Segmentation Made Simple
You’re new in your job. Or the competitive landscape has changed. Whatever the reason, you need to understand which customers to target for purposes of product development, marketing, and customer experience design. So you call one of the big firms that specialize in segmentation and brand tracking. They tell you identifying and following your target audience requires a study that requires 12 weeks and costs a couple hundred thousand dollars or more. Once you recover your composure, you ask: ‘Why does it have to be so complicated?’ Sound familiar?
Three Tips for Keeping It Simple:
As you consider alternative measures, data sources, and analytic approaches, keep these questions in mind:
1. What behavior do we really need to understand?
2. How much complexity can my managers, agencies and staff really handle?
3. What are all the potential uses of the data?
Segmentation is a vitally important tool for any business; no business can afford to move ahead without a clear idea of its likely customer, their needs and their brand alternatives. But complicated segmentation research can makes things more complicated. So here are some thoughts about what marketers really need to know.
What is Segmentation Research?
At heart, segmentation is about identifying groups of people who respond differently to marketing so that your efforts can be focused on the most attractive prospects. Segmentation first became fashionable in the 70’s just as mass brands were beginning to give way to ‘niche’ markets and brand extensions. At the same time, media fragmentation and new media research techniques were providing more precise targeting options. The question became, ‘Why treat everyone the same?’ It wasn’t long before everyone was doing ‘segmentation research’.
Segmentation research is not one technique or approach, but a term that covers a broad range of tools for defining important differences between clusters of customers – differences that matter to how they buy, what they buy, how much they spend and what they spend it on. Segmentation research can reveal unmet needs or provide insights about how to approach different people differently for the same brand or product.
One of our earliest experiences with segmentation research, was a study of the cat food market for HJ Heinz. (For perspective, this research was a mail survey every table was drawn by hand, that’s how long ago it was.) In the 60’s and 70’s, cat owners had two choices – ‘wet’ (canned) cat food or ‘dry’ cat food. The study revealed that the users (eh, buyers) of these two products were very different, especially in how they related to their cats! There was a segment of wet food users that were so passionate about their cats; they literally threw half of every can away because it wasn’t ‘fresh’ enough to provide a second meal for their darling kitty. The result? The introduction of 3- oz cans of wet cat food and the birth of an entirely new segment of ‘super premium’ cat food.
What Should be Measured?
Key Segmentation Measures:
- Demographic characteristics
- Buying behavior
- Brand Attributes for you and key brand
- Your Brand’s Relative value
- Your Brand’s Relative superiority
- Category motivators
- Purchase Intent
The first step in any segmentation study is to explore and define buying behavior. By understanding how different people behave in a category –– what they buy, how they buy, what they spend, how often they spend -- it is possible to forecast revenues and profitability by segment, and prioritize the customers that can make the biggest difference to the business.
A few years ago, we helped a major retailer define a ‘heavy online shopper’. The actual definition varied considerably from their going-in beliefs. Heavy online shoppers shopped online more frequently than previously imagined -- and they shopped more offline, too. In fact, a heavy online shopper was a shopping fanatic. This had important implications for their multi-channel marketing efforts. By targeting heavy online shoppers, they were also helping their store business, and learning about ways to connect the two to capture a larger share of the heavy shopper’s total wallet.
The second step is to measure additional information that can provide insights as to why customers behave differently. This information should include category attitudes, purchase motivations, brand perceptions, and detailed demographic profiles. By understanding the differences between groups of customers on a variety of dimensions, segmentation research also brings important similarities into focus. In researching online travel purchasers, we learned that people who buy leisure travel most frequently have many of the same attitudes toward travel planning as less frequent travelers. They both care passionately about XYZ. This aha allowed us to build communications and messaging aimed at the frequent leisure travelers that also appealed to those who travel less often. So, the company was able to leverage its marketing across more users and prospects.
How Should the Data Be Analyzed?
Cluster analysis can be used define groups on several dimensions simultaneously. This type of analysis is very powerful, but can be tricky to interpret. Our recommendation is to begin with simple crosstabs to determine what groupings make the most sense, then decide if it multivariate analysis is needed. We often find that managers respond well to cross tab tables (indexes and tables) because they are telegraphic and easy to understand.
A common way to use crosstab analysis is to compare the characteristics of a brand’s most frequent customers to less frequent customers to address the question ‘What do we have to do to make less frequent customers more like the frequent ones?’ One of our retail clients was surprised to learn the only meaningful difference between their most frequent customers and customers who come once or twice a year is their perceptions of the store’s merchandise selection. In nearly every other way, they are similar – demographically, in terms of category involvement, even the other stores they shopped. By better communicating its depth and breadth of assortment, this client will attract less frequent customers to shop more often.
Getting the Most from Your Study
The benefits of segmentation go beyond product development and messaging, so it is advisable to consider how you plan to apply the findings when designing the study. Here are just a few possible applications of well-designed segmentation research, beyond the obvious ones of product development, marketing and customer experience design:
- A Screening Tool for Online Focus Groups: Segmentation research identifies key customers. Once identified, you can invite them to participate in follow up research. (Note: This requires permission from outside panel providers, but is easy to do with members of an internal database.)
- Brand tracking. Many of the questions in a segmentation study are the same ones used to track brand perceptions and behavior. A segmentation study can double as the first way of on-going research.
- Organizing marketing and sales functions. Some companies find that it is better to organize teams around customer segments rather than products alone. This can contribute to a more customer centric organization.
- A way to analyze and forecast revenue growth. By creating models based on predictions and trends in consumer behavior, opportunities can be spotted more easily.