The Blake Project, the brand consultancy behind Branding Strategy Insider, delivers interactive brand education workshops and keynote speeches designed to align marketers on essential concepts in brand management and empower them to release the full potential of the brands they manage.
The worst data miscalculation in brand marketing is also the easiest to fix, even though it requires a new way of looking at market opportunities for brands.
It’s a miscalculation made with good intentions. Brand marketers put a high priority on focus, so complex markets have to be boiled down to their essence. Focus through simplification is smart, but oftentimes it takes a brand in the wrong direction. So it’s worth taking a second look at bringing focus to brand marketing.
As we all learned in Marketing Research 101, there are two broad, overarching ways of looking at research data – measures of central tendency and measures of variance. Admittedly, this lingo doesn’t easily roll off the tongue, so to put it another way, for any given dataset, you can look at averages or differences. You can look at the average consumer or you can look at varieties of consumers.
There are many ways of calculating averages and differences, so each category of measures includes lots of statistical techniques. But every statistic is either a measure of central tendency or a measure of variance.
The priority on focus in brand marketing means an over-reliance on measures of central tendency and an under-appreciation of measures of variance. Brand marketers want to focus on the average because that gives them a way to simplify and concentrate their resources. Brand marketers fear getting paralyzed by the ambiguity and indecisiveness often present in differences. It is hard to know what to focus on when options, scenarios, choices and situations seem endlessly numerous and diverse. Even when brand marketers factor in differences, they do so by focusing on multiple averages, which is an incomplete way of accounting for differences.
In short, brand marketers don’t realize full value of the data they collect about their markets and consumers because they are over-reliant on measures of central tendency to focus their decision-making. This is easy to fix, and doing so provides a richer understanding of how to focus resources and strategy.
Two examples are illustrative.
Example 1: Segmentation Personas. Market segmentation studies are, perhaps, the single most important type of strategic research. These studies define the structure of the market, identify high priority targets and tease out positioning platforms for existing and new brands. The output of a segmentation study is a set of consumer segments, and one of the most common ways in which each segment is summarized and understood is in the form of a persona, or a typical consumer that embodies the essence of what that segment is all about.
Personas are great tools for presentation, ideation and illustration. But they are averages, and thus represent the central tendency of a segment not the full diversity of a segment. In fact, the dirty little secret of segmentation research is that at the margins, segments get fuzzy and may even overlap to some degree. Segment members are more alike one another than members of other segments, but the strength of similarity is not the same for all segment members.
Segmentations take the variance of the entire sample and reduce it by creating discreet groups that are easier to tackle, but these groups still retain a lot of variance themselves that is not captured by personas. For example, if women constitute 65 percent of a segment, the gender of the persona will be female, but what then of the third who are men?
The solution is simple. Instead of an individual consumer persona, develop persona families. That means multiple personas per segment, but focus is retained by presenting these personas as families who share basic family traits while manifesting those commonalities in distinct ways. Thus, each segment is understood as a family rather than as an individual. Brand marketers can focus on one type of family group rather than one sort of individual. The family personas are still averages, but unlike individual personas they do not over-simplify and ignore the additional information about a segment that might provide useful insights for resource allocation and strategic planning.
Example 2: Consumer opinion. There are many types of research in which brand marketers only want to know the majority opinion – taste tests, copy tests, consumer confidence, user research, formulation optimization, design preference research, values tracking, and more. In these cases, majority rules. When brand marketers want to know where the largest part of the market lies, they look to the majority opinion.
But the majority opinion is often used to describe consumers as a whole as well, with phrases like “consumers prefer…” or “consumer sentiment is…” or “consumers believe…” This is attributing the opinions of most consumers to all consumers. When the biggest part of the market is what matters most to brand marketers, it is perfectly fine to focus on majority opinion. But this should not also be used to characterize everyone. Doing so obscures the differences that are, in fact, the truer characteristic of all consumers.
The solution here is two-fold. First, majority opinions should be used with more specific purpose. Typically, it is not the majority that matters but a majority that also crosses some critical threshold. It is the threshold that matters not the majority per se. For example, when tracking value shifts, 51 percent is qualitatively different than 80 percent even though, as majorities, both are the same quantitatively. If all you’re looking for is a majority opinion, 51 percent and 80 percent are no different. But at some point between 51 and 80 percent, as more and more consumers hold a certain value, the qualitative feel of the marketplace begins to change. It is this threshold that matters more – not when a majority is reached but when the character of the marketplace starts to have a different feel.
Second, questions that measure preference or beliefs should be used as classification questions not simply as opinion questions. For example, if 65 percent of consumers prefer A to B or are confident in the future or hold X as a strong value, that means roughly one-third do not. These consumers should be examined separately, or even excluded entirely. In addition to discovering majority opinion, such a question has classified consumers into contrasting groups that should be examined and assessed independently. This is much more informative than glossing over this crucial difference by speaking of consumers in total as if they are identical with the majority.
The biggest imperative for making greater use of measures of variance is that the marketplace itself is becoming increasingly diverse. The macro trend of the future is more difference not less. Averages will be less useful metrics for a world of difference.
More and more markets are characterized by the demographic diversity of minority-majority populations. The unifying, homogenizing dynamic of globalization is giving way to a more complicated dynamic of fiercely held local values and tastes alongside globalization. Marketing technologies and databases are rapidly overcoming the technical challenges to individually personalized products, experiences and messages.
Difference is the future, so brand marketers must approach it with a perspective on the data that brings these differences into view.
Contributed to Branding Strategy Insider by: J. Walker Smith, Executive Chairman, The Futures Company
Sponsored By: The Brand Positioning Workshop