Perceived Differentiation

What it is

Brand strategy has long leaned on the idea that a brand wins by being seen as different from its rivals. Perceived differentiation turns that idea into something a survey can check. Buyers are asked whether a brand is different from the others in its category, whether it is unique, or either, and the share who say yes is the score.

The measured version comes from Romaniuk, Sharp and Ehrenberg, whose 2007 paper in the Australasian Marketing Journal examined perceived differentiation across many categories and two countries. Their abstract reports that perceived differentiation was low, that customers still bought, and that distinctiveness belongs at the centre of brand strategy. Romaniuk is first author, and the credit for the measured finding goes to her first.

How Brands Grow, by Byron Sharp, carries the same finding into a book-length argument. It reports figures across sixteen categories, with category averages of about 11 per cent of customers seeing a brand as different, about 10 per cent as unique and about 17 per cent as either. Those figures are the book's tables, which reproduce the paper's work. They are quoted here as the book reports them.

In effect

The finding matters for a branding engagement because it changes what the early stages are asked to deliver. If buyers see few meaningful differences between rivals, a positioning document that rests entirely on a claim of difference is working against the grain of how people perceive the category. The alternative the research points to is distinctiveness, meaning cues that let buyers recognise a brand quickly, which is the subject of the Distinctive Assets page.

The book checks its own measure in three ways. It shows that brands which are genuinely different do score high on the questions. It shows the ordering that would be expected between categories. And it repeats the measure on a five-point scale, where the average is about 2.5 and no brand averages as high as 4. These checks support the claim that the survey can detect difference when it exists, so that low scores are not an artefact of the question.

The book also prints a fourth column, the share of customers who see at least one brand in the category as different or unique. It averages about 54 per cent. Read with the first three columns, it says that the finding concerns individual brands. In many categories a good share of buyers can point to some brand they see as different, and the typical brand is not that one.

What it does not say

Two things should be kept apart. The measured finding is that perceived differentiation is low and varies little between rivals. The conclusion that differentiation does not pay is a separate claim, and the tables in the book carry no price, margin or return to support it.

Researchers working with brand-tracking data have reported price premiums for brands seen as meaningfully different. Those reports appear in trade sources, rest on proprietary data, and leave the direction of cause open, since larger brands might be seen as more different because they are larger. The two sides are measuring different outcomes, perception and share on one hand, price on the other, and the question is not settled.

Practitioners also describe a middle position, in which a brand owns an attribute or association more fully than its rivals do and treats distinctiveness as a companion to that aim. That position is argued in the trade press and is not a finding. The 2007 paper's abstract reports no price result.

---