Business Analytics Blog

Business Analytics Blog

Opinions expressed on this blog reflect the writer’s views and not the position of the Capgemini Group

Mining Olympic Data Gold

Category : Sport

This is the third in our four-part Figure It Out series showcasing the potential of Advanced Business Analytics. The theme will be next summer’s much awaited London 2012 Olympics, and each week we will also feature a new cartoon focusing on a specific analytic technique.

The first two FiOs in this series, Olympic Tickets - Were you in the winning segment? and Olympic Tickets Bought - How Do I Get There? looked at the roles of Segmentation and Simulation in providing deep insight for organisations

This week we take a look at Data Mining and how it provides insight into the patterns that are hidden within data. We’ll examine how the London Organising Committee of the Olympic Games (LOCOG) could find out a bit more about the customer base and use this information to improve the chances of members of the public trying to get hold of tickets through the ‘Second Chance Sales’.

Thanks, once again, to our ‘Mistry’ Cartoonist we can show you “what the OR team can do for you” in data mining if ever you need a helping OR hand.

With a total of 8 million tickets being sold worldwide LOCOG must now also use the masses of data they collected during the ticketing application to ensure that the Olympic Games are a success from a commercial, logistics and PR point of view.

Data mining is related to Segmentation but uses the data to reveal insights about underlying patterns hidden in the data. Data mining can provide insight into the behaviour of those people who were unlucky the first time round, helping to predict what they might choose to apply for in the ‘Second Chance Sales’ Application. It can also identify patterns in successful ticket applications that might help manage Olympic Games’ logistics and revenue generation.

Relationships between ticket applications for events

As discussed in the first FiO in this Olympics themed series, the Olympics Ticketing Ballot has been the biggest ticketing exercise ever undertaken in the UK, but with 1.2 million people left empty handed and disappointed, the approach has been criticised for being unfair. LOCOG are now under increasing pressure to demonstrate the fairness of the sale of the remaining 2.3 million tickets.

Second Chance Sales open at 6am on 24th June so it is too late to improve the system any more, but hopefully LOCOG have already applied some analysis to address the situation. Looking for relationships between the events unlucky applicants applied for and other available events is one way to help applicants who were unlucky in the first ballot to succeed in the Second Chance Sales. By identifying related events that have a common market LOCOG can focus their efforts to promote the remaining tickets to the right people. If done successfully, this would ensure as many people as possible got tickets for the Games and that LOCOG are not left with any unsold tickets.

A recent Guardian Article revealed that there were 5 million ticket applications for the athletics sessions and 1.3 million for the 100m final alone. Applicants for athletics (or any other event) can be correlated with other applications to discover what events that people who applied for this sport also applied for. Additional clustering by age, geographic location could also be done to provide a detailed profile of the unsuccessful applicants. By identifying the events that applicants in each cluster are likely to be interested in LOCOG can then focus their marketing of these events on these clusters.

Clustering ticket holders

Understanding the different types of people that bought tickets for each event makes organisers better placed to meet their requirements. By knowing more about who will be going to each event it is possible to better plan logistics for the event (e.g. transport, facilities) and generate improved revenues from people attending the event.

Taking transport for example, each day of the Olympics hundreds of thousands of people will be using public transport to get to the Olympic Park and other Olympic venues. The data map below, produced by Tube Viz, shows the footfall at each London Underground station. There are currently around 2,000 people in the station during peak periods, not much compared to central London stations, but it is easy to see this map looking much different during the Olympics.

London 2012 has proposed a 45% increase in capacity on London Underground's Jubilee line, and a transport hub in Stratford. This will be designed to carry 320,000 people per hour, but how do we know this is enough?

Clustering the volume of people travelling to each event can forecast the stress hot-spots in the public transport network. By clustering people into groups by the events they are attending, the time of those events, the geographical starting point for travellers and other characteristics it is possible to implement capacity planning to best meet each group’s requirements.

Classifying the market

Lloyds TSB has estimated the London Olympics will generate £10bn in revenue for the British economy. There are various estimates for how this will be achieved; figures quoted include £1 bn for Olympic merchandise (key rings, soft toys etc.), £0.5 bn ticket sales, and £1.5 bn tourism. However, not all people will buy the same products and services. Revenue generation will depend upon understanding the market and the different classes of customer in the market.

Tourists going to the Olympic Stadium may be much more interested in buying London Olympics 2012 merchandise (such as T-shirts) while locals going to a rowing event might be more interested in going out locally for a meal and a drink after the event. At the Olympics, people attending events are likely to fall into a number of different spending classes and it is vital for organisers understand these and take them into account during planning.

Classification uses patterns in data to group people together into classes that can help focus marketing and sales. Using the characteristics of each ticket holder, such as the price of their ticket, and the event type, it is possible for organiser to classify the market at each event ensuring that the right products and services are available.

About the author

Nigel Lewis
Nigel Lewis
Nigel leads the Capgemini Consulting’s 35 strong Business Analytics team, which delivers analytical, operational and strategic modelling solutions to clients. He has 18 years consultancy experience as well as 8 years experience in the UK gas industry. Nigel has successfully managed complex projects in both the public and private sector, including capacity modelling, simulating supply chain operations, strategic business modelling to support future policy decisions, and implementing complex demand forecasting systems. Nigel is currently focussing on the development of Capgemini’s customer analytics and analytics advisory services.

Leave a comment

Your email address will not be published. Required fields are marked *.