I would like to be able to..
Replace Playerbase insights metrics with something that gives more in depth information
Highcharts offers a Sankey diagram which is good for showing changes to groups over time.
https://jsfiddle.net/4dwme1o0/1/
(Edit: Looks like this graph is flipped, more recent on the left when it's supposed to be on the right)
The data in JSFiddle was produced from a dataset of 650k sessions, of which 500k fit within last 13 weeks.
The query to produce the data took 5 seconds to query after server_id, session_end, user_id INDEX creation.
Is your feature request related to a problem? Please describe.
Playerbase tab could use some love.
The Playerbase insights show a change of how many players moved from one group to another, but a single number is not enough. Even implementing them as datapoints is limited to 4 weeks and only covers regular group
I'm writing this ticket so I don't forget to work on this, as it still needs a lot of work to get inside Plan and usable.
Another problem is the Inactive group which can get very large over time. It would be useful to separate 'Inactive' <1 and 'Not Playing' =0 groups.
I would like to be able to..
Replace Playerbase insights metrics with something that gives more in depth information
Highcharts offers a Sankey diagram which is good for showing changes to groups over time.
https://jsfiddle.net/4dwme1o0/1/
(Edit: Looks like this graph is flipped, more recent on the left when it's supposed to be on the right)
The data in JSFiddle was produced from a dataset of 650k sessions, of which 500k fit within last 13 weeks.
The query to produce the data took 5 seconds to query after server_id, session_end, user_id INDEX creation.
Is your feature request related to a problem? Please describe.
Playerbase tab could use some love.
The Playerbase insights show a change of how many players moved from one group to another, but a single number is not enough. Even implementing them as datapoints is limited to 4 weeks and only covers regular group
I'm writing this ticket so I don't forget to work on this, as it still needs a lot of work to get inside Plan and usable.
Another problem is the Inactive group which can get very large over time. It would be useful to separate 'Inactive' <1 and 'Not Playing' =0 groups.