Findings from the Responsible Gambling Summit.

Having attended the Responsible Gambling Summit in London, the conversations with the Gambling Commission, William Hill, and Sky Bet struck a chord and got the Qubit team thinking about personalizations we could build using readily available data.


How can personalization technology change the way brands react to customers whose behavior becomes abnormal, and make sure they are using services safely? Not only that; how can you do this in real-time and at web scale?

As Rob Painter from Sky Betting & Gaming put it: ‘historically people in the industry have seen safer gambling as an area of eroding revenue, but if you are a trusted brand you will earn new customers. In addition, by focusing on Responsible Gambling you become a more sustainable business as you don’t want to be generating revenue from problem gamblers.’

Benchmarking and identification of ‘problem gamblers’ (PGs) is now commonplace. Typically each customer has a ‘Responsible Gambling Score’ (RG Score), that’s created by a number of weighted criteria and behaviors they display. Depending on their ‘normal’ behavior, scores will change based on behaviors that differ from ‘normal’.

For example, a customer who usually deposits £100 each month suddenly deposits £100 each day over the course of a week. This is a key identifier of a change in behavior and, potentially, a problem. Consequently their ‘RG Score’ increases. When their RG Score exceeds the threshold set by the operator, that individual enters the PGs customer segment.

The operator will generate a report of those that have surpassed the threshold at the end of each day and a team will investigate further.

This is nothing new.

However, there are limitations. Limitations primarily to do with how fast operators can react. If you’re generating a report at the end of the day and then handing that to a specialist team, you could be losing precious time to influence customers decision-making that could be in-session.  

We want to fix this, and it’s possible now.

With Qubit, when a customer surpasses the PG threshold, we can react in real-time and start deploying personalizations to ensure the individual is safe. Of course, there will be different levels of severity, with messages and experiences that vary accordingly.

Here are some ideas for responsible gambling which are possible now:

1 - Understand more about Problem Gamblers

Target surveys to customers with high PG scores to understand how you can help. This will provide valuable qualitative data, that you can match with existing quantitative data to get a rounded picture of what’s going on.

With Qubit, you can then segment further based on the responses and tailor experiences based on how mild or severe the indicators are.

2 - Use personalized awareness messaging

Notify customers in-session if they exhibit behavior significantly different from their norm. Using social proof, you can count an individual's number of actions on site (like the number of deposits they make) and compare this to their average.

Customers can sometimes lose sight of the actions they are taking; reminders will encourage them to reflect and control their behavior.

Messages like ‘We hope you’re still having fun, just to let you know you’ve made 15 deposits today’, could be served to that effect.

 3 - Prominently signpost tools available

Personalize the navigation and content, like homepage banners, so PGs can easily access the online tools available to help them control their behavior.

Deposit limits are a good tactic to employ when thinking about responsible ways to engage customers.

 4 - Customer care support

Human interaction is one of the most successful treatments. GamCare has reported that they receive around 30,000 calls on their gambling helpline, (Source: GamCare).

Deploy a contact option, only for customers with a high PG score and exhibiting real-time PG behavior, so they can speak to a care agent immediately.  


The important aspect of all of this is that it can be achieved within a session - while the customer is on-site. As soon as the first signals are identified, operators should be using technology to deploy messaging that ensures a session and the individual is safe. A 24-hour wait before you can take action is too long. You need to be activating your data at a speed which creates more trust with your customers.

For Qubit this means working with operators to take advantage of their existing RG models and supercharge reaction times with Qubit’s real-time data. Put simply, this is how it could work.

  1. Use Imports to upload RG Scores into the Qubit platform,

  2. Qubit acts as the listening engine, the layer on site that picks up on in-session PG signals,  

  3. By combining imported (i.e. RG score) and in session data, customers can be segmented and see the experiences based on their level of risk.



Jason Fernandes

Read More

Subscribe to stay up to date.

Receive the personalization newsletter directly to your inbox.