warranting different treatment - has been practiced in packaged goods companies for well over 40 years and is increasingly in vogue for financial services providers. Banks, too, have practiced segmentation for years, initially building on the desire to avoid credit risk and loan losses rather than to grow revenues or differentiate themselves from competitors. As non-banks continue to make incursions into what was once thought to be the exclusive domain of banks, as new electronic delivery channels threaten to further commodotize bank products, and as the importance of physical proximity of branches lessens, the need for segmentation will only increase. The ability to customize offerings to individual households has been made possible through increasingly more efficient and available information technology. Those banks that have already begun to employ segmentation strategies have achieved almost overnight growth and have been catapulted into market leadership positions. Types of Segmentation One commonly used approach is life-cycle segmentation, which is based on the idea that as individuals encounter certain life events, their needs will change. An obvious benefit of life-cycle segmentation is that if you accurately predict the event and are among the first to make the offer, the overture is relevant and should be more effective. The major disadvantages of life-cycle segmentation is the low probability of any single institution being alone in its knowledge of a prospect's life event. Psychographic or attitudinal segmentation seeks to tailor advertising copy and products to the way customers and prospects view the world. Just as Nike appeals to those who wish to live life to the fullest, so some financial services providers aim their advertising at those frightened by the reality of having to manage money. The advantages of attitudinal segmentation include knowing a message will be reassuring to those sharing the attitudes being reinforced. The limitations lie in the difficulty of accurately polling specific households on their attitudes. Demographic segmentation, which uses characteristics such as age, sex, income, and marital status, is commonly used by banks and their research services. For certain types of targeting, it clearly is essential. The major flaw of demographic segmentation is the mistaken belief that securing affluent customers assures profitability. Affluent households may have high profit potential, but they also are generally more discriminating, more likely to spread their relationships among many providers, and thus have about as great a likelihood to be unprofitable as their less affluent brethren. Segmentation that groups individuals on the basis of what they actually do - that is, behavioral segmentation- bypasses attitudes, demography, and life events but includes the influences those factors have on behavior. Its advantage is that it reflects what financial services customers actually purchase. Its disadvantages are that it is usually expensive and dynamic (requiring updating) and still will cluster profitable and unprofitable customers together. Segmentation based upon profitability alone ensures the ability to invest, protect, change behavior, or demarket on the basis of customers' value or cost to you. Arguably it is the only segmentation scheme that will guarantee you are not prioritizing segments that may include big money-losers. Therein lie its advantages and disadvantages. By knowing who is profitable and unprofitable, you can create the appropriate prescriptive actions to manage your customer portfolio. On the other hand, this approach is potentially punitive and short-sighted in its treatment of unprofitable customers who have the potential for behavior change and increased profits. Hybrid segmentation schemes can combine the best of all worlds. Behavioral data coupled with profitability data can soften the bluntness of profitability segmentation and allow for refinement and exceptions to pure behavioral segmentation. Potential Uses Segmenting customers on the basis of the distribution channel they use has some clear benefits to the bank, but not necessarily to the customer. From the bank's perspective, knowing which customers use branches, telephones, and automated teller machines provides for better cost accounting and allows more targeted messaging. Though an understanding of channel usage is valuable and though delivery behavior may be a key component of customer segmentation, distribution channel usage data alone are not very useful, because customers rarely use only one channel. The real controversy, then, is whether to let each separate product line of business determine and execute its own segmentation scheme or to create a bankwide household relationship segmentation paradigm that seeks customer optimization. Each is seductive, yet product optimization has ruled largely because of the ways banks have organized and allowed product company autonomy. Product versus Household Which segmentation approach is followed or how the two are integrated is a function of the bank's overall retail strategy. If the business strategy creates intimacy and deepens relationships with selected households, a company focused on product would arguably be more responsive to customization as long as overall household profitability increased. Arguments as to whether or not this would ever occur are rampant in organizations and largely a waste of time in a world where disciplined market testing can answer the debate objectively. What need not be debated is the premise that the sum of product optimization efforts across a bank do not add up to a household optimization effort. When each separate product company within one bank acts unilaterally with the customer based upon its product agenda, the relationship can suffer. Information sharing about customers across businesses; preferential terms where warranted because of other relationships in the bank; recognizing the full relationship at the point of customer contact; and recommending solutions that lie outside of product silos are all casualties in a product optimization world. On the other hand, the complexities raised by household segmentation when applied to product companies, staff functions, and geographic management are anything but trivial. They require nearly a complete rethinking of every organization including marketing, distribution, credit management, human resources, operations, and systems. And they require customer data bases and analytics that are household-driven versus product-driven. The Best of Both Worlds The extremes of either approach are suboptimal. Product segmentation in a multiproduct organization perpetuates silos in a world where consumers increasingly expect and get integrated treatment from vendors. Best-in-practice examples regularly tout the success of Lands End, Nordstrom, USAA, and others who have taken a more holistic approach to customer management. Single product plays that are category-killers are few in number and make banks more vulnerable to providers who build relationships and use the information they gather as strategic weapons. What is needed is balance and the capability to overlay the schemes for commonality when it can be achieved. At times, the two will be incompatible and there must be selection of one over the other for prioritized customers. Mr. Evans, a senior vice president of Chase Manhattan Bank Corp., focuses on retail strategy and marketing issues.

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.