Teamwork Lets Prospect/Customer Relationship Marketing Enrich Data

Contrary to the claims of some prominent companies in our industry, state-of-the-art prospect and customer relationship management (P/CRM) is much more than "star wars" hardware and software. This is because P/CRM is a process rather than a set of sophisticated tools. No matter how advanced or tightly integrated the underlying "plumbing," the process of P/CRM involves large doses of iterative human input.

This human input is most effective when individuals with disparate skill sets work as a team to develop innovate strategies, in a collaborative effort that I refer to as "Dynamic Data Leveraging." In addition to an MIS component, the team generally includes:

  • Data mining experts, particularly those with significant direct marketing experience. It is critical for these individuals to be well versed in a wide array of analytical techniques such as predictive modeling, clustering, lifetime value analysis, and focus group and survey research.
  • Marketing and creative professionals such as agency personnel, and especially those with vertical industry expertise. Just as with the data miners, the best of these individuals also provide strategic insight into any P/CRM initiative.

To illustrate how this iterative and cooperative process of Dynamic Data Leveraging works, consider an example that can be subtitled "Making Something From Nothing: Developing a P/CRM Initiative From the Absence of Data":

A travel-by-credit services company focused its prospecting efforts primarily on the customer databases of large financial services organizations. The offers were co-branded, including a credit card tied into airline, hotel and restaurant discounts.

To sharpen the prospecting focus, a series of predictive models were built. The effort was a success, with excellent segmentation power resulting from models that crossed response with subsequent usage as well as bad debt propensity.

A "eureka moment" occurred during a review of the variables that drove the models. The following relationship was found to exist between Gender Code and performance:

Gender Code / Ratio to Average

  • Male: 104
  • Female: 94 
  • Both: 98
  • Unknown: 158

Intrigued by the behavior of the individuals with a Gender Code of "Unknown," the data miner investigated and discovered that the Gender Codes were created by a lookup table that consisted of thousands of Given Names. Individuals tagged with the "Unknown" designator had names that did not appear on the table. ("Both" represented names such as "Pat" that are common with both sexes.)

Further investigation revealed that the lookup table had been created many years ago and "because it did not represent our country's current diversity "was comprised almost entirely of Anglo Saxon names. As a result, almost all of the "Unknowns" had Asian names such as "Ming" or Middle Eastern names such as "Hassan." This was a revelation to the client, whose customer base was predominantly older couples of Western European heritage.

Intrigued, the data miner generated counts of the converted prospects "that is, customers "on the analysis file who had Asian or Middle Eastern Given Names. Numbering just under ten percent, this was a niche with sufficient heft to support a cost-effective targeting effort.

Demographic profiles were run on these Unknowns, which indicated that their life-stage was very different from that of the typical responder. The Unknowns tended to be unmarried, under thirty years of age, and living in urban Multiple Family Dwelling Units. This was in stark contrast to the typical customer, who was older, married and living in a suburban Single Family Dwelling Unit.

The Circulation Department then commissioned focus group and survey research to develop a complete understanding of the Unknowns, including their motives for responding to the offer. It recognized that this insight was critical to optimally leveraging this promising market niche.

The research revealed that these customers tended to be first-generation Americans who were either graduate students or just starting out in their professional careers. They generally took advantage of the travel-by-credit service to visit their families in other parts of the United States. This "again "was a significant departure from the typical customer, who used the service primarily to visit popular tourist destinations such as national parks.

At this point, the decision was made to convene a formal task force to fully leverage the findings. In addition to the data miner and representatives from the Circulation Department, individuals from Creative and MIS joined the team. As a result of several brainstorming meetings:

  • The Creative Department developed a new prospecting package to address the very different needs of this prospect subset.
  • MIS enhanced the outdated Given Name look-up table. A large number of Asian and Middle Eastern names were added, along with corresponding ethnicity codes. This was necessary to precisely identify the desired niche universes.
  • The Circulation Department embarked on an aggressive test program to uncover additional cost-effective prospecting sources. Specifically, lists were identified that appealed to young, single, urban Americans of Asian and Middle Eastern heritage.
  • The Creative Department worked in conjunction with Circulation to develop unique, targeted cultivation offers to existing customers. This, in turn, inspired additional focus group and survey work to gain even deeper attitudinal insight.

This iterative and cooperative effort of Dynamic Data Leveraging, which involved industry professionals with data mining, circulation, creative and MIS backgrounds, was a resounding success. Because of teamwork, an intriguing pattern in an analysis file involving the Gender Code of "Unknown "in a sense, the lack of data "snowballed into a very profitable set of targeted P/CRM initiatives.