Effective segmentation of the market is probably the single most powerful business strategy to help ensure long-term, profitable growth.
Segmentation can generate more revenue by making it possible to sell more of your product to more people, and by increasing the revenue coming from each customer.
While there is no doubt that segmentation is sophisticated ad complex, its true power is simplicity itself: ask people what they need and then deliver it.
Bank card issuers used segmentation to great advantage.
When bank cards were introduced in the 1950s, they extended the convenience of the general-purpose cards that until them were used only in travel and entertainment. The new segment wanted to use credit cards in local markets. By meeting that need, the banks obviously increased the size of the card market.
Within the bank card industry, individual issuers have chosen to offer different products to different segments of the market, further increasing overall sales. Almost everyone now offers a gold card to affluent consumers, up from only 40% six years ago. The number of gold bank cards has risen to over 30 million, from six million.
Some bank issuers have chosen to offer an entire product line to capitalize on the newly segmented markets. Citibank has more than 10 products, each designed to serve a specific group. MBNA America Bank, known for its affinity programs, is built on segmentation, with 1,500 products for 1,500 different segments.
One of my favorite examples of segmentation is the sneaker business -- or athletic footwear, as it is known today. It sold 230 million sneakers last year, up from 25 million in 1965. The annual growth rate was five times that of the population.
One company, Nike, sells more than 300 types of sneakers, clustered in 12 market categories. The price of a pair is commonly around $100l; the average in 1965 was $8.
And this is a business that was thought to be saturated in the 1960s as the baby boomers grew up.
In terms of pricing, the bank card business is going in the other direction, with interest rates reduced. We can hope that the recent round of price reductions is merely a long-overdue correction, and that we soon will return to a value-focused approach to meeting customer needs.
Price reduction may be the easiest way to segment, but it's not usually the smartest over the long haul. It is worth nothing that more than a few card issuers have been able to flourish with a high-value, premium pricing strategy, as ikn MNBA's approach to the memberships of various associations.
Then there are the airline co-branded cards, with annual fees of $50 to $100, and -- inspiration to us all -- the $350 American Express platinum card, which comes with white-glove service, access to special events, and an overdose of prestige.
Aside from increasing customers and revenues, segmentation, if properly designed, can also reduce your costs.
Customers who are self-service oriented, for example, do not seem to mind using a telephone to obtain account information from an automated response system.
Marketing expenses can also be lowered. When you know specifically whom you want to speak to, and who are the customers you want, you can focus all your marketing efforts on those people alone.
It doesn't make sense to waste money communicating to people who are not important to you, or who have different interests than the segment you are talking to.
There is, of course, a significant opportunity to control the bottom line through risk reduction.
In fact, the creditworthiness factor is probably the most frequently used segmentation factor in our business today.
The problem is that everyone is going after the same segment -- one that is not customer defined, but product defined.
One of the things that makes our business so challenging is that we have two distinct types of customers -- current ones and prospects. Each group to be addressed differently.
This means we need to segment not only the overall, external, market, but also our internal market -- the current customers.
Prospects' needs are often different from those of current users. The reasons for applying for a card are often very different from those for keeping and using one, even among customers in the same market segment.
This can make segmentation even more complicated, like playing two games of chess at the same time. But play we must.
That subtle but powerful distinction -- customer differentiation rather than product differentiation -- forms the basis of all effective market segmentation. It has direct application to the enormous marketing success achieved by three products introduced in recent years -- the Discover, AT&T, and General Motors credit cards.
People, Then Products
The first to develop a coherent theory of segmentation based on customer differentiation was Wendell Smith. In a 1956 Journal of Marketing article, he wrote:
"Product differentiation is concerned with the bending of demand to the will of supply, by advertising or promoting differences between a product and the products of competitors.
"Market segmentation, on the other hand, requires that we view a heterogeneous market as a number of smaller homogeneous markets, in responses to differing product preferences of important market segments."
Mr. Smith's point was that a business needs to market products in response to what people want. Start with the market and work backward to the product, rather than the other way around.
Discover, AT&T, and GM did not attempt to win card market share simply by coming up with a fancy slogan. Each addressed a specific need of a certain segment of the American population, and an apparently large segment, the one that is interested in discounts, rebates, special deals.
How can we translate segmentation theory into action?
The first step is to identify the segment.
The first and most common criterion is geographic. Here the market can be divided into different areas, such as city, state, or region, and according to size of each entity. One common, form of differentiation among bank card issuers is local versus regional versus national.
A second basis for segmentation is customer demographics: age, sex, income, occupation, family size, education, religion, race. Demographic variables are important, since they are often readily available and effectively predict differences in customer needs. The category of "student," for example, represents one of the biggest card markets of the past five years.
Sex is another powerful segmentation category. American Express was able to increase its market share significantly in the 1980s by going after the female market.
A more recent basis for segmentation, which can cause problems, is customer psychographics, or attitudes -- based on lifestyle, social class, or even personality type. American Express was successful with prestige marketing for many years, going after the banks followed with their gold cards.
But be careful when buying psychographic-based, geographically coded lists for direct mail, or when encoding your own files. Many issuers have found that the predictive power of these computer-generated clusters does not always meet expectations.
Finally, there is behavioral segmentation -- one of the most potent approaches, because it relates directly to a common behavior or action. This category can include such variables as spending, types of spending, benefits sought, and stage of readiness (unaware, aware, interested, desiring).
At the same time, it is dangerous to stop at behavioral segmentation.
We often confuse knowledge of the customer's behavior pattern with knowledge of the customer itself.
Therefore, it is common to use these segmentation schemes in combination with each other.
No discussion of segment definition would be complete without mentioning the vast capabilities we all have to create market segments with our customer information systems. You can even create the ultimate segment: the segment of one. But unless you can translate the micro-segments into tactics that increase the bottom line, you are wasting your time.
And how do you know when you have a viable market segment with true bottom-line payoffs? There are five criteria:
* The segment should be relatively homogenous. No construct is 100% pure, but the closer to the criteria you have set, the better.
* The segment should be of sufficient size and potentail to warrant spending internal marketing resources. How big is big enough depends on your organization's needs and goals.
* The segment should show a clear and distinct consistency in market behavior. If its reactions are inconsistent, the segment is not properly defined.
* It should be possible to reach the defined segment through available media, including direct mail, in a cost-effective manner. (This might be hard to do when psychographics is the basis of segmentation.)
* Perhaps most important, the chosen segment must meet profit objectives. This factor is too often overlooked before marketing campaigns are initiated. In our eagerness, we sometimes tend to just hope that a segment will be profitable.
While we look very closely at the risk-based components of profitability, what about the revenue side of the equation? Will the targeted customer use the card? Will those spending dollars revive? And what is the likelihood of retaining the accounts over the long term?
Fortunately, it is much easier to measure profitability of existing customers than of potential ones. By overlaping a profitability model on your relation data base, you can calculate the profitability of any segment that can be defined on your file.