Stics recently introduced a revolutionary new predictive analytics model called the Behavioral Model. It is so named because it sees into gamblers’ behavior to predict whether they will play until they run out of time, run out of money, or run out of free play before they leave.
The algorithm Stics invented for the Behavioral Model utilizes information from a huge population of gamblers to identify the expected behaviors each player exhibits while gaming. The algorithm then categorizes each player as time-limited, budget-limited, or free-play limited so effective offers can be delivered to maximize the value of each player.
Time-limited players, for instance, should only be given earned free play. When time is limited, they spend incentives first. In these cases, each dollar of free play displaces a dollar out-of-wallet before the player must leave. Players that are time-limited should only be incented with conditional offers. If not, they will not play with their own money, but just play with offer dollars.
The best players are budget-limited players—those that play until they’ve exhausted their available wallet. They should be incented with offers at a value that will keep them from playing at your competitors. Higher incentive levels can be offered to them with less risk, because they will still tend to spend all of their budget.
The Behavioral Model is offered as an option in the Standard SticsPredicts package, but can be purchased separately. SticsPredicts is a group of models which, when used together, will give a casino a more complete picture of their customers. The standard SticsPredicts package includes a Gamblers Worth Model, a Gambler Frequency Model, and a Gambler Response Model.
“On its own, the Behavioral Model is something the industry can use and make money with immediately,” says Christy Joiner-Congleton, president and CEO of Stics. “Using all the models together gives a new definition to customer insight. As a predictive analyst myself, it is easy to see that this model is unique. Observed industry generalizations ‘fall out of’ or become special cases of this complex model. So, the analytics may be gnarly, but the recommendations smack of good sense.”
For more information, visit the company’s website at www.stics.com.
