QUESTION: We’ve been using the UAI (Usage, Attitude, Image) study for our brand positioning search. That’s been for more than a decade now. But in the past 3 years, it has been of no help. Its Image section didn’t identify our brand’s differentiator versus competition.
Our major brand is a chocolate drink for children. Over the years, moms have maintained chocolate taste and flavor as their most important attribute for a chocolate drink for kids. Our UAI of the past 3 years have rated our brand as parity with the leader brand on this most important attribute. Our other competitors were rated also as parity on chocolate taste and flavor. If we’re all parity, how come the leader brand continues to have the highest market share and our other competitor brands showed differing market shares including our brand?
Two of our brand managers attended a brand positioning and brand equity seminar. When they came back from the seminar, they told us that their seminar facilitator told them that the UAI is now practically useless because “it cannot get in-depth qualitative data that the positioning decision requires.” Is this true? So should we stop commissioning our annual UAI? What research should we use in its place?
Answer: You’ve asked at least 3 different though related questions. We will tackle each question in turn.
The first question—“why is our brand’s parity positioning different from its market share relative to competitors who are all parity with each other in positioning?”—has a known standard and basic answer.
Your brand’s market share is a function of the entire marketing mix, or the 4 Ps. Positioning is just one of the marketing mix levers. So, there’s no logical basis for you to expect a one-to-one correspondence between market share and positioning ranking.
Your first question actually has a subquestion: “How can different competing brands with differing market share have parity positioning ratings?” To uncover a meaningful answer, you have to understand what can go wrong in a positioning research via the UAI Image questions and the consumer responses to those questions.
The UAI-based positioning model asks 2 sets of questions. The first is about your consumers’ priority of important qualities or product attributes for a chocolate drink for kids. The second set asks your consumers how they rate four or five competing brands including your own on those product attributes you just rated in importance. If your brand is rated superior over competitors on a top priority important attribute, then it has a superior positioning. However, if the rating is as good as competitors, then your brand has parity positioning. And if the rating is not as good as competitors, then your brand has an inferior positioning. This has happened to four of my clients, and all go into denial and blame the research, saying something’s wrong with it.
As you can guess, your positioning research can go wrong in at least two ways: On the product category attributes rated in importance, and on the ratings of the four or five brands including yours on the priority attributes. The most common error in the importance rating of product attributes is to present the UAI respondents a “generic” rather than a “specific” attribute. An example of a generic attribute is, what your UAI has found over the years as the most important attribute of chocolate drink for kids, namely, “chocolate taste and flavor.”
Suppose your brand were rated “not as good as other brands” on this attribute. If you ask your product development (PD) to correct this inferior rating, you can be sure that PD will ask you: “In what kind of chocolate taste and flavor is our brand rated lower?”
There are many kinds. Generic attributes are not actionable. So your UAI should be provided with specific attributes, and not generic ones.
What about in the brand ratings on the important attributes? The most common problem here, among others, is the use of “2 top-box scores” or the average rating. To start with, the brand rating is along a 5-point scale: (5) rate definitely better on this attribute than others, (4) rate generally better than …, (3) rate just as good as …, (2) rate generally not as good as …, and (1) rate definitely not as good as.…
Which rating is cleaner than the others? It’s the top-box score, the extremes, i.e., the rate definitely better, and the rate definitely not as good as. A consumer who rates your brand as “generally better” is rating for the sake of just giving an answer, or just to be polite, and the like. It’s a multi-motivated rating and is not as true as the consumer saying she “definitely” finds your brand as better. So avoid combining the 2 top-box scores of “% definitely …” plus “% generally.…” It’s contaminated data. It is for this same reason that you should avoid the average rating, which is also a combination of the true and the not-so-true.
The second question is, “Is it true that the UAI cannot help positioning because it cannot provide the in-depth qualitative data that positioning requires?”
In my market research book, I devoted two chapters on UAI, where I showed that the lack of depth in the UAI data is the research user’s fault. As a research user, you allowed the UAI to have no qualitative character. How? That’s because you did not probe any of the positioning responses. As qualitative research, you regard FGD (focus group discussion) or IDI (in-depth interview) to provide in-depth data. They are able to do so because FGD or IDI asks third, fourth and even fifth degree probes. If your UAI probes a positioning response even up to only a second degree, you will find that you have given your UAI those FGD-quality answers. For examples of second degree probes, please see the appropriate chapters of my User-Friendly Marketing Research book.
As to your third question—“What research can replace the UAI for positioning purposes?”—keep in mind that there are six or seven positioning research models to choose from. The UAI-based model is only one of them, although it happens to be the most popular. For the explanations and examples of those other models, please see previous Marketing Rx columns in November and December 2014. You can access these online.
Keep your questions coming. Send them to me at ned.roberto@gmail.com.