“Genius May Have Its Limitations, but Stupidity Is Not Thus Handicapped”
I’m not a Luddite when it comes to new ways of assessing brands, advertising, and marketing programs. I’m awfully proud to be able to say that my company has been at the forefront of leading-edge marketing and brand solutions. Not only have we spent the past 20-plus years developing predictive and leading-indicator metrics, but we’ve also been one of the vanguard consultancies that have invested intellectual capital investigating every new media touch point that the innovative minds of technologists and marketers have been able to envision.
And yes, we have our dark side and been known to label certain marketing programs “field of dreams” approaches: You know, the kind of program that was based on “this is really cool, so if we build it, they will buy it” logic. It seems simplistic to say it, but just because something is new or the flavor of the month, doesn’t mean that it is going to produce results. We define “results” in this case as positive behavior toward the brand, sales, profits—stuff that warms the hearts of shareholders. So we try to be evenhanded in our judgments when companies ascribe “success” by the number of people who, for example, visit a Website or watch a rerun of a commercial on their computer. They are metrics, to be sure, but very few of them correlate to the previously referred-to “results.”
Even more amazing are the lengths some marketers and researchers will go to create innovative metrics—even if they are poorly designed and don’t measure much of anything at all. In cases like that, there is no safety in numbers.
But facts are stubborn things and numbers more pliant, and that thought came to mind when an alert client e-mailed us a report (with the subject heading: “Thought You Could Use a Good Laugh!”) of a survey that counts the number of times the average American mentions a specific brand in conversation. They call it “talk share.” Sounds interesting, doesn’t it? Redolent of the well-established “share of market.” So very evocative of 21st-century word-of-mouth buzz, right?
Let’s see. We are not mentioning the perpetrators of this “research.” The quality of mercy is not strained, even in our business.
The study’s summary reports that they asked 100 people, 13-65 years of age, to recall the brands they referred to when they chatted with family, friends, and co-workers over the course of a week’s period of time. Using that number, the study asserts, advertisers can configure a brand’s “talk share” relative to its brand share. The higher the talk share, the more the market share will grow. Quod erat demonstrandum, right? Wrong. Let’s take a look at the component parts of this “research,” the design, and the output, and see what we actually have.
Okay, it represents a wide age spectrum, but even with nontraditional and “new” approaches, clients like to look at traditional age breaks, which by our count would be five (13-17 years old, 18-24, 25-34, 35-49, and 50-plus), which works out to a qualitative count of 20 respondents per group. Hardly a large enough number upon which to base any kind of share configuration. Hardly enough for a couple of focus groups, even.
Next the study claims to represent the “average American,” which has an undeniably egalitarian appeal, since it’s every marketer’s burning desire to be all things to all consumers. But sadly, this is not the case. Different products and different brands are configured for and attract different demographic segments. It is reasonable to assume that there will not only be different linguistic patterns but also subject-matter variations by age group, no matter what the sample size. Older consumers are more likely to discuss things like, say, prescription-drug brands, while younger consumers are more likely to have chats regarding popular culture or fashion brands. How do you reconcile that in your “talk share” calculations?
And what of controls regarding category or brand usage? There isn’t a marketer alive (traditional or nontraditional) who doesn’t have the most basic screener in place so that he absolutely knows that he’s talking to the “right” respondent. And what of the context in which the brand was actually mentioned? It’s one thing if respondents are talking about the brand in the most glowing of terms. But how does the talk-share approach differentiate between that and someone who is criticizing a brand? Does a conversation that involves complaining about a brand have as much weight as one from a satisfied consumer? Only half a talk-share point? (Remember, there’s no actual record of the conversations. All of this is based upon recall, so you think about which brands you mentioned five days ago.)
The perpetrators of this approach might have benefited from a conversation with brands like Martha Stewart, General Motors, Snickers, and a litany of others that in their time got plenty of prattle but very few profits. There may be a certain truth to the old public relations dictum “it doesn’t matter what they say about you as long as they spell your name right,” but that doesn’t fly when you’re trying to correlate brand mentions with market share.
We applaud our colleagues’ efforts to develop innovative, leading-indicator metrics and ROI assessments to meet the demands of the 21st-century marketplace and today’s bionic consumers. But sometimes you innovate and end up getting it wrong. And when that happens one should just consign such efforts to the trash, admit it didn’t work, and move on. It’s a complex marketplace out there, so we are patient with stupidity—but not with those who are so proud of it that they publish proof of it.
Robert Passikoff, Ph.D., is founder/president of New York-based marketing firm Brand Keys and is the author of “Predicting Market Success: New Ways to Measure Customer Loyalty and Engage Consumers With Your Brand.”
Other articles by Robert Passikoff:
The New Four P’s: Promoting Predictive Promotion Planning
Self-Interest Is the Anesthetic That Dulls Innovation
Seven Brand and Marketing Trends for 2007
Consumer-Generated Content: Let Yourself Go