The explosion of online games has resulted in the creation of a new industry: in-game currency, currently valued at $1 billion in the U.S. alone. But game developers, particularly startups that are rising and falling on a single game, are losing significant revenue as savvy social game players figure out how to “game” the system by stealing currency rather than buying it.
Normally, network security flaws enabling fraudulent gameplay are identified and solved on the backend, often by using thousands of servers at high capital and operational expense. This represents a significant cost to the developer, as IT security personnel will play back every transaction and analyze it in order to determine who the cheaters are and how they are manipulating the game.
There is a better way to ensure the viability of a new company with a product requiring in-game currency. Online game developers have vast amounts of rich, unstructured data at their fingertips that they could use to help achieve revenue stability. The data delivers an understanding of each player and what they are doing while in the game. That same data could be manipulated to identify and stop fraudulent game activities in real time.
Recently, we had the opportunity to test this premise with one of the world’s largest casual online gaming companies, which was plagued by revenue leakage due to cheating players. By using analytics to examine and model the game, we could determine common player navigational paths. We could then design mathematical algorithms that could place parameters around the average player and predict the average way he or she would progress in the game. This information determined the threshold range for what would be acceptable play; any players that fell outside of the threshold range would be flagged as being potentially fraudulent. The game developer could then choose to immediately freeze accounts for those advancing too quickly — thereby plugging the revenue hole.
Oddly enough, not every game developer chooses to ban a fraudulent player when they are identified. Sometimes, after running a basic cost-benefit analysis, the cost for the fix is higher than the loss of revenue attributed to cheating players. However, taking a broader view of the analytics, it may be possible for the fraud analysis to more than pay for itself.
For example, rather than using the extracted data merely to fix security breaches and stop cheating players, a developer might also use the data to provide marketing and sales insights. Slower game players and one-time users, quickly identified through the data analysis, might be pinpointed and targeted for promotions to entice faster and more frequent play — increasing company revenues. Other monetization strategies might also be developed through a closer examination of how players navigate the game.
For those interested in how data analytics is used specifically within the gaming genre, a good primer is now available on Amazon: Game Analytics: Maximizing the Value of Player Data by Magy Seif El-Nasr, Anders Crachen, and Alessandro Canossa.
Article by Ed Sarausad on VentureBeat/GamesBeatRead article