SBTech leverages the transformative potential of machine learning

Betting Business SBTech
Digitain LB
Share this article
  • 1
  •  
  •  
  •  
  •  
  •  
  •  
  •  

With the release of Action Bet – its latest mobile betting feature – sports-betting solutions provider SBTech married rapid-fire in-play betting with machine-learning enhanced personalisation.

 

Ian Bradley, SB Tech’s head of product, explains how new technology can enhance player experience and boost revenues for operators, and unpacks the substance between the “personalisation” trend.

Whether browsing online shopping sites or skimming through social media feeds, today’s consumers are experience their digital worlds in an increasingly personalised way. According to Ian Bradley, head of product at sports- betting solutions supplier SB Tech, the release of the firm’s Action Bet feature is simply the logical next step in this process.

Utilising advanced machine-learning and machine-vision technology, Action Bet is the product of a sustained investment in research and development – and its immediate popularity among younger punters has been evidenced in market research carried out by the supplier.

The mobile-betting feature uses a machine-learning algorithm to create rapid-fire in-play markets based on incidents that occur during a football match, presenting these to the punter as a “bet now” call-to-action.

According to Bradley, this is just the beginning of SBTech’s investigations into machine learning and personalisation – with further announcements set to come in 2018. Here, he discusses the potential of these technologies to transform the customer experience, and how a clever application can boost both revenues and loyalty.

iGaming Times: Action Betting is one example of SB Tech’s ongoing investment in R&D – particularly around personalisation. Will this continue to be a focus going forward, and can we expect more machine-learning driven solutions from you in 2018?

Ian Bradley: Very much so. Personalisation is a major focus, not just for SBTech but for many e-commerce industries. One just has to go on Amazon or see the targeted ads that appear on screen when users have searched for a particular product to see that the whole online experience is geared towards answering or trying to predict customer habits and wants. From our point of view optimising our machine- learning solutions to produce highly customised betting markets for our operators and their players is a natural and logical conclusion to this process.

iGT: The kind of machine-vision technology needed for an algorithm to recognise match events is incredibly labour-intensive and complicated. How long did it take to perfect this, and do you eventually plan to apply this to sports other than football?

IB: We leverage the game-state information from our third-party partners, then use that as a trigger for the algorithm to push the market to the user. This is leveraging our complex event processing (CEP) infrastructure which we’ve been working on for the past six months. This underpins many of the new features and enhancements we have planned. We plan to deliver fast and iteratively, so to do that we have released the feature for football, we will analyse it, learn from it and adapt it so we can optimise the feature before rolling out across more sports.

iGT: ‘Personalisation’ has become a major buzz-word in the industry lately. What does this actually mean in terms of user experience, and what kind of user information does Action Betting work with?

IB: As mentioned, most e-commerce companies work with personalisation as a key aim: it is now part of the user experience and interface of the website; whether it is an online sportsbook or Amazon.

This enables the merchant to address their needs in a targeted way that individualises the site for them, reminding them of what they bet on or purchased and acting as a reminder of previous activity as they ease into their online session and look at upcoming opportunities. This fits in with the consumer feeling comfortable and familiar with what they are being presented.

Action Betting works by using the match stats and game-state information supplied by our data partners to offer in-play markets that are contextual and relevant to the action. If there is a corner, it can offer odds on the “corners market”, if there is a dangerous free-kick it can be over/under or goal-scorer markets, with the player taking the free-kick involved. Action Betting can then recycle that player’s information and suggest those markets to them when they next log on.


Share this article
  • 1
  •  
  •  
  •  
  •  
  •  
  •  
  •  
Digitain LB