How Brands Grow
Page 13
Meaningful differentiation
Differentiation is deemed to be an essential part of marketing strategy. It is taught more like religion than hard-nosed business. Differentiated brands are supposed to inherit all the customers and profits; it's all or nothing and we are told we must 'differentiate or die' (Trout & Rivkin 2000).
Marketing texts use a motivational perspective – they talk of a meaningful, perceived difference that provides buyers with their reason to purchase and be loyal to a brand (Aaker 2001; Kotler 1994). Undifferentiated new entrants to a market are supposed to be likely to fail because no customer is motivated to buy them (Davidson 1976). It is believed that established brands need to maintain their point of difference in order to stay desirable to their customers. Breakthroughs in perceived differentiation – achieved either through product features (e.g. Apple's original, candy-coloured iMac) or image-building advertising (e.g. Marlboro man) – are seen as pathways to growth.
The marketing literature explicitly states that a brand's differentiation has to be perceived by customers and must be valued. This valued difference doesn't have to be a product feature, and can be symbolic or emotional. Some marketing texts even suggest that a brand's differentiation can be based on what, in reality, are meaningless product features, such as Folgers 'flaked coffee crystals' (Carpenter et al 1994) – so long as these features are noticed, believed and valued by consumers. Perception is reality, as many marketing commentators point out. Others like Kotler and Aaker are emphatic that the difference must be meaningful to consumers.
The advertising principle of promoting a 'unique selling proposition' (USP) (Reeves 1961) reflects this theory. Advertising that does not give buyers a reason to buy the product is thought to be ineffective. Although, a few have challenged this orthodoxy by arguing that advertising can work effectively without a USP or a means of persuasion (see Chapter 9).
Yet, in spite of these strong beliefs, or perhaps because of them, very little is known about brand differentiation. Consider these questions:
•How different are brands perceived to be?
•Do buyers need to perceive a meaningful difference to repeatedly buy a brand?
•Are some brands far more differentiated than their rivals? Does this mean their buyers are more loyal? Are these brands more profitable? Are they growing faster?
In spite of all the marketing books that preach differentiation, they offer scant evidence to answer these basic questions.
Real world differentiation
There are very good reasons to doubt the textbook theory of differentiation. Much of the empirical evidence that is presented in this book does not sit well with the theory.
In the real world differentiation undoubtedly exists. The idea that brands are identical commodities is fiction; for a start, they have different names that can be used by buyers to develop loyalties and show preference60. Equally importantly, there is a tremendous amount of situational differentiation, for example:
•this brand is here now, while the others are not
•I know where this bakery is
•I have an account with them
•this one has my size
•this is the only one in red
•I'm in the mood for chocolate
•it was the one I noticed
•it was the one that came to mind
These situational differences affect all brands; this means that they all enjoy some differentiation. But is there also differentiation at the brand level? Do some buyers (across time and buying situations) see and value a difference?
The first thing that should make us question the importance of differentiation is that loyalty does not vary much between rival brands (see Chapter 2). Loyalty definitely exists, but it is a characteristic of consumer behaviour, rather than being driven by brand differentiation. This means that all brands should benefit from loyalty. But if there is substantial differentiation at the brand level, then why doesn't loyalty differ much between brands? Why are niche brands, with small but highly loyal customer bases, not commonplace?
The second thing we should consider is that if brand-level differentiation is strong then a brand should appeal more to a particular type of person who particularly values the brand's style of differentiated offer. Yet, as discussed in Chapter 5, brand-user profiles don't differ much – competing brands in a category sell to very similar customer bases (Kennedy & Ehrenberg 2000; Kennedy & Ehrenberg 2001a). Brands with vastly different prices (and quality) have different user profiles; expensive brands tend to be bought by wealthier people. But within their competitive set, the brands' user bases are similar: Versace's buyers are similar to Gucci's, and Nike's buyers are similar to those who buy Adidas's.
Differentiation theory suggests there should be a great deal of market partitioning (where brands share more or fewer customers than they would be expected to according to their respective market shares). Brands that are very differentiated from one another should share very few customers; brands that are positioned in a similar manner should share many customers. Yet the widespread fit of the duplication of purchase law, which states brands share customers with other brands in line with their relative shares, shows that partitioning is generally slight (see Chapter 6).
If brands vary in their degree of differentiation, we might expect to see them exhibit different price elasticity, as customers of more differentiated brands would be less sensitive to price. However, price elasticity seems to vary more with the context of the price change (see Chapter 10). This again suggests that brands within a category have similar levels of differentiation.
Finally, we must consider the highly sucessful, empirically grounded mathematical model, the NBD-Dirichlet (Goodhardt, Ehrenberg & Chatfield 1984). This model effectively assumes that brands compete as undifferentiated choice options of varying popularity. The real world conforms to this model, with the NBD-Dirichlet successfully predicting a wide variety of brand performance metrics in dozens of product categories, different countries and across time (Ehrenberg, Uncles & Goodhardt 2004).61
Every category has brands that differ in price and quality (referred to as ‘vertical differentiation' by economists). Sometimes a category contains brands that are vastly more expensive than the other brands in the category. This price difference shows up in the dissimilarity of the different (i.e. cheaper and more expensive) brand's user bases. The price difference is also exhibited in the more expensive brand's departure from the duplication of purchase law and NBD-Dirichlet benchmarks. The fact that this evidence is not often seen for other brands suggests weak levels of differentiation at best.62
This body of theory and empirical evidence does not support the traditional role of differentiation presented in much marketing literature, nor does it support a 'perfect competition' (commodity) model. Differentiation undoubtedly exists, but empirically grounded theory suggests differentiation is best thought of as a category level (rather than brand level) phenomenon. In summary, differentiation theory is like the neo-classical, economic, perfect competition model, in that it describes an abstract ideal world and not the one that real brands compete within.
Icon brands
It's not just the buying of everyday, run-of-the-mill brands that causes the patterns reported above. These laws also apply to 'icon brands', which are brands that are supposed to represent the height of marketing excellence and consumer bonding. For example, Nike is considered (by the CEO of Saatchi and Saatchi, Kevin Roberts (2004, pp. 78-9)) to be a Lovemark brand, which is a brand that inspires 'loyalty beyond reason'. Yet an analysis by Dr John Dawes (2009) clearly shows that Nike conforms to the double jeopardy law and the duplication of purchase law: Nike's buyers are not 100% loyal; they give it no more loyalty than it should receive given its market share (i.e. Nike's receives less loyalty than Adidas in some markets). Nike sells to the same sort of people as other sports brands, to similar demographics, social classes and psycho-geo-demo-graphics (see Dawes 2009). These patterns in real
buying behaviour raise serious questions about the degree to which buyers perceive rival brands to be different.
Perceived differentiation
So how do buyers choose between branded options? Do buyers need to perceive that the brand they are purchasing is different? Do they need another reason to buy the brand (other than offering the benefits of the category, e.g. it tastes like ice-cream or it works like a credit card)? Systematic study across product and service categories, countries, survey methods and questions types reveals two robust patterns:
1.Buyers of a brand perceive very weak differentiation – yet this does not stop them loyally buying that brand.
2.A brand's level of perceived differentiation is very similar to their rivals.
For example, Table 8.2 reports the proportions of brands' regular buyers who agree that the brand they buy is different or unique.
Table 8.2: Brand user perceptions of differentiation in the soft drink (UK) and banking (Australia) categories
Soft-drinks Different (%) Unique (%) Either (%)
Banks Different (%) Unique (%) Either (%)
Coca-Cola 8 13 19
ANZ 12 4 15
Diet Coke 9 8 15
CBA 12 12 19
Pepsi-Cola 7 10 15
NAB 8 12 12
Fanta 8 5 12
Westpac 9 6 11
Pepsi Max 9 10 19
St George 26 16 32
Schweppes 6 9 13
Canada Dry 10 9 17
Average 9 9 16
Average 13 10 18
Source: Ehrenberg-Bass Institute. See also (Romaniuk, Sharp & Ehrenberg 2007)
All the markets we have examined, whether using our own or others' data, follow a similar pattern: an average of 10% of any brands' users think their brand is different. It could be that these responses reflect apathy when responding to research questionnaires; however, there are three indicators that this is not the case. First, within the markets there were exceptional brands that had obvious functional differences. For example, Aldi supermarkets, which do not stock national brands, scored 67% for “different”; while in the fast food category, sandwich-only brand Subway achieved 50% for “unique” when put in a competitive set with McDonald's, Domino's and KFC. So respondents could indicate when something was perceived as differentiated. However, most brands were seldom perceived as such. Second, the results show expected differences between markets; for example, soft drinks are more differentiated than water. Spirits and skincare – two categories we deliberately chose because they are considered to be highly image driven and supported by vast amounts of brand advertising expenditure – both showed among the highest levels of perceived differentiation (although still low). And, third, other perceptual metrics in the same surveys obtained far higher scores.
To check whether this finding was simply an artefact of the measure, we also tried a forced choice 5-point scale (where 1 is not at all descriptive and 5 is extremely descriptive). The average score was 2.5: few brands scored higher than the neutral midpoint of 3, and none had a mean as high as 4 (a score that would indicate respondents somewhat agreed the brand was different). Therefore, our finding that consumers hardly see their brands as differentiated is consistent across both a pick-any or a rating scale measure. 63
There were some systematic deviations from this general pattern, where brands consistently obtained higher response levels. This was for small, higher-priced brands. This is consistent with the notion that more differentiated brands will be smaller, or, less popular brands are seen to be different. There were also some brand-specific deviations; however, these tended to be linked to large functional differences and less to image differentiation. So the majority of buyers do not explicitly state that they perceive their brand to be differentiated from other brands. Therefore, it is questionable whether perceptions of brand differentiation drive buyer behaviour. Buyers don't need to see differentiation to buy a brand, or to keep on buying it.
Table 8.3: Examples of category level results
Category Current users who perceive it is different (%) Current users who perceive it is unique (%) Current users who perceive it is either (%)
Spirits
20 27 36
Supermarkets
25 21 31
Skincare
17 21 30
Ice-cream
14 11 20
Fast food
16 13 20
Banking
13 10 18
Soft drinks
11 9 18
Condiments
10 9 17
F&V drinks
11 8 16
Ready sauces
9 7 14
Information technology
9 10 14
Soups
8 5 12
Yoghurt
8 5 11
Cars
9 6 11
Water
6 6 10
Electronics
4 6 8
Average 11 10 17
Source: Romaniuk, Sharp & Ehrenberg 2007.
Table 8.4 – showing some of the brand-level data from the information technology category in Table 8.3 – illustrates this with the example of Apple, a poster child for differentiation and customer loyalty (as discussed in Chapter 7). Apple's level of perceived differentiation is higher than other computer brands, but it is not high. Most (77%) of Apple users did not perceive their brand to be different or unique. This seems surprising, given that Apple computers look different and use a different operating system64. However, most computer users have little technical knowledge and few would know what an operating system was. Apple Macintosh is a personal computer (PC), with a graphical user interface, a mouse and keyboard, it runs software (e.g. Microsoft Office), sends email, surfs the web, stores files, prints and so on. It is not unreasonable that its users see it as doing very much the same things as other computers. They certainly buy it to do the same things. Few apparently bought it because they wanted a computer that was different; they bought it to be a good useful computer – just as buyers of any other computer brand. 65
Table 8.4: Computer brand users' perceived differentiation
Users of Different (%) Unique (%) Associating the brand with either (%)
HP 4 8 11
Toshiba 11 5 13
Compaq 10 10 18
IBM 16 13 18
Apple 15 25 27
Dell 5 5 5
Average 9 10 14
Source: Y&R, BAV (Brand Asset Valuator), 1999.
Note: Only a quarter of Apple users think the brand is either different or unique.
Many marketing texts instruct marketers to strive to make customers perceive valued differences between brands. Yet even marketers of highly successful brands seem to have failed. This leads to the conclusion that perceived differentiation plays little part in the success of brands. It is certainly not a case of 'differentiate or die'; if it were, most of the brands we buy would be gone.
The literature on consumer behaviour has for decades focused on customer perceptions of brand features as the main reason why one brand is chosen over others.
This emphasis is misplaced. If buyers of a brand don't often think their brand is different or unique, then presumably this is not the main reason they buy it. We need to look elsewhere for explanations of why customers buy one brand and not another.
It seems counterintuitive that buyers do not often notice that the brands they use are somehow different from other brands in the market (given that some of the brands are functionally different). This shows that brand choice is a trivial task, as is the choice of deciding whether or not to buy from the product category. So buyers do not spend a great deal of time comparing brands in a category, and so differentiation (relative to other brands) is not noticed. This realisation is counter to many models of information processing; buyers are thought to weigh up brands on the merits of their relative attributes (e.g. Alpert 1971; Fishbein & Ajzen 1975; Green, et al 1981), which implies that buyers know about t
he differences between brands. However, what empirical evidence shows is that buyers seem to know something about the brands they use, and very little about the ones they don't. This results in the much lower response levels from non-users (compared to that of users).
The main implication is that it isn't essential for marketers to convince buyers that a product is different before they buy it. This should take a considerable weight off marketers' shoulders, because our data shows that such a task must be near impossible. Instead, marketers should focus on achieving the things that do make customers buy a product.
Brands are different from one another, and they are perceived as such, but differentiation plays a small role in how brands compete. Consider three brands that are clearly functionally different (and maybe different in other ways too): Pizza Hut, McDonald's and KFC. Respectively, they sell pizzas, burgers, and fried chicken. So we should expect that McDonald's primarily competes with other burger providers (Wimpys, Burger King, Hungry Jacks, Carl's Jr, etc.); that Pizza Hut competes with other pizza chains (Dominos, Pizza Pizza, Little Caesars, Papa John's, Pizza Haven); and that KFC competes with other chicken outlets (Harold's Chicken Shack, Bojangles, Oporto, Chesters, Red Rooster etc.). According to textbook logic, McDonald's, KFC and Pizza Hut would not compete against one another (or very indirectly). Of course this is nonsense; they are direct competitors; they compete in one big fast food category. Rather than working to differentiate themselves and splinter this fast food category, they actually try to reduce their differentiation; for example, McDonald's advertises that it offers chicken (burgers), KFC advertises that it offers (chicken) burgers and Pizza Hut advertises that it sells chicken (pizza). 66