Predicting the monetization percentage with survival analysis in free-to-play games

Understanding and predicting player monetization is very important, because the free-to-play revenue model is so common. Many game developers now face a new challenge of getting users to buy in the game rather than getting users to buy the game. In this paper, we present a method to predict what percentage of all players will eventually monetize for a limited follow-up game data set. We assume that the data is described by a survival analysis based cure model, which can be applied to unlabeled data collected from any free-to-play game. The model has latent variables, so we solve the optimal parameters of the model with the Expectation Maximization algorithm. The result is a simple iterative algorithm, which returns the estimated monetization percentage and the estimated monetization rate in the data set.

[1]  Christian Bauckhage,et al.  Predicting Purchase Decisions in Mobile Free-to-Play Games , 2015, AIIDE.

[2]  Oliver Hinz,et al.  Making Digital Freemium Business Models a Success: Predicting Customers’ Lifetime Value via Initial Purchase Information , 2015, Business & Information Systems Engineering.

[3]  M. Pagano,et al.  Survival analysis. , 1996, Nutrition.

[4]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[5]  Tapio Pahikkala,et al.  Measuring Player Retention and Monetization Using the Mean Cumulative Function , 2017, IEEE Transactions on Games.

[6]  Christian Bauckhage,et al.  The Playtime Principle: Large-scale cross-games interest modeling , 2014, 2014 IEEE Conference on Computational Intelligence and Games.

[7]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[8]  Kyung-Joong Kim,et al.  Game Data Mining Competition on Churn Prediction and Survival Analysis Using Commercial Game Log Data , 2018, IEEE Transactions on Games.

[9]  Andrew Nealen,et al.  Exploring Game Space Using Survival Analysis , 2015, FDG.

[10]  G. McLachlan,et al.  The EM algorithm and extensions , 1996 .

[11]  Rüdiger Zarnekow,et al.  Counting Customers in Mobile Business - The Case of Free to Play , 2015, PACIS.

[12]  Djoerd Hiemstra.  Probability Smoothing , 2009, Encyclopedia of Database Systems.

[13]  Sam Devlin,et al.  Predicting player disengagement and first purchase with event-frequency based data representation , 2015, 2015 IEEE Conference on Computational Intelligence and Games (CIG).

[14]  Cristina V. Lopes,et al.  Gate Me If You Can: The Impact of Gating Mechanics on Retention and Revenues in Jelly Splash , 2015, FDG.

[15]  New York Dover,et al.  ON THE CONVERGENCE PROPERTIES OF THE EM ALGORITHM , 1983 .

[16]  Jan Sedivý,et al.  Predicting Players Behavior in Games with Microtransactions , 2014, STAIRS.

[17]  R. Kay.  The Analysis of Survival Data , 2012 .

[18]  Tapio Pahikkala,et al.  Playtime Measurement With Survival Analysis , 2017, IEEE Transactions on Games.

[19]  Bruce G. S. Hardie,et al.  Probability Models for Customer-Base Analysis , 2009 .

[20]  Xian Liu,et al.  Survival Analysis: Models and Applications , 2012 .

[21]  H. Akaike.  A new look at the statistical model identification , 1974 .

[22]  Dirk F. Moore,et al.  Applied Survival Analysis Using R , 2016 .

[23]  M. May.  Bayesian Survival Analysis. , 2002 .