Strict Standards: Declaration of JParameter::loadSetupFile() should be compatible with JRegistry::loadSetupFile() in /home/content/94/5783694/html/intelligencia.co/libraries/joomla/html/parameter.php on line 512

Warning: Cannot modify header information - headers already sent by (output started at /home/content/94/5783694/html/intelligencia.co/libraries/joomla/html/parameter.php:512) in /home/content/94/5783694/html/intelligencia.co/plugins/system/jat3/jat3/core/parameter.php on line 107
Casino & Hospitality

My Account

Open/Close

Thanks For Your Feedback!

Email
Subject
Message
Five plus seven is? (answer as number)
Open/Close

Facebook Comments

Click to open/close fB Comments
RSS

Intelligencia.co

Casino & Hospitality

Intelligencia executives have been involved in the casino and hospitality for the past five years. With staff based in Hong Kong, Macau, Manila and Australia, Intelligencia's consultants have been intimately involved with projects at casinos and sports books throughout the region.

Intelligencia has been involved with BI, CI, marketing automation and analytics implementation at several casinos in Macau as well as offsite with clients in both Australia and the United States. Intelligencia's executives are some of the most knowledgeable consultants in the field of data mining, analytics and business intelligence.

These are the questions casino executives need to have answered when it comes to predictive analytics in the gaming industry:

  • How much is a patron worth, how much can we expect a patron to lose in the future, and who are the most valuable patrons?
  • What patrons come together?
  • What patrons are most likely to abuse an offer?
  • What patrons are most and least likely to respond to an offer?
  • What offers perform the best?

Once patron worth has been defined, the business can then use data mining and modeling to estimate predicted worth into the future. Simple metrics based on historical behavior, such as Average Daily Theoretical Loss or Average Trip Theoretical Loss, will produce fairly accurate predictions of future worth. However, advanced predictive models are able to predict worth with more accuracy and power by accounting for both patterns in behavior over time and relationships between predictive inputs that exist within casino data. There are a variety of techniques that are used to develop models to predict future worth, the most common being regression models.

Multiple regression models are the most commonly used for this because they utilize a variety of predictors and the relationships between those predictors to predict future worth. For example, a model built to predict future gaming trip worth might be generated based on historical information about theoretical win, actual win, credit line, time on device, nights stayed, and average bet. Regression models can also be built using such categorical variables predictors as gender, ethnicity, age range, or other demographic variables. Developing separate models based on categorical variables, such as separate models predicting worth for slot and table players might produce models with less error and better predictions. Regression models are particularly effective because the model can be used to score historical data to predict an unknown outcome, which is worth in this case, within a certain degree of confidence.

Identifying most valuable patrons

In addition to developing models to predict future worth, there are other analytical methods to determine a patron’s value to the business. One way to identify the best patrons is to try and separate the skilled gamblers from the unskilled. It is possible to look at whether a patron is usually a loser or winner. A quick and easy way to evaluate a player’s skill is by calculating the percentage of trips where the player actually lost money.

Although slot machines are not really skill based, we can still differentiate between patrons by looking at the strategies and behaviors of slot players. One quick and easy way to separate slot patrons is compare how much play they have on participation machines relative to owned machines. Since casinos have to pay a certain percentage of win or handle to the slot manufacturer for participation games, patrons that primarily play non-participation games are slightly more valuable to the casino. A slightly more complex metric for slot players is to look at their average bet relative to the maximum bet on the games they play. Usually, the maximum bet has to be played in order to be eligible for jackpots and progressives. Given two patrons of similar theoretical worth, the one that plays closer to the maximum allowed bet is more likely to hit a jackpot than the one who doesn’t. Usually the patron with the higher average bet would seem to be more valuable, but since the lower bet patron is less likely to hit a jackpot, the lower bet patron might be a lower risk. This metric could be useful on its own, or could be used as either a predictor in a model for future worth or a decision tree predicting whether a patron will respond. These are just a few examples of how data mining, along with predictive modeling, can provide useful information to differentiate between players that might otherwise seem very similar.

Identifying patrons at risk of abuse

Predictive models of worth will likely take into account the factors that predict whether a guest will play on a future trip. It is also a good idea to build a separate model to identify patrons that are likely to use a future offer and not play at all. Decision trees and logistic regression are common statistical methods used to identify patron characteristics that predict the likelihood of a patron (or segment of patrons) to abuse an offer.

Some factors that are likely predictors of abuse are age (younger patrons are more likely to abuse), gender, and history of abuse. Additionally, survey data (e.g., from follow-up surveys after a patron’s visit) that is linked to individual patrons can be used to identify other predictors. If a patron thought they were treated unfairly and had a bad experience in the past, they might take an offer for a free room and/or free play as revenge for that bad experience. When they come in they take all the perks available from the offer then go elsewhere to gamble and spend money so they can later get a better offer from the other casino. By identifying the patrons at risk of abusing offers, the business can decide how to market to those risky patrons. For instance, someone might reach out to the patron to try and rectify the situation if they had a bad experience. Instead of sending them the general offer for free play, they would be sent an offer that requires them to play to a certain level or they won’t get the free play.

Optimizing offers

In addition to predicting the future worth of patrons, it is important to know which marketing campaigns are the most effective for driving response, revenue, and profit. In general, certain offers are better than others, and specifically certain offers will be better for certain patrons. The common components of marketing involve offers for rooms, restaurants, retail, and gaming (i.e., free play). While knowing the probable future worth of a patron is critical for determining the reinvestment level for which a patron is eligible, patrons’ behaviors and interests can be used to identify the offer(s) that will be most appealing to each patron and generate the most profitable response.

A/B Testing

The most basic way to identify the best offer is through A/B testing. A/B testing involves testing two different offers against one another in order to identify the offer that drives the highest response and the most revenue/profit. More advanced statistical methods can be used to generate likelihood of response scores and classification scores. Some of the more common statistical approaches are logistic regression, decision trees, and discriminant analysis.

Essentially, these statistical methods use historical data to find the factors that are related to whether a patron responds. Those factors can then be used to assess the likelihood of response based on the similarity of a patron profile to that of responders. These methods have historically been used in direct marketing analysis to identify the best types of offers and the most likely responders. In order to build accurate and predictive response models, historical data about response is required. The likelihood of response might be a broad measure of response that refers to the likelihood a patron will respond to any offer, or it might be specific to the likelihood of response to a specific type of offer. Additionally, it’s a good idea to select test segments of patrons for the purpose of continually testing new offers. Doing so will help to ensure that there is a large amount of response data that can be used to build models and continually improve the efficacy of marketing. Effective response models will help identify which patrons are most likely to respond to an offer, and in turn to which offer patrons are most likely to respond. There are at least three main uses of response modeling that can improve marketing results:

  1. Identify the likelihood of patrons to respond to the offer
  2. Identify the offer(s) to which patrons are most likely to respond
  3. Predict when a patron is likely to return

Identify likelihood of response

In addition to developing models to predict future worth, there are other analytical methods to determine a patron’s value to the business. One way to identify the best patrons is to try and separate the skilled gamblers from the unskilled. It is possible to look at whether a patron is usually a loser or winner. A quick and easy way to evaluate a player’s skill is by calculating the percentage of trips where the player actually lost money.

Although slot machines are not really skill based, we can still differentiate between patrons by looking at the strategies and behaviors of slot players. One quick and easy way to separate slot patrons is compare how much play they have on participation machines relative to owned machines. Since casinos have to pay a certain percentage of win or handle to the slot manufacturer for participation games, patrons that primarily play non-participation games are slightly more valuable to the casino. A slightly more complex metric for slot players is to look at their average bet relative to the maximum bet on the games they play. Usually, the maximum bet has to be played in order to be eligible for jackpots and progressives. Given two patrons of similar theoretical worth, the one that plays closer to the maximum allowed bet is more likely to hit a jackpot than the one who doesn’t. Usually the patron with the higher average bet would seem to be more valuable, but since the lower bet patron is less likely to hit a jackpot, the lower bet patron might be a lower risk. This metric could be useful on its own, or could be used as either a predictor in a model for future worth or a decision tree predicting whether a patron will respond. These are just a few examples of how data mining, along with predictive modeling, can provide useful information to differentiate between players that might otherwise seem very similar.

Identify likelihood of return

In addition to having some information that helps determine to which offers a patron is most likely to respond, it would be nice to know exactly when a patron was planning on making their next trip. Although we might not know exactly when a patron is likely to return, fortunately we can make a pretty good prediction about it. There are a variety of methods that range in complexity that can be used to assess when a patron will return, including frequency analysis, regression, and survival analysis. Knowing when a patron is likely to return is beneficial as it helps to identify patrons that haven’t made a trip in the expected amount of time and are at risk of leaving. First, the business needs to have an idea of the average or median time between trips. This might need to be segmented based on geography, worth, or even historical frequency. Patrons that have not made a trip within the decided amount of time for their segment are subsequently flagged and dealt with appropriately.  

Historical data can help identify segments of patrons that are expected to make trips weekly, monthly, quarterly, annually, bi-annually, and so forth. Marketing can integrate information from predicted worth, optimal offers, and time to next trip to maximize campaign success in a number of ways. The business can save money by adjusting the frequency of offers for patrons that are not identified as likely to come back for longer periods of time. Instead of sending the patron monthly offers, they can sent quarterly offers with longer valid windows that allow more time to book. Or, for example, if the patron only comes annually around his/her birthday, we might only send an offer annually around the patron’s birthday. Conversely, campaigns might be created with the goal of increasing the frequency of visits from higher worth patrons. Casino marketing should have the goal of generating trips sooner than expected and converting patrons into more frequent visitors. Additionally, time to next trip analysis can be used to identify when it has been too long and the business is at risk of losing the patron. In this case it might be useful to send an offer using “last chance” or “we miss you” messaging. The offer might also need to be slightly better than what the guest has received in the past. By knowing when a patron is likely to return, casinos can adjust marketing strategies appropriately in order to save money on mail costs, retain guests, and increase loyalty.

Table Games Revenue Management

Because table games take such a prominent role in a casino's bottom line throughout the ASEAN region, a Table Games Revenue Management (TGRM) system can be a great revenue driver, as well as a powerful analytical tool that helps with optimization throughout a casino's property. The question of when to open a table, at what minimum price, and in what section of a casino can have a huge affect on the company's bottom line. Many costs, such as the labor required in dealers and pit managers, are somewhat fixed, while others variables, such as hand speed, depend on how many seats are filled at the table, but these costs can be modeled to create not just a table minimums data set, but also a labor schedule to ensure the right number of tables are open at the right time, at the right price, in the right area, with the right staffing levels. 

As business levels increase, casino shift managers currently examine the number of players at a table and determine if opening additional games would be prudent. As games open, and fill up, the determination is then made if raising minimum bet levels is needed. This works at a reactive level, but is far from optimized. This potential loss of profits could be quite substantial over time. 

By predicting the next time period’s expected demand in terms of head count, then applying the percentage of players at each average bet level, the number of players expected at each betting level can be predicted. After optimizing and considering overall house advantage, profit can be maximize for the next time period in a proactive manner.

Intelligencia is also working with its vendor partners to add additional variables, such as geolocation (to understand on-property demand) or social media data (to understand approaching demand (pun intended)) to better compile the demand side of the equation, which, we believe, is currently ignored because, once seven seats at the table are filled, demand constraints aren't taken into account. 

Check with us to learn more about our TGRM models to see how they can increase optimization on your casino floor. 

clients

                                                 

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

Save

&