Firing up customer behavior

Powergen, now E.ON UK, is one of the U.K.’s leading energy suppliers, operating over nine million electricity and gas accounts across the country. The company produces electricity from a portfolio of worldclass power stations and is also one of the leading names in “green” power generation. With power in constant demand, energy suppliers are in an extremely competitive business. Consumers are constantly offered lower prices by rival companies as well as special deals to switch from their existing suppliers. A competitive edge is often retained by factors unrelated to tariffs. For example, customer relationships are key, and good CRM practice, if implemented correctly, can be a significant advantage for an energy supplier.

Outsourcing out of tune

In 2005, the Customer Insight team at Powergen was outsourcing the majority of analytical work, in order to better predict trends in customer behavior. However, outsourcing was proving expensive and difficult to use. Stewart Robbins, customer knowledge manager at Powergen, explained: “As the energy market became increasingly competitive, it was clear that we needed to move up a gear in our analytics processes and stay up to date with the competition. “Outsourcing marketing activity didn’t provide much visibility of the processes involved. The previous processes required specially trained analysts who, although they could use the technology, weren’t necessarily in tune with what we wanted to achieve.”

A clearer picture

As a company, Powergen needed to establish a clearer picture of what customers really wanted from their power supplier. It looked for a solution with a low set-up cost which was simple to install and didn’t require specially trained users. The company chose to invest in IBM SPSS predictive analytics software, which analyzes past behavior in order to predict how customers are likely to behave in the future. Improving call center processes was fundamental to Powergen’s success. Using IBM SPSS Statistics, the company was able to identify patterns in how customers were behaving on calls and analyze the reasons for this behavior. In addition, Powergen aimed to cut down the number of times customers needed to use call centers. Robbins continued: “It was in everyone’s best interest to cut down on the number of calls customers needed to make. Having fewer calls means happier customers and allows everything to run more efficiently, especially at busy times of the year. Using predictive analytics techniques, we were able to establish links between customer activities. For example, if customers were calling about meter readings, we were able to tell if it was appropriate to discuss paying by direct debit at the same time.”

As well as customer behavior, Powergen used IBM SPSS Statistics to analyze customer feedback from surveys, calls centers, and communications such as letters and emails. This information was used to improve the design and tone of power bills. “It might seem like a small matter,” said Robbins, “but we discovered that the way power bills were written was extremely important to customer satisfaction levels. We used IBM SPSS tools to analyze behavior and establish which customers were most likely to default on payments. As a result, customers who have loyally paid their bills on time for years won’t receive a red reminder just because they’ve gone on a three-week holiday and missed their usual payment date.

“Similarly, commercial and domestic customers obviously have very different priorities, and we’ve tailored their bills accordingly. It’s all about being more sensitive to customer needs and tailoring our activity to individuals as far as possible.”

A core analytical tool

Making marketing and CRM activity more specifically targeted has inevitably saved money. However, using IBM SPSS Statistics has also reduced overhead due to the quick and efficient system. The analytics team has been able to look at a particular problem or situation, develop and execute on findings quickly, and then move on to the next step. Robbins continued: “Using IBM SPSS Statistics as one of our core analytical tools, we have reduced unnecessary marketing activity, which in turn saves money and helps prevent us from targeting the wrong people with inappropriate campaigns, which could be potentially damaging to our reputation.

“We started the project with just a few licenses, which has grown significantly over the last two years. “IBM SPSS Statistics has more than paid for itself in the short time we’ve had it and, as a result, we’ve started to use text mining tools from the company to look at customer records in call centers. In the analytics team at least, IBM SPSS Statistics is the fundamental tool in our toolkit, and has more than proved its value.” The company has even been able to improve the accuracy of meter readings by analyzing previous statistics for the property in question.