MobilityWare Uses CDP and Predictive Scoring to Learn What Makes Some Games So Irresistible, Play After Play?

January 27, 2020

What makes a game stay appealing, play after play after play? That’s what mobile games purveyor MobilityWare wants to know — and it’s using a Customer Data Platform (CDP) and predictive analytics to help figure that out.

If you’ve ever played games on your phone, chances are you’ve played a MobilityWare game — either its flagship Solitaire or any of the many other games the company offers, such as BlackJack, Spider, or Jigsaw Puzzle. Founded in the 1990s, the company’s continued success and platform transitions are due in part to its deep understanding of its players and their customer journeys. And lately, MobilityWare’s creative use of customer insights is getting an assist from Arm Treasure Data enterprise CDP.

How to Improve Customer Lifetime Value? Keep Players Playing

One of the toughest things for a gaming company to do is keep its best customers playing, but that’s pivotal in increasing customer lifetime value (CLV). And the answer is usually highly dependent on the game, the demographics of the players, and even individual customer journeys and player histories. Teasing out player incentives or other customer retention strategies requires careful customer data analysis, the kind that CDPs can quickly and automatically perform.

The other problem games sellers face is how to monetize their games, and a major concern with any contemplated change is not cannibalizing revenues from current play. MobilityWare makes money from in-app purchases and ads, plus rewarded video.

Predictive Scoring & Modeling Help Dissect the Churn Problem

The challenge that the MobilityWare team faced was this: How do you figure out when a customer is likely to quit playing? And which player incentives could delay or stop customer churn?

Chris Densmore, director of analytics for MobilityWare, set out to get answers. He focused primarily on “dedicated players,” who had installed the game more than two weeks ago, but who had not played in the previous two weeks.

“Dedicated players can be incentivized, they’re more valuable, and we have a larger behavioral dataset because they’re more likely to have played a lot in the past,” Densmore says.

“Also, we can do more to influence the outcome, which is important,” he adds.

Densmore used MobilityWare data and analytic features from the company’s Arm Treasure Data CDP. Then he chose a logistic regression model, because it can still produce valid results even if some variables are correlated — as they usually are in real, live users — and because logistical regression makes it possible to interpret contributing factors as well. Another bonus: The coefficients that come out of regression can be easily placed in SQL scripts to produce predictions.

Customer Data Insight #1: Coins Don’t Work Like Boosters

Densmore’s work got some surprising results. Rewarding people with a coin that can be redeemed within the game wasn’t the clear winner. It did reduce churn, but had a negative impact on monetization. People tended to hoard them for unspecified “later use,” rather than play more and use them in the game. And who wants to take steps that hurt revenue if they don’t have to?

Customer Data Insight #2: Gamers Want Boosters, Not Coins, and They Pay More If You Help Them

The real surprise was that giving someone a “booster,” or a small assist or tool they can use to win, was almost as effective as coins in combating churn, PLUS it increased ARPU (average revenue per user) by more than 450 percent for those with a probability of churning between 60 and 80 percent.

A little bit of help — but not so much that it devalued the accomplishment of winners — was the key to reducing churn and renewing player interest in the game. And without the CDP data, it would have been tough to get the customer insights that pointed the way to the right customer experience. With it, hypothesis testing and predictive modeling pointed the way to higher profits.

Just think how many other product “churn” problems could be informed by similar analysis. Many other areas where fashion or trends are short-lived come to mind. For example, Shiseido uses a CDP to analyze loyalty data and combine it with other data feeds to predict when people might be open to new beauty products — or even a major makeover. Telecommunications companies, which live or die by the difference between new customer acquisition rate and churn, could also use CDPs plus predictive analytics to gain insight about what makes someone switch to a new plan or provider. The insights that CDPs can provide are only just beginning to be harnessed for better marketing and even improved product design. Predictive analytics are clearly more than a fad, and with the help of CDPs, they are fast becoming an easier-to-use tool in every marketer’s workshop.