Increasing Credit Card Acquisition Through Customer Understanding
Use a four-step approach in database management to acquire new customers: market research; customer segmentation by attitudes; developing the optimum messaging for the most promising customer group; applying insights to predict new and potential customers.
This can be demonstrated by the recent experience of a major British bank had. It wanted to launch a co-branded credit card that would not require expensive and undue marketing. While previous commissioned research provided statistics, the bank remained without actionable direction.
Here's how the challenge was tackled.
Know the customers' individual "hot buttons." It was first critical to identify these hot buttons. This meant looking at a lot of communication messages rather than a few, as is usually done. A professional facilitator helped the bank's development team create 225 elements in a full-day session. About 150 respondents, as well as an optimization method known as IdeaMap, identified the hot buttons for customers. These hot buttons were then optimized into compelling offers. To find hot buttons in a realistic context, the group tested 100 different propositions and by making them unique, 15,000 possibilities were tested. Furthermore, each customer saw systematically varied combinations of features/offers. This systematic variation, called experimental design, was critical, because it identified what specific concept elements were "turn-ons" and "turnoffs" for each customer.
Discovering new segments of credit card customers. Conventional ways of dividing customers don't really reflect the way customers differ in their reactions to credit card offers. Attitudes are the gateways to success.
There were two major attitude segments and a third group of individuals who did not care about anything that the bank would tell them. None of the three groups was particularly interested in credit cards, as such. What made the difference was what the bank would provide, and the language used to communicate the offer.
The first segment (40 percent of customers tested) was highly responsive to financial benefits (e.g., low interest rates) and not particularly interested in other offerings (e.g., gifts). These were not prime prospects - it would be too expensive to create a card for them.
The second segment (35 percent) strongly responded to a number of lifestyle offers, such as a chance to participate in trips to certain events. They didn't want merchandise, and they were only modestly interested in low annual percentage rates and better financial terms. These were the prime prospects. They could be satisfied and it would profitable to gain them as customers.
Creating an optimum offering and sending it to the right person. The optimum concept for the lifestyle group comprised the elements (messages) that would turn on these customers. With the knowledge of their hot buttons, creating this concept became quite easy. The bank held back reserve elements for this segment to promote activation of the card, as well as messages to send after six months and one year to refresh the relationship.
A cross-tabulation of the individuals in the segments showed that they cut across various demographic segments, such as age, income, etc. Since simple targeting would not work, it was necessary to create a decision tree to enable the bank to estimate the odds that a specific person would belong to the lifestyle segment and then concentrate on sending the right message to people who were deemed to fall into that segment. The results were promising - showing a fairly good increase in response rate for acquisition. (Lifts in response rate vary in these studies from 40 percent to more than 150 percent.)
Implications for direct marketers:
1) Know the customer on an ongoing basis, through actionable studies of customer attitudes. It is possible to obtain a lift in response rates using consumer reaction to and optimization of concepts. But this requires ongoing knowledge of changing customer tastes and wants, and research that can be immediately turned into optimal concepts and offerings.
2) Combine customer needs and wants with database information to create a more powerful model. Rather than renting databases and developing models based upon past behavior, incorporate knowledge of customer wants and optimal messages into the modeling.
3) Work with large customer base sizes (e.g., on the Web) and create smaller segments that approach 1:1 marketing. It may be feasible to create individualized offers for credit cards by creating segments of very small sizes. These very small segments allow for the increased personalization of the credit card offer. n