[Skip Global Navigation]


Customers Home



Telenet, the largest provider of broadband cable services Belgium, offers modern, accessible communications, information and entertainment services to more than 1.6 million homes, plus services to business customers in Belgium and Luxembourg. Telenet is at the crossroads where television, the Internet, and telephony converge. The provision of iDTV (integrated Digital Television) is intended to ensure that all consumers can benefit from this convergence.

One of Telenet’s priorities is to respond correctly to its customers’ current and future needs, using the different channels available for customer contact. Until recently, however, one of those channels – the Customer Care Call Centre – was being markedly under used.


Koen Puttemans, marketing intelligence manager at Telenet, explains: “Seeing that we as a company are enjoying unrestricted growth, that we have the broadest possible customer profile and work with complex technology, a smoothly operating Customer Care Call Centre is essential. Naturally, the basic task of the call centre is to answer our customers’ questions correctly. However, we wanted to broaden the tasks of our call centre operatives to allow them to make relevant and interesting offers to our customers as well.”

Telenet did not wish to have its call centre workers randomly showering customers with offers, so the company started to consider the possibilities offered by predictive analytics. Predictive analytics is set of technologies that makes a more targeted customer approach possible by analysing specific details about customers and providing the insight gained to those making the contacts.

From its support logs, Telenet had a large amount of historical customer data available. The company needed a solution that would use this data so that call centre workers could make targeted offers.

To limit the burden on its workers, the solution had to be as easy and automated as possible. And it had to allow a worker to make a commercial offer to a customer who was only requesting technical support.

First, the customer’s problem had to be solved quickly. Only then would there be a good starting point for a commercial conversation. However, the question that had to be answered was: how to ensure that the call centre workers had the correct information to make a relevant offer?


Telenet decided to approach a number of suppliers who would design a predictive model and also provide the required consultancy/support in the initial phase. In early 2006, Koen Puttemans and his team concluded their search for a supplier when they chose SPSS Inc.’s data mining solution, PASW Modeler.

“We chose the SPSS Inc. solution for various reasons,” he says. “The intuitiveness of the PASW Modeler interface is very important, and our users learned to work with it very quickly. Furthermore, the solution is ideal for rapid integration into our hardware and software environment, and with our database. We also believe it was very important that we get along with the SPSS Inc. team. Our contacts were enthusiastic and open, and completely focused on developing the best possible model, as we knew in advance that a ready-made solution would not be possible.”

In early 2006, Telenet’s marketing intelligence team and SPSS Inc.’s consultants created seven models – three for cross-selling and four for up-selling. The models were then implemented and rolled out in the customer care environment. In August 2006 the project went live.

Interested in Telenet? Download the PDF


Thanks to PASW Modeler, Telenet can now identify its customers much more efficiently. When a customer calls, the software displays pop-ups that provide useful information as a starting point for the call centre worker. Following each call, the customer’s details are updated.

Incorporating the correct parameters is an important part of the process. When should the call centre workers start up-selling or cross-selling and when should they not? Telenet did not wish to ignore the intuition of its workers, because the human factor remains extremely important in relationships with customers. Therefore, in addition to a thorough training programme, Telenet also provides coaching and support for its operators.

At Telenet, the Customer Care Call Centre is now one of the most vital parts of the whole company. Sales from the call centre doubled within six months of the launch of the sales support pop-up, and the strength of the software is reflected in the sales figures for Internet, telephony, and iDTV.

“This was the first time that we have fed analytical results back so directly and precisely into our operational processes. We have learned that the success of such an initiative not only depends on good analytics, but also on the way in which it is supported and encouraged on the operational side. Both elements are necessary to achieve the added value offered by these analyses,” Puttemans concludes. ”If you can combine the two, you can achieve an excellent return within a short period of time.”

Ultimately, this project is part of a broader process within Telenet for using customer profiling and data mining to work more efficiently and effectively and, therefore, achieve better corporate results. The team of analysts is now looking specifically for correlations between purchasing behaviour and other customer information, based on several hundred potentially predictive factors that may provide an insight into the chances that a customer will agree to a commercial proposal.