What parameters can be used to segment players based on the results of the first session?

What are the common parameters can be used to immediately after the end of the first day of the player as accurately as possible to assess the likelihood that the player will become active? And is it even possible?


Of course every game has its own specific parameters that can be used as a criterion for evaluation of the player. But perhaps it is possible to estimate using only basic metrics such as the duration of the first session, the number of quests/tutorials that the player passed in the first session, number of sessions on the first day after registration or something like that?


If we had historical data, we could take the most active segment of the audience, and calculate what parameters these players was different in the first day from the average values. But as yet, I want at least some feedback about the quality of player engagement and evaluation of traffic sources at the earliest stages.
October 3rd 19 at 03:57
2 answers
October 3rd 19 at 03:59
Was an article on this topic. To learn from the experience of the author, you can that in advance to choose the settings hard — in any case need to mesh and training volume. There are subsequently a good metric for games like rpg. But only the metrics, and the values for them in General (for all games at once) it is hardly possible to specify, even approximately, to an acceptable number of errors both genders.
October 3rd 19 at 04:01
This article was already read, but thanks anyway for the answer.

Find more questions by tags Big dataClusteringData mining