity represented in both the first and the second activity information.
11. The computer-implemented method of claim 10 , wherein identifying a first type of activity represented in both the first and the second activity information includes:
identifying at least one respective instance of the first type of activity in both the first and the second activity information, each respective instance of the first type of activity performed during an initial subscription time period.
12. The computer-implemented method of claim 10 , wherein determining a probability that the current subscriber will terminate software service subscription further comprises:
determining the probability based in part on a type of job title present in profile data of the particular current subscriber.
13. The computer-implemented method of claim 12 , wherein the type of job title represent a senior level job title.
14. The computer-implemented method of claim 10 , wherein the first type of activity comprises an average number of searches in the software searches.
15. The computer-implemented method of claim 10 , wherein the first type of activity comprises an average number of software service profile views.
16. The computer-implemented method of claim 10 , wherein the first type of activity comprises an average number of messages sent within the software service.
17. A non-transitory computer-readable storage medium whose contents, when executed by a computing system, cause the computing system to perform operations, comprising:
comparing first activity information of a plurality of current subscribers of a software service to second activity information of a particular current subscriber of the software service, the first activity information representing at least one type of activity of the software service performed by the plurality of current subscribers the second activity information representing at least one type of activity of the software service performed by the particular current subscriber;
determining, based on a first weight assigned to activities of the first activity information and a second weight assigned to activities of the second activity information, a probability that the particular current subscriber will terminate software service subscription, the probability based at least on a difference of a first performance level of the at least one type of activity in the first activity information and a second performance level of the at least one type of activity in the second activity information, the determining including calculating a first product of the first weight and the first performance level and a second product of the second weight and the second performance level and summing the first product and the second product; and
causing display of a prompt to the particular current subscriber, the prompt indicating a particular type of activity to be performed by the particular current subscriber to reduce the probability that the particular current subscriber will terminate software service subscription, the particular type of activity being an activity that was previously performed by the plurality of current subscribers but not by the particular current subscriber.
18. The non-transitory computer-readable storage medium of claim 17 , wherein comparing first activity information of at least one former subscriber of a software service to second activity information of a current subscriber of the software service includes:
identifying a first type of activity represented in both the first and the second activity information.
19. The non-transitory computer-readable storage medium of claim 18 , wherein identifying a first type of activity represented in both the first and the second activity information includes:
identifying at least one respective instance of the first type of activity in both the first and the second activity information, each respective instance of the first type of activity performed during an initial subscription time period.
20. The non-transitory computer-readable storage medium of claim 18 , wherein determining a probability that the current subscriber will terminate software service subscription further comprises:
determining the probability based in part on a type of job title present in profile data of the particular current subscriber.