“From player development opportunities to slot floor operations, properly applied artificial Intelligence (AI) models can identify and predict trends, insights and anomalies and reveal previously hidden opportunities to drive incremental profit throughout a casino’s operations. AI-driven applications produce recommendations, measure outcomes, and learn from that process to improve future recommendations.
Why do I need AI?
We are in a transformative time in the gaming industry, with new systems and sources of data coming online now, more than ever, AI matters.
Today’s Business Intelligence (BI) landscape is completely changing. Operators need to seek solutions that are designed from the ground up to do more than just provide them with reporting. The reality is operators don’t need someone to tell them their worst performing game. They already know that. Operators should seek a scalable platform that uses all the property’s data, combined with new data sets and third-party data to identify underlying trends, and provide recommendations before performance trends make it clear, which games or players are showing signs of becoming underperformers, or which show potential to become high performers with attention.
One of the biggest challenges for any operator who wants to install a new technology is the answering the question your boss always asks, “what is the ROI?” check out these case studies:
Case Study 1 – The Super Spender Model
This AI application first identifies a population using AI models to (1) predict player response to offers from player spending and behavior and (2) identifying low frequency – but high worth characteristics. These are combined in another multi-layered AI “ensemble” algorithm. The test list has a high confidence (85%+) that the recommended Free Slot Play offers will yield the anticipated revenue. So, offers can be targeted very efficiently.
The model focuses on the 20% of the top players with the highest confidence levels.
- Property A (Mid-Size Casino/Hotel in a High-Frequency Locals Market):
- The test group demonstrated a 2% average lift average in play across the after vs. before comparison periods.
- The control group experienced a 12.3% average lift in play across the after vs. before comparison periods.
- The property (using only players who had play in both periods) experienced a 12.45% average lift in play across the after vs. before comparison periods.
- Property B (Regional Casino in a highly competitive market):
- The test group demonstrated a 6% average lift average in play across the after vs. before comparison periods.
- The control group experienced a 13.3% average lift in play across the after vs. before comparison periods.
- The property (using only players who had play in both periods) experienced a 1.9% average lift in play across the after vs. before comparison periods.
Case Study 2 – High-Potential New Cardholders
Using a combination of play and demographic data, an AI predictive algorithm identifies new cardholders with high potential. A quick and individualized recommendation is generated for Player Development teams to act on.
- The retention rate increased 20% for cardholders identified and given offers in a test group, versus a control group not given an offer within 10 days of enrollment.
- Players targeted within 10 days saw a daily average theoretical profit increase of 54% compared to those in our control group.
- The recommendation also provided the property with a way to quickly develop relationships with new high potential guests.
In both case studies the test and control groups become highly valuable inputs to a new iteration of the AI models to further increase the accuracy of the predictions. By retraining the models on the results, the next set of predictions will be even better, and more profitable.
The future is here, and it includes AI. Properties can no longer afford to not embrace a technology that creates increased play and adds money to the bottom-line.”
