The Myth of the Average Customer

There’s an old joke that goes something like this: A statistician has one hand in a bucket of ice water and the other in an open flame. He says, “On average, I feel comfortable.” Silliness aside, there is a lesson in this for market researchers. Often, thinking in terms of the “average customer” washes out extremely valuable information. Rarely does an average customer visit a Web site, any more than any given family has 2.1 children.

For any given business, there are always several varieties of the “typical customer.” Finding these varieties and adapting your business to them is the goal of segmentation. That said, this article will explain the concept of segmentation and its importance to Internet businesses.

Why Segment?

Your customers consist of diverse groups of people, which is why a one-size-fits-all approach to satisfying your customers is bound to fail. On the other hand, it usually isn’t practical to target individual consumers.

Segmentation finds a middle ground by identifying a small number of homogeneous groups (known as segments) of customers with similar characteristics. This common ground could be based on many factors, including product performance needs and wants, demographics, Internet experience, or value based on purchase behavior.

Once a segment has been identified, a savvy marketer has several options. For instance, he or she could develop different promotions targeted to each group’s distinct needs. He or she could incorporate mass personalization into his or her Web site so users with different Internet experience levels see different versions of the site. Or high-value customers could be provided with a special customer-support phone number that doesn’t have seven levels of voice prompts. The possibilities are endless!

What to Measure

Successful decision making is a function of knowing your customers, and understanding market segments is a critical component of this knowledge. The key to identifying valuable segments is to find variables that are useful in helping you make better marketing decisions.

If the segmentation does not show a difference in preferences, then it will not likely be useful for making a marketing decision. Take gender, a popular demographic segmentation variable. A marketer faced with deciding what colors to offer for next year’s pickup trucks might segment the pickup-truck buyer base into two groups, one for men and another for women. If men like blue and women like green, then the marketer has found a preference difference that can be used when marketing pickup trucks. However, if the gender segmentation finds that men’s and women’s preferences for color are the same, the marketer does not have anything unique that can be used when marketing separately to each group.

Another strategy is to group users together based on their actual behavior. On the Web, data such as frequency of use and purchasing history is often readily available to marketers. Segmenting your customers based on the categories of products they traditionally buy provides you with the opportunity to create special offers for each segment that are relevant to that group’s wants and needs, though not necessarily appropriate for other segments.

The Power of Segmentation

Segmentation can be a powerful tool for Internet marketers. For example, suppose an online clothing retailer has identified two major user segments: fashion slaves and bargain hunters. Fashion slaves are most interested in owning the latest styles and are not particularly price-sensitive. Bargain hunters, on the other hand, don’t mind wearing 20th-century styles as long as they get a good deal. These segments could be identified using purchase-history data. A Web business can capitalize on this segmentation by dividing up the Web site into areas aimed directly at certain segments. Similar promotional strategies can be implemented even more easily by using email promotions.

Think about the huge advantage segmentation gives online retailers over brick-and-mortar establishments, which are much more limited in the number of ways they can set up their stores.

Implementing Segmentation

Segmentation can be simple or complicated. If you know that age is a major factor determining customer preferences, then it is a no-brainer to divide up your analysis of any data according to age groups. Segmenting users by multiple variables, such as survey responses, involves some complicated procedures. Fortunately, you don’t have to know anything about them. All you need to know is that the method used is called cluster analysis.

To carry out segmentation, you will have to get your hands on some software that does cluster analysis. (Any standard statistical software, such as SAS or SPSS, will do the trick.) But I won’t get into those details, which will depend on the specific software you use.

Segmentation is a good example of a fundamental marketing concept that becomes even more powerful when applied to the Web. On the Internet, it is more feasible than ever before to target real segments instead of nonexistent average customers.