by Byron Sharp
When marketing succeeds in increasing a brand's market share, then buying propensities change across the board. This tells us that marketing has the best chance of being successful when it has as much reach as possible. Marketing is particularly successful when it reaches light and non-buyers of a brand.
Therefore, marketing that seeks to increase sales by targeting heavier buyers is unlikely to succeed. Loyalty programs are a classic example of a strategy skewed towards heavier buyers of a brand. These programs are are still very popular with marketers; and yet with thousands of such programs running it is difficult to find even a few claiming great success. The serious academic studies report the same finding: loyalty programs generate small or no shifts in market share (see Chapter 11). Price promotions, in contrast, do increase sales, but they are also skewed towards heavier buyers. In spite of a lot of volume being sold on deal, promotions stimulate far less incremental volume (see Chapter 10). Many of the sales would have occurred anyway, to customers who already buy the brand regularly. When the price promotion ends, sales instantly collapse back to their normal level. This is a good example of how difficult it is to bring about sustained growth by targeting only heavy buyers. If this weren't the case, then we would routinely see brands that had different buying distributions, and unusually high loyalty metrics for their market share levels. But we don't.
Conversely, marketing that targets light buyers of the brand, and/or non-buyers, has far greater chance of success. This is because such marketing has great reach. Most of a brand's buyers are light buyers, and because it's nearly impossible to simply target light buyers, heavy buyers tend to get hit too. And heavy buyers are far more likely to notice advertising, visit stores and notice publicity.
A classic example of brand growth is due to winning distribution in a new geographical area. This reaches buyers who, because it was inconvenient for them, were previously light buyers (or even real non-buyers) of the brand. These previously light/non-buyers start buying the brand and its penetration metric jumps up. The new store also improves the availability of the brand for existing heavier buyers, when they happen to be in the neighbourhood. So over time they also buy a little more often. This helps nudge up average purchase frequency (and other loyalty metrics). This large rise in penetration and small gain in loyalty fits the double jeopardy pattern we saw in brand performance metrics in Chapter 2.
Conclusion
Now you know how brands grow and how marketing, when it is successful, affects buying rates. The implications counter much fashionable marketing thought, and yet are clear and simple:
1. Acquisition is vital for growth and maintenance.
2. Reaching all buyers is vital, especially light, occasional buyers of the brand.
This is a recipe for clever mass marketing, which is not surprising because it was clever mass marketing (not CRM, relationship marketing or loyalty programs) that built the majority of today's leading brands. But it need not be a recipe for unsophisticated mass marketing. Reaching all potential buyers of a brand, at the right time and at low cost, is tricky and there is much to learn. The digital revolution is creating new opportunities to reach consumers in different ways, at different times – to be more relevant, and fit in better to their heterogeneous lives. There are great opportunities for sophisticated mass marketing.
Our next chapter takes a closer look at the buyers we have to reach. Who are they? How do we identify them?
Chapter 5: Our Buyers Are Different
Byron Sharp
The good news is that your customers are just like your competitors’ customers, and their customers are like yours. This means their buyers are up for grabs. So, target the whole market – all sorts of different people.
Who are our buyers? Who are our potential customers?
Common marketing ideology states that differentiated brands should sell to different groups of people, or should sell for different occasions or uses. In which case, an important question for brand managers is, “What sort of people buy my brand?” The answer to this question provides insight into how to market a brand. It also allows for a brand's current sales potential to be estimated: if a brand sells to a unique customer base it can't sell to everyone and its growth potential has clear boundaries.
In 1959 a business school professor at the University of Hawaii gave personality tests to people who owned Ford and Chevrolet cars (see Evans 1959). This sounds like an odd thing to do, but the theory was that Americans in the 1950s took their cars very seriously, cars were deeply symbolic purchases, and marketers had invested in building distinctive brand images. It was expected this would show up in interesting ways in the personality profiles of the respective customer bases. Executives from car companies and academics had expressed this belief.
The results were indeed interesting. The personality and demographic profiles of Ford and Chevrolet owners were essentially identical. The lack of difference shocked the marketing world. At first people argued about the results. Then the finding was repeatedly confirmed with different samples, techniques and product categories. After that everyone decided to ignore the discovery. As Winston Churchill said, “Man will occasionally stumble over the truth, but usually manages to pick himself up, walk over or around it, and carry on.” (Klotz 1996).
Nearly 50 years later a series of vastly more comprehensive studies were undertaken examining the profiles of the customer bases of competing brands. These studies profiled hundreds of brands, across dozens of categories (from cigarettes to computer games to mortgages) over time (Hammond, Ehrenberg & Goodhardt 1996; Kennedy & Ehrenberg 2000; Kennedy, Ehrenberg & Long 2000, Kennedy & Ehrenberg 2001a; Kennedy & Ehrenberg,2001b). Just as importantly, the studies used hundreds of variables capable of describing buyers: demographics, psychographics, attitudes, values and media habits. The key discovery of these studies is that competing brands sell to the same sort of people. Within each brand's customer base there is a lot of variation (i.e. different types of people), but each brand has the same variation.
Table 5.1 presents car ownership data from a different decade (1990s), but still the same pattern is clear. The customer bases of these UK car brands are quite similar to each other – even in the newspapers they read. Rover buyers, in spite of driving an up-market brand, still read the tabloid newspaper, the Sun, far more than they read the more intellectual newspaper, the Telegraph. This is also true for the owners of all the other car brands, and reflects the fact that the Sun is considerably more popular than the Telegraph. The greatest difference in the customer base is household size. This reflects the fact that larger households buy brands of cars that are physically bigger (more seats with a larger boot). Sierra and Cavalier are somewhat physically larger vehicles.
Table 5.1: Car brands sell to similar types of people, UK 1990s
Gender (%) Household size (%) Owners who read newspapers (%)
Owners of: Men Women 1 - 2 people 2+ Sun Telegraph
Rover 52 48 40 24 20 9
Escort 50 50 34 27 26 5
Sierra 50 50 25 34 28 4
Cavalier 51 49 29 33 24 6
Average 51 49 31 30 25 6
Source: TGI.
The similarities in the customer bases include the values that the different car owners hold. Table 5.2 shows a delightful collection of consumer values. Respondents may be fibbing when 98% of them say they don't judge people by the car they drive but, if so, your honesty doesn't depend on the brand of car you drive.
Table 5.2: Owners of different car brands hold similar values, UK 1990s
Owners of: I cannot bear untidiness I keep up with technology I judge people by the
car they drive A car is only to get from A to B
% agreeing
Rover 16 11 1 18
Escort 19 9 3 21
Sierra 17 9 2 17
Cavalier 17 10 3 17
Average 18 10 2 18
Source: TGI.
Let's see how this generalises to very different condition, beer brand
s in Canada (see Table 5.3). In spite of the differences in price, origin (from Mexico to Toronto) and brand images, these beer brands have the same customer profiles. Modelo and Corona skew a little bit towards women and younger drinkers, but that's about the only substantive difference.
Table 5.3: Beer brands’ consumers
Gender (%) Age group (%) Earnings (‘000s)
Buyers of Men Women <29 50-59 <$30 <$50
Coors 58 42 23 22 21 36
Canadian 71 29 30 19 26 35
Bud 67 33 32 19 30 27
Oland 69 31 34 17 24 38
Blue 69 31 30 20 28 28
Sleeman 71 29 24 20 18 40
Corona 52 48 34 13 24 35
Modelo 51 49 33 12 25 35
Miller 69 31 38 17 27 39
Average 64 36 31 18 25 35
Source: Data courtesy of Dee McGrath, Molson Breweries Canada; region and year hidden.
Now let's look at credit cards (see Table 5.4). These are in a product/service category where there can be substantial differences between products. These differences might appeal to particular market segments and not to others.
Again, the user profiles of the brands are very similar. Every credit card brand has a customer base that's about half female and half male. Barclays MasterCard has a customer base that is 60% male compared to the average of 54%, but this is a small difference that has no practical marketing significance. No sensible marketer would alter their marketing strategy based on this data. Also, around 40% of every brand's customer base is aged between 35 and 54 years old; this percentage is 47% for Access NatWest – but again, this is a small difference that has no practical significance.
Table 5.4: Demographic profiles of credit card brands, UK
Gender (%) Age (years) (%)
Credit card Male Female 15-19 20-24 25-34 35-44 45-54 55-64 65+
Barclaycard Visa 51 49 1 4 17 23 22 18 14
TSB Trust card 51 49 1 3 18 20 21 19 19
Access NatWest 54 47 1 2 16 24 23 21 14
Access Midlands 52 48 1 3 18 24 21 20 13
Barclays MasterCard 60 40 1 3 15 21 23 20 17
Access Lloyds 54 46 2 3 18 22 22 20 13
Bank of Scotland Visa 56 44 1 2 17 21 26 19 14
Midland Visa 53 48 1 2 15 24 24 19 15
TSB MasterCard 56 45 1 6 18 20 21 18 16
Co-op Bank Visa 56 44 2 2 17 17 23 20 20
Average brand 54 46 1 3 17 22 22 19 16
Source: TGI.
These sorts of minor difference – some simply due to random sampling variation – were the norm in all the research studies. This is illustrated in Table 5.5, which shows the average size of the average deviation from the brand norm in more than 40 product/service categories summarised by Dr Rachel Kennedy (see Kennedy & Ehrenberg 2000).
Table 5.5: Deviation from the brand norm
Category Demo Media Values Average
Cigarettes
4
4
6
6
Cat food
3
1
4
4
Mints
3
1
3
3
Toothbrushes
3
1
3
3
Private health insurance
4
2
3
3
Sweets
3
1
2
3
Crisps
3
1
3
3
Toilet soap
3
1
2
3
Package holidays
3
1
3
3
Dry batteries
3
1
2
3
Other chocolate
3
1
2
2
Kitchen rolls
3
1
3
2
Nuts
3
1
2
2
Chocolate bars
3
1
2
2
Toothpaste
2
1
2
2
Toilet paper
2
1
2
2
Computers
3
2
2
2
Record shops
3
1
1
2
Store retail cards
3
1
2
2
Computer games
3
1
2
2
Vitamins
3
1
2
2
Liquid detergent
3
1
2
2
Grocers
3
1
2
2
Yogurt
3
1
2
2
Light bulbs
2
1
2
2
Car tyres
2
1
2
2
Stain removers
2
1
2
2
Car insurance
3
1
1
2
Coffee
2
1
2
2
Home contents
2
1
1
2
Paint
3
1
1
2
Shampoo
2
1
2
2
Airlines
2
1
1
1
Camera film
2
1
1
1
Headache tablets
2
1
1
1
Cars
2
1
1
1
Credit cards
2
1
1
1
Mortgages
2
1
1
1
Fuel
2
1
1
1
Retailers
1
1
1
1
Average 3 1 2 2
Source: Kennedy & Ehrenberg 2000; Data courtesy of TGI.
The categories are ranked by size of the deviation, but as you can see there isn't great variation between them. In each of these categories the typical deviation is small, apart from some rather obvious differences, such as:
•Scottish newspapers have more readers in Scotland than the average newspaper brand.
•Children's television channels have more children viewers than the average television channel (though still almost half the viewers of these channels are adults).
•Very expensive brands have fewer poor people in their customer base.
Large exceptions are worthy of management attention, but of course they are undoubtedly well known (e.g. that's why we use names like 'children's channels').
Extreme targeting
Maybe the real-world looks like this because marketers haven't targeted special audiences. Maybe if they tried harder reality might match textbook theory? But the two following case studies suggest that even
when marketers try, in very overt ways, to target special audiences, they end up with normal-looking customer bases (so long as they are successful in winning market share). By definition, the more market share they gain, the more normal their customer base becomes.26
No girls
I first saw a Yorkie chocolate bar in a London corner store and was surprised by the packaging: it featured the slogan, “It's not for girls!” and had a picture of a girl that was crossed out. This packaging looks like a serious attempt to gain a segmented customer base: men.
Of course, this is cheeky British humour, designed to get the brand, and its television advertising, noticed. 27 However, decades before Nestle adopted this slogan, Yorkie ads consistently stressed the large chunks of chocolate28 and very much targeted blue-collar, male workers. Most of the original advertising featured truck drivers enjoying Yorkie bars29. After a long gap, or a period of unmemorable advertising, the “It's not for girls!” campaign was launched – perhaps as a return to the brand's historic roots. As one commentator noted, “From the ads they seem to be targeting not ‘British men’ but British, large, bearded, macho, builders. That's got to be a limited market ...” (Redfern, 2002). But either the market got the joke or the targeting failed because this is what the user base of Yorkie looks like today:
Table 5.6: Percentage of Yorkie bar buyers: male versus female over different product variants
Gender Total brand Yorkie -
Milk
Yorkie - raisin & biscuit Yorkie -
the nutter
Male 56 57 46 40
Female 44 43 54 60
Source: Kantar Worldpanel.
Table 5.6 shows that the demographic breakdown fluctuates across Yorkie varieties, partly due to random sampling variation, but the obvious pattern is that a lot of women buy Yorkie – they make up about half of its customer base.