Devise a single metric to measure retention of listeners using the service on a daily basis for the last month

feature_engineering
business_analytics
statistics

#1

I have dummy data for a music streaming service, the data is as follows:-

Monthly uniques = the number of unique listeners on each platform for the entirety of a given month

All-time uniques = the total number of unique listeners that have ever used that platform at the end of a given month (i.e. from platform launch until the last day of that month)

Hours = the aggregate number of listener hours across all listeners

Daily uniques = the number of unique listeners on each platform per day

What is the best method to devise a single metric to measure how successful we were at retaining listeners using the service on a daily basis during December 2014, only using the data above?


#2

Hi @rohan2387,
I can suggest that you could take out groupby of the users per month, on weekly and a day of the week basis then compare it to the average users in the respective group by. This should give you a brief report of all facts and analysis. And then, according to your business case, a metric can be defined to measure success.

P.S - I have done something similar before with opensource chat channel. I was measuring the success with an average lifetime of a contributor and average engagement of social chats on weekday basis per month. Do have a look at the report and let me know if it helped. Thanks!