Modelling and Advanced Research
Download Technical Note
The Technical Note below is the first in a ongoing series around Playnomics’ perspective on predicted LTV for players of online games.
We rely on a broad array of behavioral scoring algorithms to parse out player behavior, one of the most important being the calculation of predicted player lifetime value.
Playnomics is especially interested in behaviors that determines a player’s long-term value. An accurate prediction of lifetime value enables disproportionate resources to be allocated to retain the best players and keeping them loyal.
The note covers the potential factors to account for in calculating player value. In the future, we will cover subjects such as segmentation of players by LTV and reward systems that encourage the maximization of best player behavior.
Playnomics researchers have worked with a variety of game publishers, from casual browser based games, to social facebook games, to client download MMOs to help analyze and optimize key decisions in acquisition, production, marketing, community management, and monetization.
At it's core all of our R&D efforts apply the same method, to first segment and classify users by their historic behaviors, and then make predictions about those segments.
The Playnomics team is constantly advancing our algorithms and methods. We are looking for interesting data sets and invite you to contact us if the following areas of study of interest to you.
Key areas of interest include:
Behavorial segmentation of Players
Any gamer knows not all players are created equal. Who is the core fan base that spends money? Which players are competitive and create a challenging environment to play? Which players are social and drive new players to join? Understanding the core groups in your player population can give insights into a healthy game ecosystem and context for valuing players aside from monetization.
Predicted lifetime value calculations
The value of a player in a specific game may be very different based on context. Once you get past the population of whales and big spenders, how do you discern the freeloaders and the trolls from the players who are helping to create a flourishing environment?
Virtual item recommendation and pricing
As free to play games expand in popularity, targeting advertising of virtual items is needed to help users sift through the growing number of options. How do know which item to recommend and at what time in their game experience? Which item attributes determine those that sell well and can be used to inform production and pricing decisions?