第1章
The Marketing Revolution You Never Saw Coming
Ever wondered why Coca-Cola spends billions on advertising when everyone already knows what it is? Or why Apple's customers aren't actually more loyal than Dell's? When Byron Sharp's "How Brands Grow" hit shelves in 2010, it sent shockwaves through the marketing world. Praised by executives at Coca-Cola and Nielsen as revolutionary, this evidence-based marketing manifesto challenged virtually everything marketers thought they knew about building brands. The book has become required reading at companies like Google, Mars, and Unilever, with the Financial Times calling it "a scientific investigation that upends the conventional wisdom about how consumers buy." Sharp, a professor at the University of South Australia and director of the Ehrenberg-Bass Institute, doesn't just offer opinions-he presents decades of rigorous research that demolishes marketing's most cherished myths. His findings are so counterintuitive yet empirically sound that they've been described as marketing's equivalent to Copernicus proving the Earth revolves around the sun.
第2章
The Shocking Truth About How Brands Actually Grow
Imagine you're the marketing director for a major brand. You've been taught that success comes from differentiating your brand, building deep relationships with loyal customers, and targeting specific segments. What if everything you believed was wrong?
The fundamental discovery that underpins Sharp's research is startlingly simple: big brands have more customers than small brands. While this might seem obvious, its implications are profound. When comparing brands of different sizes, their penetration metrics (how many people buy them) differ dramatically while their average purchase rates vary only slightly. This pattern, known as the "double jeopardy law," shows smaller brands get "hit twice": they have fewer buyers who also buy slightly less often.
This law has been documented across thousands of brands, dozens of countries, and over decades of research. From washing powder in the UK to cars in the US, the pattern holds true regardless of category or country. Even supposedly unique "niche" brands follow this pattern.
The strategic implication is clear: growth comes primarily from increasing penetration (getting more customers), not from increasing loyalty among existing customers. When brands grow, they don't convert a small group of loyalists into super-fans-they simply acquire more customers who buy at roughly the same frequency as their existing ones.
This directly contradicts conventional marketing wisdom. A meta-analysis of 207 advertising tests showed that campaigns targeting penetration were the only ones correlated with larger sales effects. Analysis of 880 IPA Effectiveness Awards entries revealed campaigns targeting penetration were twice as likely to report large improvements in sales and profits than those targeting loyalty.
Yet despite this evidence, more marketing campaigns still aim for loyalty than penetration. Why? Because marketers have been trained to believe in myths about customer loyalty that simply aren't supported by data.
Consider the math: For Toyota, which typically gets half its sales from returning customers, reducing defection to zero would only gain them 50% more sales (about one percentage point of market share). Meanwhile, with about half of all new car buyers switching brands annually, 50 points of market share are available through acquisition. The potential gains from acquisition are fifty times greater than from retention.
The evidence is clear: brands grow primarily by acquiring new customers, not by increasing loyalty among existing ones. This isn't just a theoretical observation-it's a fundamental law of how markets work.
第3章
The Light Buyer Revolution: Who Really Matters Most
If you're like most marketers, you've probably been taught some version of the "80/20 rule"-the idea that 80% of your sales come from 20% of your customers. This belief has led countless brands to focus their marketing efforts on heavy users while neglecting occasional buyers. But what if this conventional wisdom is fundamentally wrong?
The reality is far more nuanced. For most brands, the heaviest 20% of customers typically account for about 50% of sales-not 80%. Professor Gerald Goodhardt's research established what he called the "20:30:50 law": the heaviest 20% of buyers account for 50% of purchases, the middle 30% account for 30%, and the lightest 50% account for 20%.
This distribution means light buyers collectively contribute substantially to a brand's success. Data from Coca-Cola reveals the average buyer purchases about 12 times yearly, but this average is misleading-half of all Coke buyers purchase just once or twice annually. A similar pattern exists for Pepsi and across virtually all consumer categories globally.
Even more surprising is the "law of buyer moderation." This statistical phenomenon shows that today's heavy buyers will naturally become lighter buyers in the future (without any real behavior change), while today's non-buyers and light buyers will become heavier. Analysis of a leading US tomato sauce brand showed that despite stable overall sales, 14% of volume came from households that didn't buy at all the previous year, while heavy buyers' contribution dropped from 43% to 34%.
This occurs because of natural variation in purchase timing-some years customers buy once, other years twice as much. This "wobble" means many consumers are misclassified based on a single year's data.
The distribution of buying rates for all brands follows a negative binomial distribution (NBD)-a mathematical pattern discovered by Andrew Ehrenberg in 1959. This highly skewed distribution explains why brands have many more light buyers than heavy ones.
When brands grow or decline, they simply move from one weight of this distribution to another, with changes in buying propensities occurring across all buyer groups. Growth comes from gaining more heavy buyers, more medium buyers, and many more light buyers.
Even brands with 100% market share would still have many light buyers because category buying itself shows a skewed distribution. For example, most UK households buy toothpaste three times or less annually, with 11% not buying at all in a given year. For individual brands like Colgate, these figures are even lighter, with 86% buying three times or less and 59% not buying at all.
This explains why marketing strategies targeting only heavy buyers typically fail, while approaches that reach light and non-buyers have greater success potential. The evidence clearly counters fashionable marketing thought, showing that reaching all buyers is essential, especially light, occasional buyers of the brand.
第4章
The Myth of Consumer Tribes: Your Buyers Aren't Special
One of marketing's most cherished beliefs is that different brands attract fundamentally different types of people. We've all seen those perceptual maps showing how Brand A appeals to young, urban professionals while Brand B attracts suburban families. But what if these distinctions exist primarily in marketers' imaginations?
A groundbreaking 1959 study found Ford and Chevrolet owners had essentially identical personality profiles, contrary to expectations. Later comprehensive studies across hundreds of brands and categories confirmed that competing brands sell to the same types of people.
Sharp presents extensive evidence showing that competing brands sell to remarkably similar customer bases. Car brands like Rover, Escort, Sierra, and Cavalier all have nearly identical demographic profiles in gender, household size, and even newspaper reading habits. Similarly, beer brands in Canada show minimal differences despite variations in price and origin. Credit cards, too, have virtually identical user profiles with negligible demographic differences between brands.
Across more than 40 product categories, the average deviation from the category norm is tiny-typically just 2-3 percentage points.
Even when marketers deliberately attempt to target specific demographics, they typically end up with normal-looking customer bases. The Yorkie chocolate bar's "It's not for girls!" campaign, despite overtly targeting men, still attracted a customer base that was 44% female. Similarly, a credit card featuring baby imagery and donations to a maternity hospital showed only slightly higher female usage (63%) compared to non-cardholders (58%), with no meaningful difference in family status between users and non-users.
Brand users also hold similar values regardless of which competing brand they buy. The main difference is simply that people have opinions about the brands they use and know little about brands they don't use. Sharp calls this the "my mum" phenomenon-just as everyone thinks their mother is special while sharing similar sentiments, brand users feel similarly about their chosen brands.
Even product variants designed for specific segments show surprisingly similar customer bases. Regular and diet soft drinks maintain consistent market share proportions across all demographic groups. Similarly, bottle sizes and even specialized products like low-allergy fabric conditioners follow double jeopardy patterns rather than serving distinct niches.
This discovery that brands sell to similar customer bases is actually positive for marketers. It means brands aren't constrained to specific niches-they can potentially attract any category buyer. When research shows a brand "skewing" toward certain demographics, marketers should question why rather than assuming that's their natural audience.
These findings suggest tremendous growth potential for brands, as competitors' customers are essentially identical to their own. However, this also means competitors can just as easily target your customers, highlighting the ongoing importance of effective marketing.
第5章
The Competitive Reality: Everyone Is Your Competitor
Traditional marketing theory suggests brands compete only within narrow segments-luxury cars compete with other luxury cars, not with economy models. Philip Kotler's influential marketing textbooks present this segmentation approach as the evolution of marketing from mass approaches to targeted strategies. But does this reflect how consumers actually behave?
Analysis of customer sharing between brands reveals a striking pattern: brands share customers with other brands in proportion to those brands' market size. This "duplication of purchase law" shows that all brands in a category share more customers with large brands and fewer with small brands. The pattern holds remarkably consistent across categories, with only minor deviations indicating market partitions.
For example, in ice cream, every brand shares about 38% of its customers with market leader Carte D'Or, while premium brands like Ben & Jerry's and Haagen Dazs show slightly higher-than-expected sharing with each other.
This law provides valuable insights for defining product categories based on actual consumer behavior rather than product features or manufacturing processes. It helps marketers understand their true competitive set, predict where new brands will steal sales from, estimate cannibalization effects, and benchmark customer defection patterns.
Narrow category definitions often give brand managers a false sense of security and artificially limit growth targets, whereas duplication analysis reveals the broader competitive reality.
Perceptual maps often suggest dramatic market partitions based on brand image, but duplication analysis reveals these divisions are exaggerated. Using flavored milk as an example, while perceptual mapping suggested distinct segments (working-class males, children, health-conscious women), actual purchase data showed most brands sharing customers according to the duplication law. Only brands with substantial functional differences (like zero-sugar options) showed true partitioning in customer behavior.
Companies shouldn't worry about having similar brands that sell to similar populations. This is normal market behavior-even Coke has Diet Coke and Coke Zero, while Mars has Mars Bar and Snickers. What matters is brand distinctiveness-ensuring brands are easily recognizable and distinguishable visually, even if they don't compete as differentiated brands.
Markets function largely as mass markets with only minor fragmentation, typically addressed through brand variants. The toothpaste category, often cited as segmented marketing, actually represents "product-variety marketing"-a type of mass marketing. Colgate's various toothpaste formulations don't use different distribution channels or dramatically different pricing; they appear on the same shelves and in mass media advertising.
Brand managers should think like "sophisticated mass marketers," recognizing heterogeneity within mass markets while seeking avenues for broad reach rather than niche positioning.
第6章
The Loyalty Delusion: How Consumers Really Buy Brands
Despite claims that advertising manipulates consumers into irrational brand preferences, psychological studies show branding effects are actually quite weak. Even in artificial laboratory conditions, branding's psychological impact is surprisingly limited.
Brand loyalty appears naturally in all markets, even commodities. Tucker's 1964 experiment showed people develop loyalty even when identical bread loaves were given different labels. Car buyers show approximately 50% repeat-purchase rates despite years between purchases and numerous available options.
This loyalty occurs because buyers typically consider only about two brands when making purchases, with one usually being their current brand. Even television viewing shows loyalty patterns-households with access to hundreds of channels still watch only about a dozen regularly, demonstrating how people naturally restrict their repertoires across all categories.
While loyalty exists universally, it's rarely exclusive. The data shows only about 13% of a brand's customers are 100% loyal, with smaller brands having even fewer exclusively loyal customers (following the double jeopardy law). Categories purchased less frequently show higher levels of 100% loyalty, but this is largely because infrequent buyers have fewer opportunities to display divided loyalty. Most consumers practice "polygamous loyalty"-regularly purchasing from a small repertoire of preferred brands.
Larger brands disproportionately attract light category buyers-a statistical effect known as the "Natural Monopoly Law." For example, occasional soft drink buyers are more likely to choose Coca-Cola, not because they prefer larger brands, but due to probability. The data shows this clearly: Heinz buyers purchase tomato sauce four times annually, while buyers of the smaller C&B brand buy tomato sauce eight times yearly (though C&B accounts for only 1.2 of those purchases). This means smaller brands are bought by heavier category users, while larger brands capture more light buyers.
Real-world loyalty is far more mundane than marketing textbooks suggest. While loyalty exists everywhere, it's divided and strongly driven by opportunity rather than passion. Consumers develop routines that make purchasing easier, but these habits are rarely exclusive to one brand. From cars to canned soup, routine results in passionless brand loyalty-buyers typically have several brands they routinely purchase.
Sharp critiques the marketing industry's obsession with creating passionate consumer commitment. He particularly criticizes academic research on "brand love," noting that even research attempting to prove emotional branding found that only 4% of Australian beer drinkers reported feeling "deep affection" for their preferred brand-and even these "lovers" only purchased the brand half the time.
Consumers face hundreds of thousands of brands competing for attention, making brand choice relatively trivial compared to category purchase decisions. Brands function as a "necessary evil" that add complexity but also enable purchasing routines. These habits create passionless loyalty that makes buying easier-almost automatic.
第7章
Differentiation vs. Distinctiveness: The Real Path to Brand Growth
Marketing textbooks unanimously preach differentiation as the centerpiece of strategy, citing authorities like Levitt, Kotler, Aaker, and Trout. Yet real-world competition shows more competitive matching than meaningful differentiation. Sharp notes the irony that marketing textbooks themselves show almost no differentiation while preaching its importance.
When examining brand perceptions, once usage effects are removed, there's little evidence of brands having unique images-all brands tend to be seen similarly by those familiar with them, with only slight double jeopardy patterns where smaller brands score slightly lower. Looking at Martin Collins' research, Sharp shows that when brands are ranked by familiarity, all brands gain very similar perception scores with only minor differences.
Sharp dismantles the concept of brand personality, noting that early research by Franklin Evans found no difference in personality traits between Ford and Chevrolet buyers. Despite this, marketers persisted with the idea, particularly after Jennifer Aaker's 1997 brand personality scale. However, empirical evidence shows consumers are reluctant to view brands as people-only about 5% associate human characteristics like "ruggedness" with brands, and these perceptions are weakly held.
Despite marketing theory suggesting successful brands should have unique image attributes, Sharp cites research examining 130 brands across 13 categories showing people rarely (only about 3% of the time) see a single brand as exclusively associated with a particular image. More successful brands don't have proportionally more unique associations, nor do customers with greater brand preference hold more unique associations.
Sharp acknowledges differentiation exists but argues it's largely situational rather than brand-level. Situational differentiation includes factors like availability, location convenience, or simply what comes to mind at purchase time.
He presents several empirical patterns contradicting strong brand-level differentiation: loyalty doesn't vary much between brands, brand user profiles are similar within competitive sets, brands share customers according to their market share, and the NBD-Dirichlet model successfully predicts brand performance metrics despite assuming brands compete as undifferentiated options of varying popularity.
Sharp debunks the myth that "icon brands" like Nike represent exceptional marketing excellence and consumer bonding. Despite Nike being considered a "Lovemark brand" inspiring "loyalty beyond reason," research by Dr. John Dawes shows Nike conforms to the double jeopardy law and duplication of purchase law. Nike's buyers aren't 100% loyal and give it no more loyalty than expected given its market share.
Research across multiple categories, countries and methodologies shows two robust patterns: 1) Buyers perceive very weak differentiation yet loyally buy their brands anyway, and 2) A brand's level of perceived differentiation is very similar to rivals'. Typically only about 10% of any brand's users think their brand is different. Even Apple, "a poster child for differentiation," scored low-77% of Apple users did not perceive their brand as different or unique.
Sharp concludes that perceived differentiation plays little role in brand success, contradicting the marketing maxim "differentiate or die." Most successful brands thrive despite minimal perceived differentiation.
Instead of differentiation, Sharp argues for distinctiveness-helping brands stand out so buyers can easily recognize them. Distinctive brand assets-including colors (Garnier green), logos (McDonald's arches), taglines (Nike's "Just do it"), symbols, celebrities, and advertising styles-help consumers notice, recognize and recall brands.
Unlike differentiation claims, these distinctive elements can be legally protected through trademarks. Distinctiveness benefits both marketers and consumers by reducing cognitive effort in cluttered environments, making products easier to find without requiring conscious thought.
Two critical criteria determine a distinctive asset's value: fame (how many people associate the brand with that asset) and uniqueness (how many people only associate that brand with the asset). Building strong distinctive assets increases the number of identification triggers for a brand, improving advertising effectiveness and making products easier to notice in shopping environments.
第8章
Mental and Physical Availability: The Twin Pillars of Brand Success
The key marketing task is making a brand consistently easy to buy for all potential customers by building both mental and physical availability. Brands primarily compete through these two dimensions, with even product innovation working primarily by enhancing these factors.
Modern consumers lead hectic lives with technology cramming more activities into limited time. Most customers are very light, occasional buyers who rarely think about brands. Marketers face the challenge of capturing attention from people preoccupied with more interesting things.
Contrary to popular belief, people aren't replacing traditional media with digital alternatives-they're simply adding more media consumption to their lives. A typical viewer watching just 90 minutes of television daily encounters over 60 ads, plus many more through other channels. This realistic exposure level still overwhelms consumers' attention capacity, with most advertising receiving minimal notice.
Faced with overwhelming choice, consumers adopt simplifying strategies. Rather than optimizing, they "satisfice" by settling for satisfactory options and effectively decide not to consider most available brands. Most brands are ignored, and sometimes no evaluation occurs at all-consumers simply reach for familiar products.
A brand's sales are primarily determined by how many consideration sets it failed to enter. Contrary to traditional marketing theory, evaluation is less important than we think because: buyers' memories are imperfect and variable; consumers only consider a tiny subset of known brands; and evaluation criteria change depending on circumstances.
Mental availability (brand salience) is a brand's propensity to be noticed or thought of in buying situations. It's more than simple awareness-it's based on the network of memory associations in buyers' minds. These associations include physical attributes, usage occasions, emotional connections, and visual elements. The more extensive and fresher these memory networks, the greater the brand's chance of being considered.
Traditional awareness measures using single cues poorly predict how often a brand will be noticed in actual buying situations. Mental availability depends on both quantity (number of associations) and quality (strength and relevance) of memory links to a brand. When a brand scores well on awareness but sales disappoint, the problem is often that buyers simply don't think of it when shopping.
Different cues mean different competitors might be considered by buyers at any time. Competitive options aren't limited to the same product category-"something to wake me up" might conjure coffee, Red Bull, Pepsi, or even a brisk walk. Marketers should think of "cue competitors" rather than functional look-alikes.
Physical availability means making a brand easy to notice and buy for as many consumers as possible across diverse buying situations. While high physical availability is essential for high market share, it doesn't guarantee it-many widely distributed brands still have low sales due to lacking mental availability or having limited appeal.
Mental and physical availability constitute intangible, market-based assets that underpin a brand's financial value. Created through marketing activity, these assets cost money to build but provide security of future profit. They make marketing more productive-advertising works better when existing memory structures exist and when the brand has strong physical availability.
For established brands, these assets create marketing inertia, making effects harder to measure as responses become sluggish. Neglected brands with substantial residual mental and physical availability can be revitalized for tremendous gains, as demonstrated by McDonald's comeback and the Queen Adelaide wine revival.
第9章
The Seven Rules for Marketing That Actually Work
Despite marketing's complexity, scientific study reveals regularities that allow for prediction and insight. For branded competition, seven strategic rules emerge:
1. Continuously reach all category buyers through distribution and communication. This means reaching both existing customers and potential customers, including light and non-buyers of your brand.
2. Ensure the brand is easy to buy by removing barriers to purchase. This involves understanding what makes brands genuinely convenient in consumers' lives and addressing both physical and mental obstacles.
3. Get noticed often because reaching consumers means nothing if your brand isn't noticed. Emotional content in advertising primarily works by gaining attention, with ad liking increasing sales effectiveness by encouraging more attention. Clever, likable creativity helps advertising get noticed, but what we see depends largely on what's already in our heads.
4. Build brand-linked memory structures by working with what consumers already have in their minds. Brand image research exists precisely to understand these existing structures so communication can be crafted accordingly. For established brands, refreshing existing memories is paramount, with consistency across campaigns being essential.
5. Create distinctive assets that allow consumers to direct their natural loyalty to specific brands. Branding ensures advertising refreshes memory structures for the right brand and makes brands more noticeable in cluttered environments. Mental associations act like coat-hangers for other brand memories-iPod's white earbuds, the Jolly Green Giant, Nike's swoosh, all serve as instant brand identifiers.
6. Be consistent yet fresh because a brand cannot be distinctive without consistency. The art of advertising lies in telling the same story repeatedly in fresh, entertaining ways. Even slight inconsistency confuses occasional buyers, which explains why packaging changes often trigger dramatic sales drops.
7. Stay competitive with mass appeal, removing reasons not to buy. When consumers shop, they primarily notice and consider a very limited set of options-often just one brand-with most alternatives eliminated before any conscious evaluation. This means "reasons not to buy" can be more damaging than "reasons to buy" are helpful.
Brand growth ultimately comes down to investing in market-based assets-improving mental and physical availability. Simply spending more on advertising isn't enough; the investment must build lasting brand assets that enhance future marketing returns. Creating distinctive memory structures makes all future advertising more effective by reducing brand ambiguity. Similarly, expanding physical distribution amplifies marketing effectiveness.
Most marketing departments don't yet see themselves as custodians of crucial market-based assets-mental and physical availability. Despite talk about building strong brands, brand managers lack specialized knowledge about managing these assets and rarely measure them properly. Instead, they waste time on "esoteric quackery" like segmentation, differentiation, and brand personality.
The future for marketing departments and consumer brands is bright-if they adapt to media fragmentation, embrace empirical marketing laws, and reject "anything goes" marketing strategies. By developing expertise in market-based assets, marketing can earn greater organizational respect since no other department measures or manages these intangible assets.
