Analyzing a Sports Book for Customers

By Andrew Pearson October 20, 2015

Many sports books are technologically unsavvy. With thin margins and difficult customer acquisition, sports books don’t have enough money for expensive IT solutions. Or so they think. The reality is a sports book only makes money when it completes a bet and SAP HANA allows sports book to accept many more bets than a normal database system. In the highly competitive field of sports betting, sports books must fully understand their customer risk profile and BetWatch™ allows just that. Sports books also need to be customer-centric and a solution like BetWatch™ gives a sports book’s CRM department to really understand their customer (down to an individual bet ROI level) so that they can market to the customer in a way that gets traction.

Dr. Don Forbes and Qualex Asia MD Andrew Pearson came up with the initial idea in the summer of 2014 and it was presented to the SAP Startup Focus Group in June 2014. It was quickly approved and development began immediately.

BetWatch is an analytics solution developed for the sports betting industry. BetWatchallows sports books (SB) to understand a bettor’s gambling habits and, more importantly, how their bets affect the sports book’s fiscal bottom line; this data reduction exercise provides a succinct and quick view of the customer base that can then be used to make risk based decisions, complement multiple marketing strategies, as well as, potentially, spot problem and/or addictive gamblers. The approach of our solution is to highlight the bridge between using analytics to derive knowledge from the information available that goes beyond merely slicing and dicing the data. SAP HANA Predictive Analytics Library (PAL) was used to create the necessary segments via the k-means algorithm. Once the segments are created, HANA is used to keep track of betting activity in real-time using the k-nearest neighbor algorithm.

Using this segmentation model, BetWatch™ tracks individual bets and alerts an end user in the sports book’s trading department when a segment bets outside the SB’s “comfort zone”. A segment that doesn’t consistently beat the book can exceed the trigger threshold without triggering an alert; it is only when a customer from a segment that has been marked as “watch” wagers an amount large enough to exceed the SB’s threshold level that alerts will be triggered. This way, the SB can know in real time what the profile of their risk and reward is for every event it is laying bets on. The SB will also have the ability to understand its customer base more deeply as well as make strategic decisions around risk, marketing, and customer management as a result of these segmentation models.

The BetWatch™ solution can assist in the following areas:

Nefarious individuals and criminal elements always have the first mover advantage, while fraud and AML programs are encumbered by resource limitations, low levels of awareness at the executive level, persistent silos, high prosecution thresholds and outdated strategies. A new approach is necessary and BetWatch™ can be one part of a sports book’s tools to root out these corrupt practices.

The key to understanding these groups is a recognition that pursuing one alert at a time; setting obvious thresholds; favoring siloed “point solutions” for channels and products and overlooking the links to the broader network will permit the criminal enterprises to commit fraud over long periods of time and inflict huge losses despite the detection of individual nodes, which are expendable. To effectively address this new adversary we must “out network the networks” by breaking down individual and industry silos; employing better technology tools than those utilized by the bad guys such as real time detection, powerful data analytics and, above all, link seemingly disparate criminal acts that individually fall under our thresholds but in the aggregate total up to multi-million dollar losses.

The second element of BetWatch is the Customer Experience Management (CEM) component that allows a SB to gain not only a 360 degree view of its customers, but also a deep understanding of the customer’s spending habits as well as the SB’s marketing activity and spend to attract and keep its customers.

Having received certification in late September 2015, BetWatch™ has only been in the market for about 4 weeks and we are only now talking to customers and potential customers about it. We have spoken to one client who has expressed interest in adding certain components to the product that could help in spotting problem gamblers as well as in spotting fraudulent activity. We have also received interest from VC firms as well as other companies that see our offering as a complementary service to what they offer, but they see the value of the product as is rather than it requiring any changes. We have plans for a phase II of the solution and this will tap into the SAP hybris marketing suite as well as the CEM module and we are discussing this with one client in the ASEAN region as well.

Each and every day sports books are faced with the daunting task of creating and listing odds for the hundreds of events that they are willing to take action on. Because gamblers can be a very sophisticated breed, they can quickly figure out if a sports books odds don’t accurately reflect the true value of the bet. When gamblers spot a price that they feel is out of line, they can hammer it and spread the word to their friends and/or syndicate members. BetWatch™ segments a sports book’s customer base so that it can not only spot the gamblers who consistently beat them, but also keep tabs on them and offer the sports book a way to manage the risk associated with these potentially winning gamblers.

Although we have yet to deploy the POC, in terms of long term integration with our solution and our customer, there is potential. We have spoken to several potential clients who are looking for a “World Class CRM” system and we believe BetWatch™ could be part of that solution.

The solution we developed requires real time capability and an in-memory system like HANA was needed to give the sports book the ability to see each and every bet that is coming in and alert the risk management department about bets that might hurt their bottom line. Our solution employed PAL and Infinite Insights within HANA’s libraries.

The sports betting industry is not a very sophisticated industry so is ripe for a solution such as ours. The visibility that BetWatch™ gives on the sports book’s clientele is unrivaled. Not only do sports book get to see things on a micro level, i.e., each bets ROI, but an overall pattern of what types of bets the customer prefers to make, as well as which ones he typically wins on. Sports books can use the information gleaned from BetWatch™ to make risk management and customer intelligence decisions that few solutions can match. BetWatch™’s ability to segment customers into winning and losing segments also helps a sports book keep a close eye on the bottom line.

Because the sports betting industry is so competitive, sports book will need to get a deeper understanding of their customers. BetWatch™ allows a sports book to really know its customers, not only in the typical sense of drill-down dashboards that reveal a customer’s betting behavior, but also predictive abilities that group customers into segments that might reveal both micro and macro behavior.

BetWatch™ can help a sports book understand what the predicted final tally of bets will be on a particular sporting events.

In-memory technology allows sports book to take considerably more bets than standard IT technology. The type of bet offerings should also increase exponentially and this should allow a sports book to make offers that its customers will find novel and, hopefully, hard to resist.

In terms of advice for other startups, we would say give yourself some time and reach out to SAP for help, if needed. We had specific integration problems with SAP HANA and Lumira this necessitated the use of SAS VA as the dashboard front end. Because we had experience with this solution, we were able to cut down development time, so take what you know, integrate it with what you’re learning and get the product out the door. Aim to shoot for the stars, but when reality calls, turn the gun down and shoot the fish in the barrel.


Originally published at on October 20, 2015.




Andrew Pearson is the MD of Intelligencia, a software consulting company. Speaker, author, columnist, Pearson writes about IT issues like AI, CI, and analytics.

Love podcasts or audiobooks? Learn on the go with our new app.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Andrew W. Pearson

Andrew W. Pearson

Andrew Pearson is the MD of Intelligencia, a software consulting company. Speaker, author, columnist, Pearson writes about IT issues like AI, CI, and analytics.

More from Medium

What is Data Lake?

8 Women in Data & Insight to Inspire You in 2022

Data Travel from Raw to Wisdom

Data Triggers