Game Analytics for Player Engagement and Retention

Developing a profitable game in today’s market is a challenging endeavour. Thousands of commercial titles are published annually, across a number of hardware platforms and distribution channels, all competing for players’ time and attention. The nature of digital games mean that highly detailed information about players and their interaction with the games can be captured and used to create individual profiles. These profiles can be used to benefit the players, providing personalised recommendations and incentives to improve their enjoyment, and to benefit companies giving them insight into how to improve the design and user experience in order to maintain engagement, retain players, and increase monetisation.

DC Labs is the leading authority in this area of game analytics, with a combined publication track over 100 papers and numerous collaborations with the game industry, with notable strength when it comes to analysing the behaviour of users, players and customers in the context of interactive entertainment, in a way that is accessible to commercial partners.

We are using machine learning techniques to develop actionable behavioural profiles of computer game players, helping game designers to find out how their games are being played and the kinds of challenges players are encountering which cause them to disengage and stop playing the game.

We are building recommender systems to help players find games that suit their interests, and help industry find players that might be interested in their games. Within games we are developing real time advisor systems that provide help, training and direction to players, using AI and analytics to advise players on best choices for equipment and kit, such as weapon choices to match their playstyles. These tools have implications for both esports and for assisted play.

Using player profiles to forecast future player behaviour, and predict the players who will stay engaged and /or monetise helps companies to understand where best invest their time and effort. These prediction models take considerable resources to develop and are currently only within the realms of large companies. DC Labs researchers have significant experience in the application of novel methods for the prediction of player behaviour, and are working to develop accessible tools which combine prediction and profiling in order to derive easily actionable insights from advanced machine learning.