iGB Q&A: Duncan Ross, Director of Data Science, Teradata

25 February 2014

Competition within the online gaming industry is intensifying and companies need to find ways to get ahead and maintain a leading position. In our digital and online connected world, where information can be tracked and manipulated to the nth degree, data is more valuable to companies whatever the sector, than ever before. And the use of Big Data and analytics is increasingly being seen as the answer.

Look outside igaming to sectors such as travel and leisure, entertainment, e-commerce and even traditional land-based gambling, and hundreds of large organisations are using Big Data in a huge variety of ways to learn more about their customers and customers’ behaviours and, in turn, acquire new and retain existing customers, drive efficiency, and boost the bottom line.

Ahead of his appearance on a webcast in association with iGaming Business entitled ‘Big data, big decisions: why analytics is the future of the gaming industry’, iGaming Business spoke to Duncan Ross, Director of data science at Teradata, about why gaming companies are increasingly looking to Big Data as a means of gaining a competitive edge.

iGaming Business: Tell us more about your role at Teradata and what it involves.

Duncan Ross: My main role is to understand the potential value from data using analytical techniques and sophisticated mathematical approaches, but with a business focus at heart. I talk to business people to find out what the pressing issues are and how data within organisations can be grouped and classified in order to understand behaviours and, crucially, predict behaviours. It’s all about finding useful ways to try to change business approaches and change behaviour.

The first thing is to find out what your key business problem is. For a bank it might be understanding risk, for a telecoms or gaming business it is likely to be about maximising the Average Revenue Per User (ARPU). It is about being able to predict what individuals people might do that can give you a key advantage on the competition.

iGB: What is the main business issue people ask you to examine?

DR: It is fundamentally about customers, improving their experience and, perhaps more importantly, influencing their behaviour so they become more loyal, remain on websites or in physical locations for longer and essentially spend more with you rather than other brands.

Customers are customers whether it is across retail, travel, or banking but there are some key things that affect the gaming industry in particular. For example, platform behaviour. How should gaming companies best approach and deal with customers when they are able to play on multiple platforms? How do you best link those together and understand their full pattern of behaviour?

iGB: How has the use of analytics changed in recent times?

DR: There are valuable lessons to be learnt from other industries such as retail where the only mechanisms they had to look at this previously were customer surveys. Big Data has changed the ways we approach this today. More recently customer loyalty cards have been used to monitor behaviour and act accordingly depending on that behaviour.

In the past we only knew what people did at the checkout and even then we didn’t know who they were, what their buying patterns were at particular times of the day, in particular seasons, around particular events and so on. Today, using existing technologies, we are able to pinpoint exactly where they’ve been in store and exactly what they’ve been looking at, when, how long for and whether they have purchased those goods as a result of a particular promotion, or physical store location, for instance.

We now have incredibly rich data, location in particular, as a result of our constant and increasing use of mobile phones and devices. Take this online and we are able to build an even more sophisticated picture. With access to the right data, and there is a huge amount available, every single action can be tracked and analysed. We can see exactly what offer or campaign a customer’s cursor hovers over, what browser and operating system they are using, how they got to the page – whether it is from a search engine, directly or via an affiliate site and then we link all these sessions and information together and understand groups of sessions, how much and how frequently they gamble, how and when they are likely to deposit, for what reasons, at what time of day, and as a result of what triggers.

By employing the right analytical tools and expertise gambling companies can find commonalties to what makes their customers bet or not bet. For example, we can monitor individual activity of someone actively looking within the football section of a site and explore the reasons why they remained on the site, looked at various offers but then did not place a bet. What are the common features of those people who don’t make the leap to depositing and transacting and what results can be achieved when comparing them to others that do?

It may be that the journey to deciding on what to bet on and making a bet is too complex, or it may be that our competitors were offering better odds and our customer was checking multiple operators. Using Big Data you can identify these areas and re-design your offering so that potential customers are more likely to turn into depositing, loyal players. The beauty of analytics and using it within an online industry such as gaming is that you can also run experiments to validate various theories and alter behaviours. Through those experiments you are then able to identify which ones are the most successful and quickly act to ensure they remain in place.


iGB: How do you view the gaming market from a data perspective?

DR: Gaming has done some great analytical work around fraud and it has taken time to get to that point. For example, in online poker where people can collude to gain an edge illegally, we have done a lot of work with various organisations to detect those patterns. The bigger prize, however, is customer understanding and being able to identify and address customer behaviours in order to extend life time and wallet share. The same kinds of techniques that have been used to combat fraud and collusion can be used for customer retention and wallet share. It’s about making that leap, identifying business problems and allowing your data to be used to improve results.

In the US several companies, including the land-based casino and entertainment groups, have taken active steps to better understand their customers both from a gaming and non-gaming point of view. They’ve asked themselves some fundamental questions in order to remain ahead of the game and the fierce competition for leisure dollars. The most common analytical questions are: how can we defend our position in a competitive marketplace against gaming brands and other competitors; how can we better retain customers; what elements are successful and not successful and crucially, what can we do better to stop people leaving our properties.

iGB: What best practice examples of using Big Data could lend themselves to the gaming industry?

DR: Customer life-stage analysis – companies can build up patterns of understanding on the ways typical customers, down to very specific segments, use certain products, how this develops and then stops using that or those products. In spotting the triggers between these stages you can increase the length of time individual customers stay online, increase their value and minimise churn.

Gaming companies need to look at their current practices, their pain points and what internal and external data they are able and willing to open up and that can be used to develop solutions. They need to look at areas such as how easy is it to participate online or on mobile, what goes from logging on to becoming regular player, and the path from acquisition to regular behaviour.

Personalisation is also a key area for the gaming industry to examine closely as is linking together behaviours between devices in order to offer a more successful multichannel experience.

iGB: Where and how should operators start when looking at using Big Data more effectively?

DR: They need to be able to accept that data can help and that it can be extremely valuable, then start from a business problem. A lot of people try and start from an overly technological approach – the key is specifying a business problem and being as clear as possible. For example, customers ‘defecting’. This can mean very different things to different sectors. Ask yourself what you mean by someone churning. It is not as straightforward in gaming as it is in telecoms for instance, where customers are obliged to say they’re leaving to go to another network. In gaming someone might make bets on a fairly irregular basis but still be a customer, so at what point do you decide they have left? This is one of the key analytical questions you have to answer.

Duncan Ross leads Teradata’s International Data Science team and is responsible for developing analytical solutions across a number of industries. He is a regular speaker at major data conferences and has been named in Big Data Republic’s Big Data influential Tweeters.

Click here to register for the webinar and hear more from Duncan Ross and a host of big names form the gaming industry speak on a live Big Data webcast on Thursday April 3rd at 2pm GMT. Live questions will be taken during the webcast.