Can you predict what your customers will do next?

Client Technical Leader for UK Public Sector, IBM UK Ltd.

Businesses are under increasing pressure to meet heightened customer expectations in this digital age. Customers expect a consistent cross-channel experience and good customer service, strongly influenced in these expectations by family and friends. Businesses must respond by utilizing all available targeting information to earn and sustain customer loyalty and drive up profitability.

Making History 8.jpgThe Championships at Wimbledon are now underway. Fans, commentators, coaches and players are afforded a wealth of data and insight into every match. IBM SlamTracker (first introduced in 2008) provides fans with the hands-on customer experience they crave by highlighting the “Keys to the Match” for each player and charting progress as the match progresses in near real time, virtually putting them into the coach’s shoes.

The “Keys to the Match” are created using predictive analytics and offer fan-friendly insights into a match by identifying the key performance indicators that will likely affect a player’s performance, specific to each match. Each player on court must work out for themselves how they will adapt their play to win. During the match, fans are able to use the IBM SlamTracker to monitor a player’s progress against their keys in near real time.

To develop decision models for the IBM SlamTracker (available to fans across multiple digital channels) IBM utilizes 41 million data points from eight years of Grand Slam tournaments. Today at Wimbledon, 48 highly trained tennis analysts are capturing data like the direction of the serve, the speed of the serve, the return shot selection, the number of strokes in the rally (forehand and backhand) and detail of the point ending stroke. Data quality and accuracy is key; ultimately these statistics are being captured and shared around the world within four seconds.

Similarly in business, information about a customer’s activity is captured and combined with customer records and details about products and services to develop decision models. These models are then used to make real-time operational decisions, for example to calculate the next best action that the organization should take with a particular customer. Such actions aim to maximize retention (for example, by using a promotion or customer service) and increase customer lifetime value (by cross- or up-sell) by responding to customer needs and sentiment during the interaction or through events, attributes or other signals that suggest the likelihood of a specific outcome.

When actions are taken in business, the outcome must be captured to measure the benefits. Such data also offers insight for further performance improvement.

As David Henderson from CIO UK recently commented: “there are five possible lessons CIOs can learn from the Wimbledon technology experience,” one being “combining real time and historical data to drive insights[—]consider opening up this data to partners to foster innovation.” Imagine what this approach could do for your business; what “keys” are most important for you and your clients?  

Related resources