Studying Video Gamers for Improving Customer Engagement
Michigan Ross professors find video game players respond to different incentives depending on their level of play.
Customer engagement is one of those ideas that’s often talked about, difficult to define, and even harder to measure.
But it’s critical to figure out for makers of video games, who operate in a noisy, crowded, and competitive industry.
Three Michigan Ross professors who got access to player data for a popular first-person shooter game took a novel approach and built a model that measured customer engagement based on game play.
They found that players with different levels of engagement respond to different incentives for continued play. Using their model to match players — as opposed to the random matching commonly used in the industry — resulted in higher customer retention.
“We got excited about this not only because of the rich data, but because there is little research on game play behavior,” says Manchanda.
The three professors studied the game play data of 1,309 players from October 2011 to March 2014 — that’s 710,212 unique rounds. Analyzing the data, they identified three levels of player engagement — low, medium, and high, and showed the players at different levels of engagement responded to different motivators for continued play.
The authors found that players at the highest level of engagement aren’t interested in being challenged. They care more about achievement (rank and score) and continuing to dominate.
At the lowest level of engagement, both achievement and challenge had a modest effect on continued play.
At the middle level, which was the majority, players responded strongly to both achievement and challenge.
“That was the surprise and it was hard to articulate before we saw the data,” says Manchanda. “They want to play to better themselves, not just to score a higher rank.”
“This also highlights the importance of tracking the evolution of players’ engagement level, because the same game-play result may affect different players in different ways, depending on which level of engagement they are at,” says Huang.
Using that information, the authors developed an algorithm that matched players in real time based on their level of play and their likely motivators.
“The key here is the algorithm must be scalable and fast, as decisions need to be made in seconds, i.e., real time,” says Jasin.
Their algorithm improved customer retention by 4 percent to 7.8 percent compared with the random matching that’s standard in the industry.
“The improvement is even more significant for players who have been playing the game for a while,” says Jasin. “This is particularly important because those are the people who are hard to retain.”
Switching from random matching to the algorithm would take a company some time and effort to set up. It has to match players correctly within seconds to be effective. But the increased customer retention should be worth it, the professors say.
Media Contact: email@example.com