Q & A: the new data--and analytics-driven site selection.

• Position: Marketing News - Dialogue with Peter Harvey - Interview

Peter Harvey is the founder and CEO of Intellidyn Corp., Hingham, Mass. The company specializes in optimizing target marketing, risk screening and multichannel integration for several industries, including financial services. Prior to founding Intellidyn Corp., he held senior marketing positions with Bank One, Chase Retail and GE Capital.

Q: What are some of the ways that banks today are reaching beyond traditional branch site selection methods to ensure optimum expansion opportunities?

A: Some savvy bank marketers are now using an array of new data sources, advanced database mining and sophisticated analytics techniques to help their institutions determine branch sites with the greatest potential for growth and return on investment.

One large foreign bank is now utilizing this technique to start from scratch and drive its expansion into the U.S. market. Borrowing sophisticated targeting techniques used by top direct marketers, this bank begins by determining the characteristics of the "best customers," typically defined in terms of profit and tenure. That information is analyzed across the nation's most comprehensive and up-to-date transaction, credit, demographic and behavioral consumer databases to identify characteristics of target customers.

It is the combinations or interactions across these databases that uniquely identify individuals that most closely resemble an institution's best customers (e.g. homeowners by their property characteristics, plus credit and purchase behavior, over time). To achieve this depth of business intelligence, the analyst has to interact with multiple data sources at the household level. If you can create highly discriminating profiles of individuals, you can size the density and select sites and markets. But you have to go further.

Q: What do you mean by going further?

A: The most basic reason that clones of your best customers would elect to do business with your institution is that they are in the market for one of your products or services. You can use your marketing models to identify this possibility (i.e. score their likelihood to respond/convert).

There is ample data from companies like Claritas that can help you determine whether prospects already have a competitive product relationship, enabling you to estimate the likelihood of ownership at the local market level. Cross tab the likelihood that they need your product with the likelihood that they own this product to identify high-density markets...