Creative Segmentation: How David Can Take on Goliath
Ben DiSanti, Co-Founder & Managing Partner, DiSanti, Hicks + Partners;
Ken Hicks, Co-Founder & Managing Partner, DiSanti, Hicks + Partners;
Greg Filler, Director of Business Analytics & Category Management, Merrick Pet Care
Edited By Ananya Gupta, Medill IMC Class of 2015
Published on 10/21/2015
“We have a definition in our heads of what an advantage is — and the definition isn’t right. And what happens as a result? It means that we make mistakes. It means that we misread battles between underdogs and giants. It means that we underestimate how much freedom there can be in what looks like a disadvantage.”
— Malcolm Gladwell
Introduction to Segmentation
“Customer segmentation is the practice of dividing a customer base into groups of individuals that are similar in specific ways relevant to marketing, such as age, gender, interests, spending habits and so on.
Customer segmentation allows a company to target specific groups of customers effectively and allocate marketing resources to best effect.”1
Segmentation procedures typically include deciding on an underlying driver differentiating groups of customers on a number of criteria such as:
- Behaviors: purchase patterns
- Attitudes: how/what they think about competitive brands
- Occasions: when they choose or consume a brand
- Modes: overall approaches to how they make decisions
Next comes the decision on how data will be gathered and analyzed, and once completed, how the findings will be operationalized within an organization.
This process is often the territory of large companies with significant budgets to follow through the above approaches with appropriate rigor.
Smaller organizations take this on from time to time as well in an effort to compete with their larger brethren, but it does stretch the limits of their marketing/ research budgets.
However, we strongly believe smaller organizations can still achieve a deep understanding of customers leading to a highly effective segmentation scheme for a fraction of the cost and time. We have developed one such method, which we refer to as Creative Segmentation.
Very simply, Creative Segmentation is any attempt to segment a customer base(s) by applying creative thinking that feeds into a regimented research process. Furthermore, using the Malcolm Gladwell vernacular, Creative Segmentation turns a disadvantage (smaller company with limited resources and budget) into an advantage.
Creative Segmentation is relatively easy to understand and perform. If anyone is familiar with the George Clooney movie “Up in the Air,” or just if you travel a great deal, there is a form of segmentation we all conduct upon entering a security line at the airport. We have a tendency to look at who is in each line ahead of us and make a judgment call as to which line looks like it will move the fastest.
While this is an example of a simple, qualitative segmentation, the critical learning here is we are already experienced in mentally running through several scenarios from our past, uncovering key insights and then making a current assessment.
Does this mean Creative Segmentation is not quantitatively rooted? On the contrary: many times this approach employs secondary data where the sample sizes and data sets are larger than those in many proprietary studies.
Our approach is an insight-driven segmentation that starts with the customer/ consumer purchase decision and then works its way backwards to customer data allowing us to define customer segments and predict purchases.
Often times a variety of key characteristics work together to predict purchase. When someone is in the market to buy an expensive sports car, it is clear that a mix of demographic (upper income, likely suburban or comfortable country, have a three-car garage), attitudinal (enjoy driving, going fast), behavioral (have purchased sports cars in the past, always attend local auto shows) and media usage (avidly read automotive magazines) factors all play a key role.
Each of these characteristics alone may not predict purchase all that well. But when woven together, they become a powerful basis of customer segmentation with the additional benefit of helping us decide the appropriate vehicles in which to reach them and possibly the right type of message to convey as well.
This is because while each piece is important, only in conjunction do they have true value. I have to be a car enthusiast. I need to keep up on the latest automotive trends. It also helps if I have a history of buying high-performance sports cars and have lots of garage space to keep my “baby” dry on rainy or snowy days. But most of all I have to have the means to purchase this type of car.
The following section provides two examples of creative, insight-driven segmentation. In each case, we reveal the challenge faced and how we employed a Creative Segmentation solution that led to actionable output for marketing.
Building Segmentation on a Limited Budget — Secondary Data, Customized Solution
“For a multitude of reasons, it can be difficult for clients to share internal data that is critical for actionable segmentation. An effective segmentation strategy can still be devised using an overabundance of 3rd party marketing data that is readily available to the database marketing community. By combining this rich data with a clear understanding of the client’s objectives, we can deliver a jumpstart to a diverse marketing program that can be easily tested, measured, challenged, and improved.”
— David Nugent, VP Data Solutions, American Spirit Corporation
Company A: A shopper-marketing agency
The challenge: A major agency competitor had just completed a traditional segmentation around consumers who use their mobile devices to shop. This was a very extensive study, which provided detailed profiles about different types of user segments. It was a mindset segmentation with a behavioral overlay.
A situation like this can be daunting for any smaller agency like Company A. So what was done to remain competitive in the marketplace?
Since the budget wasn’t available to conduct a study like Company A’s competitor, a creative approach was needed.
The solution: Company A sought to combine learning about shopper behavior with how consumers use their mobile devices. The Simmons/Experian database served as the major resource since it offered both a shopper segmentation as well as a mobile user segmentation.2 These two segmentations were cross-tabbed against each other. Company A realized that by doing this there was the inherent risk that the two segmentations may (and likely would) not align — especially since there were six Shopper segments and five Mobile segments.
Figure 1 provides a visual example of Company A’s hypothesis regarding the result of combining these two data segmentations (Ideal Solution). Upon analyzing the data closely, the hypothesis had to be refined and is shown as the Reality Solution in Figure 2.
To reach this final outcome, a ‘Light Switch’ Method was employed — i.e. turning factors/attributes on and off to see how each affects the different groupings. By doing this, there were three factors that were key distinctions between the segments — how the target used their cell phones, their approach to shopping overall and how they viewed pricing/deals (Figure 3). Verification of this came by eliminating the factor/ attribute and examining how it affected the defined segments. In effect, this ensured that the segments were better defined and more mutually exclusive.
The final output included “Key Themes” (Figure 4) for each segment, along with more detailed profiles about drivers and descriptors for each as well. The end result was an actionable segmentation that could be leveraged by Company A, in a way that would better position them against competition.
All of this was completed at a small fraction of the cost of the study conducted by Company A’s major competitor. There was also an added advantage in that the segmentation could more easily be updated, since Simmons refreshes their data twice a year.
Getting Creative With Data — Database Manipulation
Company X: Real estate investment company
The challenge: How can Company X effectively reach millionaires? More specifically, the target was high net-worth investors. And to add to the task, there was a need to identify those high net-worth investors that would be most likely to incorporate a new real estate investment tool into their portfolio. In effect, the task was to segment a group that comprised no more than 1-3 percent of the total U.S. population.
Company X’s goal was to increase active participation in a new online crowdsourced funding mechanism for real estate.
The solution: Utilizing data mining techniques against third-party syndicated data, four consumer segments were identified that demonstrated one or more of the following characteristics:
1. Interest in real estate
2. High net-worth or disposable income
3. Keen on staying current with financial and investment trends
Taking advantage of access to hundreds of demographic and psychographic attributes of U.S. consumers through the Mosaic Consumer Segmentation Tool, Company X systematically isolated traits they felt best described ideal prospects. Mosaic is a “household-based segmentation system that classifies all U.S. households and neighborhoods into 60 unique Mosaic USA types and 12 groupings that share similar demographic and socioeconomic characteristics.”3
To further customize the segmentation, Company X identified key differentiating traits related to the new product and behaviors/attitudes that the target should exhibit.
Some of the traits isolated included:
- Website visited: Business & Finance – Real Estate
- Highest level of discretionary spending
- Presence of cash management accounts
- Active investor
- Agreement with “I find advertising for financial services to be interesting”
- Agreement with “I read the financial pages of my newspaper”
- Agreement with “I’ll pay any price for good financial advice”
Within these traits, Company X identified Mosaic consumer segments that indexed the highest and recognized crossover trends.
The examination of crossover trends revealed four dominant consumer segments:
- Comfortable and Coasting: Upscale retirees and empty-nesters in comfortable communities
- Platinum Prosperity: Wealthy and established empty nest couples residing in suburban and in-town homes
- Status Seeking Singles: Younger, upwardly mobile singles living in mid-upscale metro areas balancing work and leisure lifestyles
- Babies Living in Bliss: Established families with active lives in affluent suburbia
The final step was to comb through the schema to eliminate borderline households from each segment.
Key Takeaways for Designing Creative Segmentation Solutions
By now we hope you have a good understanding about more creative solutions to segmentation. This approach works well when:
- There is a limited budget: The two techniques discussed in this paper (Secondary Data/Customized Solution and Database Manipulation) are examples of how segmentations can be completed on a budget. The outcome of these segmentations can be directly applied to marketing challenges, and the learnings can help to differentiate a brand from its competitors. More importantly, it can narrow the gap between those companies with large research budgets and those with limited funds.
- There is a difficult challenge that traditional segmentation techniques cannot easily address: More traditional methodologies are continually being challenged to accomplish more and more. At some point, it becomes more advantageous to explore new research techniques. This paper highlights one such technique (Secondary Data/Customized Solution) not bound by attitude, approach, occasion or behavior in isolation.
In summary, this approach, if effectively implemented, allows the Davids to successfully take on the Goliaths with less cost, time and resource commitments. [END]
- Rouse, Margaret and Tim Ehrens. “Customer Segmentation Definition.” SearchCRM.com. TechTarget, May 2015.
- “Delivering the Mindset of the American Consumer.” Experian.com. Simmons® National Consumer Study, 2015.
- “Consumer Lifestyle Segmentation for the United States.” Experian.com. Mosaic® USA, 2009.
(American Spirit Data Solutions offers access to Mosaic cluster data and the broader Experian database.)