Brand Knowledge

Consumers store knowledge about brands in their memories as associations linked to the brand name (eg, stylish, good value, red stripe).  This brand knowledge influences the chance that the brand will be accessible in buying situations, as well as whether or not the brand is judged acceptable for that purchase occasion.

Most of the time we pay attention to individual brand associations.  Is my brand considered good value?  Or refreshing?  However, another dimension is the “richness” of the network of associations, or how many memories are linked to the brand name.  If we think about all associations across all customers, this is the brand’s total knowledge. These associations form the basis of the brand’s salience and so constitute a large part of its equity.  They are at the heart of probably all commercial brand equity monitors.

A small stream of research has found that the size of the brand’s network of associations is positively related to brand choice, irrespective of the nature or importance of the specific attributes that make up the network.  This supports the theory that we retrieve information through ‘activation’ spreading down links in memory, so the more links, the better the chance of retrieval of the brand name (Collins and Loftus 1975).

However little is known about the distribution of this brand level knowledge across consumers.  Is it concentrated in the minds of a few highly brand familiar consumers?  Or is it spread out evenly over the entire customer base?  Does this vary for different brands, depending on whether the brand is large or small, or the nature of the marketing strategies employed for the brand in the past?   

One technique used to analyse the distribution of buying data is the Pareto share.  This is a relatively simple metric to calculate that provides a basis for comparing the concentration of (typically) sales across brands within a market.  Here we apply this same calculation to brand knowledge data.

Pareto Share

In the marketing literature, Pareto share is defined as the percentage of sales coming from the top 20% of a brand’s customers (Schmittlein et al. 1993), this is typically less than 60% (Romaniuk & Sharp 2007).

Just as there are heavy and light buyers of the brand, there are those who hold a very rich set of associations and those who hold very few.  However no studies to date have investigated how brand knowledge is distributed across a brand’s customer base.  By drawing an analogy between purchase frequency (the number of times the brand is bought) and the number of associations stated about a brand (eg the number of associations customers have for Coca-Cola), a Pareto share  can be calculated.  Specifically we are examining: how concentrated is the distribution of a brand’s knowledge?

Method

For this research we drew on associations measured with a ‘free choice, pick any brand-attribute association’ measure.  This approach aims to measure the presence of a link between a brand and an attribute and is commonly used in market research to measure brand associations.   Attributes were either written or verbally provided to respondents, who were asked to indicate which brands, if any, they associated with each of the attributes.  The number of associations for each person were then tallied and all of a brand’s customers ranked from the highest to the lowest, based on their number of brand associations.  Calculations were performed to find the percentage of total brand associations (ie, summed across every customer) that came from the highest 20%, when ranked by knowledge levels.  An example of the interim calculations is available in Table 1.

This analysis was conducted for each of 208 brands in the 28 markets.  Examples of the results from the top 5 and bottom 5 brands in two markets (Soft drinks and Electronics) are shown in Table 2, where we can see that the Pareto share for Coca-Cola is 47%, while for Sony it is 48%.

Results

A wide range of markets were included in this study.  16 were consumer packaged goods, 3 were durables, 7 were services and 2 were retail stores. Despite the obvious differences between markets, the results were remarkably consistent:

• The Pareto shares within markets did not vary much, with deviations typically of the order of 4 or 5 percentage points.
• The actual size of the Pareto share is very similar to that found for sales, with the top 20% holding just less than half of the brand’s knowledge.
• The vast majority of small share brands had similar Pareto shares to the larger brands in the market.
• There was a noticeable effect that brands with penetrations less than 10% did sometimes have higher Pareto shares.  These instances were few, and most likely to be explained by random sampling variation.
• The key difference between the brands in the market appears to simply be how many customers know/recall anything.

The similarities within any market become even more apparent when we compare larger and smaller brands within the same market.  For example in the mineral water market, Evian (47%) had a similar Pareto share to a brand with half its penetration, Spa (46%).  Likewise, Heinz soups (43%) had a similar Pareto share to New Covent Garden soups (43%) despite a nine-fold difference in size.  In service markets the pattern was also similar with St. George and Commonwealth banks having the same Pareto knowledge share (33%) despite large differences in market share.

When was Pareto share more concentrated?

There were a few purely methodological scenarios when Pareto share did appear more concentrated.  These were when more respondents with very little knowledge were included in the analyses such as:

• If the customer classification was ‘ever used’ then concentration was higher than if it was ‘used in the last 3 months’.
• If non-customers of the brand are added to the analysis, as these people know very little about the brand.  This effect was most noticeable for small brands,  where sometimes only 20% of category consumers know anything about the brand – thus giving that brand a concentration of 100%.   In contrast, the concentration of larger brands varied little when their non-customers were added to the analysis.
• Similarly, the Pareto share is affected by the more attributes in the set because this provides a greater opportunity for those with richer associations to express these and distinguish themselves from people who know very little, resulting in higher Pareto share.

Discussion

Brand managers can now refer to the light half/heavy half of the brand’s knowledge base.  They can be reasonably confident that this is applicable not only to their brand, but to other brands in the market.  This new empirical generalization holds across brands in 28 different markets from soft drinks, to soups, from banking, to tourism – across different times and countries.

Strategic Implications

If there had been no pattern in the Pareto shares, this would have suggested that this metric was of little value.  However, the consistency of the Pareto share within and across markets has considerably more important implications.  It suggests that this Pareto share is something of a “constraint” that marketers have to work with, regardless of the history of the brand, the share the brand is at currently, or  the share objective.  Therefore, when choosing between brand knowledge building strategies, marketers would be best advised to choose something that worked with this “constraint” rather than adopt a strategy that tried to alter it.   Strategies that  would have the effect of changing the concentration of brand knowledge, (for example targeting a small group of the customer base and working to substantially increase their knowledge levels) will be less likely to be successful.  We simply don’t see situations where large brands have highly concentrated customer bases.  Large brands have more ‘heavy knowledge holders’ but these are counter-balanced by more ‘light knowledge holders’.    There are simply brands that are more or less famous.

Implications for share maintenance

Every brand has a ‘heavy half’, that 20% of users who hold almost half of its total knowledge.  These customers are important for a brand to have.  However, the brand also has the remaining 50% of knowledge distributed over 80% of its customer base.   These are people who buy the brand currently, but do not recall a great deal of knowledge about the brand.   They know less or find it harder to retrieve their knowledge.  These people aren’t far away from forgetting the brand altogether.  This will be evident not only in market research surveys, but also in buying situations.  This means that it is important for marketing communications and advertising to reach the wider customer base to maintain the accessibility of this current customer knowledge, particularly when competitors are also advertising.  Should these customers fail to retrieve the information they have, then this will have negative consequences for the subsequent sales of the brand.

Implications for brand growth

The difference between larger and smaller brands provides insight into how a small brand will (or will not) change, should it achieve any share growth objectives.  This research includes some of the world’s largest and most successful brands in Coca-Cola, Sony, Microsoft and Heinz, and these brands had the same concentration of brand knowledge as brands one-tenth their size.  Therefore, brands can vary 10 fold in the number of customers they have, but this will not change the distribution of the knowledge within each brand’s customer base.  This is despite differing levels of brand performance, and the myriad of strategies that would have been used to achieve this performance.

Implications for research

Our new empirical generalisation is a blow to commercial brand equity measures that portray brands as stronger or weaker, based on perceptual responses.  In terms of depth of knowledge, it makes sense to talk of larger, more salient brands compared to smaller, less salient brands.  However, we find no support for the idea that it is important (or even possible) to cultivate a disproportionately greater core of knowledgeable brand buyers for whom the brand is highly salient. Therefore brand equity models that encourage the building and monitoring of the presence of these brand ‘champions’ are misleading marketers about the factors that separate more successful brands from less successful.