A Marketing Goldmine: Using Customer Data
February 06, 2008
How online marketers can transform e-commerce data into a goldmine
By Sheldon Gilbert
It's practically common knowledge that the key to increasing business today is data—the more information a company has, the better its competitive advantage in the marketplace. By taking Internet click-trail data from servers, companies can analyze consumer behavior, identify their interests and pinpoint their affinities for specific products and services. This treasure trove of information can also be used to predict what your customers will buy, when and at what price point. You just need to know how.
Data Mining for Gold
Most marketers have a natural sense of what their customers want. However, great marketers seek to validate their assumptions with "proof' rather than simply relying on their intuition and experience. But, although many are certain that the data lies somewhere on the Web server, they don't know how to extract it.
The first place online marketers tend to seek assistance is often with the IT department. Although the network managers, IT administrators and Webmasters may know where to find this critical data, they don't know how to analyze it or derive intelligence. Under pressure from the marketing department, a motivated engineer may develop a crude system for identifying customers with similar behavior patterns or transactional histories. However, the ability to understand how to use that information is usually beyond their reach.
Upgrading the Old Art
The two most prevalent approaches to segmentation of customers involve the use of demographic and psychographic data. Demographic data includes information such as income, education, race, age, home ownership, etc. Psychographic data examines individual lifestyles and behaviors, including interests and values. These segments are usually based on figures gathered during customer transactions or from information provided by the customer upon registration for a Web site or service.
The problem with both of these isolated approaches is that they give very poor insight into what the customer really desires and in predicting his or her future behavior. For example, marketers often draw inferences that an individual with $100,000 income will be interested in buying a designer handbag because she can afford it, even though there is no direct evidence to support the customer actually wants it.
The third approach is behavioral observation. While applied less frequently, it is the most valuable. By observing the behavior of customers during the online shopping experience, marketers can get a more accurate profile of a customer's true interests and intentions for future purchases. Behavioral targeting uses information collected on an individual's Web-browsing behavior—such as the pages visited, searches made, objects enlarged, click through paths and advertisements viewed. Once these behaviors are observed, they can be coupled with the demographic and psychographic data collected.
Science of Segmentation
Then a segmentation scheme must be developed. Ultimately, the segmentation schemes should be designed to find the most valuable sets of customers. These may be found by dividing your customer base by patterns such as recent purchases, loyalty program registration or dollar value spent. Most segmentation schemes can benefit from basic RFM segments (recency, frequency and monetary) and visitation (browsing behavior). Some modality for all of these can be combined to identify the top five groups of customers.
Monetizing Your Data
Once the segments have been established, marketers must understand the dollar value associated with the members of that segment. (If a business class traveler is 20 times more profitable than an economy traveler, do you know the best place to put your marketing dollars?) If you've created multiple segments and they all have the same value, then these segments may not be useful. Although some segments may tell you that your customers have divergent interests, it is critical to understand how much money and profit they generate for your business.
The high value segments—or those that generate the most revenue or profits for a company—are the most important to pay attention to. If your top segment of customers generates an average of $500 annually per person for your company, but there are only 1000 members, your marketing efforts might be focused on moving your second tier revenue generators, with 100,000 members (that average only $200 annually per person), into the next highest segment.
Investing for Profit
When you know which customers are 'top value" for your company, you can tailor marketing programs for them and invest in those relationships. Having the insight and intelligence regarding the value of each segment enables marketers to develop programs that are appropriate for their contribution to your company's bottom line, or designed to migrate groups from the bottom to the top of your segmentation ladder.
Regardless of the scope of data mining a marketer performs, a little intelligence can be used to generate a lot of revenue. When you know which customers are "top value" for your company, you can tailor marketing programs for them and invest in those relationships. By focusing on growing that sweet spot of customers, marketers will be able to exponentially maximize their ROI.
Sheldon Gilbert is the founder and CEO of Proclivity Systems, developers of Proclivity Mail, a predictive engine to help clients anticipate and forecast consumer purchasing behaviors online. He can be reached at sgilbert@proclivitysystems.com.
Sales & Marketing Management Magazine
This article is brought to you by Sales & Marketing Management, the leading authority for executives in the sales and marketing field.
|
|
RSS
