Mental availability and physical availability

Origin

Byron Sharp published How Brands Grow: What Marketers Don’t Know with Oxford University Press in 2010 [1]. Chapter 12 carries the two terms. Mental availability is the probability that buyers “notice, recognize and/or think of a brand in buying situations” [1][2]. Physical availability is the ease of finding and buying it. The book’s argument is that these two are what budgets should build. What “easy to buy” means has moved since 2010: in McKinsey’s 2024 B2B Pulse survey of nearly 4,000 decision makers in 13 countries, buyers used an average of ten channels to interact with suppliers, up from five in 2016, and among sellers that offer e-commerce it now brings in 34% of revenue against 17% from in-person sales [30].

E-commerce (sellers that offer it)34%

In-person sales17%

Video conference12%

Being easy to buy is no longer a question of distribution depth alone; buyers used an average of ten channels in 2024 against five in 2016, and the channel that needs no salesperson now carries the most revenue. The remainder is spread across email, mobile apps, e-procurement portals and online chat, which the report does not break out.

Source: McKinsey & Company, B2B Pulse Survey 2024, nearly 4,000 B2B decision makers, 34 sectors, 13 countries, Truth 3 and Exhibit 6. Redrawn by Brave New Media.

Unit: % of revenue

Show the data

Sales channel	% of revenue, 2024
E-commerce (sellers that offer it)	34
In-person sales	17
Video conference	12

The problem Sharp was solving was a decade of loyalty-first management. Reichheld and Sasser had told Harvard Business Review readers in 1990 that “companies can boost profits by almost 100% by retaining just 5% more of their customers” [18], and Aaker’s Managing Brand Equity (1991) had put brand loyalty first among five equity assets [20]. Sharp’s answer: brands do not grow by deepening loyalty, they grow by adding buyers.

The regularity underneath the book is double jeopardy, which William McPhee described in Formal Theories of Mass Behavior (1963) for actors, comic strips and radio presenters before anyone applied it to brands [5][9]. Ehrenberg, Goodhardt and Barwise formalised it for buying in the Journal of Marketing in 1990: “a small brand typically has far fewer buyers than a larger brand. In addition, its buyers tend to buy it less often” [6]. Uncles, Ehrenberg and Hammond (1995) tied the regularities together in the NBD-Dirichlet model [7]. Sharp’s own step before the book was brand salience, defined with Jenni Romaniuk in 2004 as “the brand’s propensity to be noticed or come to mind in buying situations” [8]. Mental availability is that idea renamed and widened. How Brands Grow Part 2 (2016) took the laws to emerging markets, services and luxury [3]; Romaniuk’s Building Distinctive Brand Assets (2018) is the operating manual for the mental half [4].

What the evidence shows

Ehrenberg, Goodhardt and Barwise (1990) reviewed the panel evidence, the theory and the known exceptions [6]. Graham, Bennett, Franke, Henfrey and Nagy-Hamada (2017) took the law into long-run continuous buying, emerging markets, capital purchasing and house-of-brands portfolios. Across 32 differentiated replications, some 35 years apart, they “find no boundary condition to the operation of the Double Jeopardy characteristic” [10].

The B2B evidence is the part a Brave New reader will want. Romaniuk, Dawes and Faghidno (2021) published an Ehrenberg-Bass report with new survey data on UK business banking and US business insurance, plus a table of eight earlier B2B studies [9]. In banking, Barclays, used by 41% of firms surveyed, held 5.7 of 17 possible products per customer and 27% solely loyal customers; Metro, used by 11%, held 3.2 products and 3% [9]. The report does not state its sample size.

% of B2B customers (penetration)% of its customers solely loyal

Barclays41%27%

HSBC30%29%

Natwest25%27%

Lloyds25%34%

Nationwide21%5%

Santander19%24%

Halifax16%8%

RBS13%10%

Metro11%3%

TSB10%10%

Read down the penetration column and the loyalty column falls with it; Nationwide and TSB are the deviations the report itself flags, and they are the exceptions, not the rule.

Source: Ehrenberg-Bass Institute for the LinkedIn B2B Institute, Romaniuk, Dawes and Faghidno, The Double Jeopardy Law in B2B shows the way to grow, 2021, n not stated, Table 1. Redrawn by Brave New Media.

Unit: % of the bank's customers solely loyal

Show the data

Bank	% of B2B customers (penetration)	% of its customers solely loyal
Barclays	41	27
HSBC	30	29
Natwest	25	27
Lloyds	25	34
Nationwide	21	5
Santander	19	24
Halifax	16	8
RBS	13	10
Metro	11	3
TSB	10	10

The companion report in the same series asked 609 UK business banking buyers and 616 US business insurance buyers a second question, whether they would refuse to use each brand, and found the barrier to acquisition lies elsewhere: rejection averaged 11% across 22 UK banks and 7% across 17 US insurers, while the share of buyers unaware of a bank ran from 16% for Barclays to 80% for Handelsbanken [28]. A small B2B brand is small because most potential buyers do not know it, not because they have decided against it.

% unaware% who would refuse

Barclays16%10%

HSBC17%10%

Lloyds Bank22%11%

NatWest23%8%

Santander29%11%

Halifax34%8%

RBS34%19%

Metro Bank53%11%

Yorkshire Bank54%7%

Deutsche54%12%

Standard Chartered62%10%

Handelsbanken80%11%

Rejection sits near 10% whatever the bank's size; the number that moves with size is the share of buyers who have never heard of it. The barrier to acquisition is absence from memory, not a verdict against the brand.

Source: Ehrenberg-Bass Institute for the LinkedIn B2B Institute, Romaniuk, Brand Rejection in B2B, 2021, UK business banking, n=609, Table 5. Redrawn by Brave New Media.

Unit: % of business buyers unaware of the brand

Show the data

Bank	% unaware	% who would refuse
Barclays	16	10
HSBC	17	10
Lloyds Bank	22	11
NatWest	23	8
Santander	29	11
Halifax	34	8
RBS	34	19
Metro Bank	53	11
Yorkshire Bank	54	7
Deutsche	54	12
Standard Chartered	62	10
Handelsbanken	80	11

The durable B2B case is aircraft. Bennett, Anesbury and Graham (2018) used 9,000 purchase records from 51 airlines over ten years: Boeing, bought by 86% of airlines, averaged 101 purchases per buyer; Embraer, bought by 16%, averaged 37 [9][31].

Kim, Singh and Winer (2017) used A.C. Nielsen’s US household panel, 2004 to 2009, and measured the share of each brand’s dollar sales delivered by its top 20% of buyers across 238 brands in 22 categories, roughly 18,000 households [12]. The mean Pareto ratio was 0.73 with a standard deviation of 0.07; category averages ran from 0.64 for liquid detergent to 0.89 for cigarettes, and under Nielsen’s narrower brand definition the mean fell to 0.65 [12]. They note the rule “is not precisely .80” [12].

Sharp and Graham read similar data differently. On Kantar data for about 600 UK brands in 40 categories, the top 20% deliver around 40% of sales in a quarter, “over 50%” in a year and “just over 60%” over five years, and 80% of a brand’s buyers make five purchases or fewer in five years [13]. Both figures sit well below 80; the gap is windows and brand definitions.

80/20 folklore

Liquid detergent (lowest category mean)64%

Mean, narrower brand definition65%

Mean, volume definition72%

Mean, 238 brands, dollar sales73%

Cigarettes (highest category mean)89%

The number moves with the category, the brand definition and the measure, which is the point: 80/20 is a slogan, not a constant, and Sharp and Graham's shorter windows on UK data give lower figures still.

Source: Kim, Singh and Winer, Marketing Letters, 2017, A.C. Nielsen US household panel 2004 to 2009, 238 brands, 22 categories, n about 18,000 households, Tables 2 and 3. Redrawn by Brave New Media.

Unit: % of brand sales

Show the data

Item	% of brand sales
Liquid detergent (lowest category mean)	64
Mean, narrower brand definition	65
Mean, volume definition	72
Mean, 238 brands, dollar sales	73
Cigarettes (highest category mean)	89

The evidence for mental availability as a cause, rather than for double jeopardy as a pattern, is thinner. Romaniuk and Sharp (2004) set out the construct and a multi-cue measure [8]. Keller’s customer-based brand equity model (2001) had already put brand salience at the base of its pyramid: salience “influences the likelihood that the brand will be a member of the consideration set”, with depth (ease of recall) and breadth (the range of situations in which the brand comes to mind) [21]. Breadth is mental availability under another name. Keller then builds meaning, feelings and resonance on top; Sharp treats salience plus distribution as the whole job.

Binet and Field’s meta-analysis of 996 IPA Databank campaigns (2013) lands on Sharp’s side for reach and elsewhere for mechanism. Their fame campaigns turned excess share of voice into growth at an annualised efficiency of 1.4 against 0.3 for other campaigns [23]. Reach is common ground. The mechanism is not: Sharp’s account is memory refreshed, Binet and Field’s is emotional priming that builds over years [23].

The challenge

Three serious critiques and one loud one.

The first is the loyalty camp. Reichheld’s numbers are claims from company casework, not panel findings; The Loyalty Effect (1996) says “even a small improvement in customer retention can double profits” [18][19]. Aaker’s pyramid is a theorist’s structure [20]. Neither has met double jeopardy on the same data and won.

The second is the academic critique of the Dirichlet’s generality, and it comes from inside the model. Fader and Schmittlein (1993) showed across product markets in Japan and the United States that high-share brands have significantly more loyalty than the Dirichlet predicts, and traced it to consumer segments created partly by distribution [14]. Jung, Gruca and Rego (2010) replicated this across US packaged goods and found excess loyalty positively related to market share in 86% of categories, a pattern some authors call triple jeopardy [15]. Scriven, Bound and Graham (2017) sorted the recurring deviations in 62 FMCG categories into three types and concluded that “brand performance remains close to Dirichlet prediction” [16]. The model bends, it does not break, and when it bends it favours the big brand.

The third is scope. Sharp, Wright and Goodhardt (2002) themselves split repeat-purchase markets into repertoire markets, where the familiar patterns hold, and subscription markets such as insurance and banking, where most buyers use one brand at a time. The distinction “turns out to be an important boundary condition for some well-established generalisations”, and their analysis “does not extend to markets for durable products” [11]. The B2B evidence since then is real but narrower than the FMCG base: two Ehrenberg-Bass surveys with unstated samples, one aircraft dataset, and eight older studies, six from before 2001 [9]. Double jeopardy travels. The full Dirichlet toolkit, and the claim that availability is the only lever worth funding, have been tested far less outside the supermarket. Formisano, Pauwels and Zarantonello (2020) put it this way: marketers should “not oversimplify Dirichlet evidences by thinking of availability as the only (costly) response to all marketing challenges” [17].

The loud one is Mark Ritson’s 2018 Marketing Week column: Sharp’s claim that it is “impossible” to value brand perceptions “is not just mistaken, it is a rejection of true scientific method” [27]. It added no data.

We could not confirm whether the 16th edition of Kotler, Keller and Chernev’s Marketing Management (2021) covers double jeopardy or the Dirichlet at all [22]. Cialdini did not address mental availability. The availability heuristic the term echoes is Tversky and Kahneman’s (1973), a rule about judging frequency from ease of recall [25]. The nearest Cialdini gets is Pre-Suasion (2016), chapters 2 and 3, on attention and the principle that “what’s focal is causal”; that is attention in the moment of persuasion, not a brand’s standing in memory across buying situations [24].

Where practitioners get it wrong

Awareness is reported as mental availability. Sharp’s 2011 note is blunt: recall and recognition scores are “flawed by the use of a single, a-situational, cue”, usually the category name [2]. A brand can post high prompted awareness and still be absent from memory when a buyer needs a supplier for a specific job. The measure Romaniuk and Sharp propose uses one cue per buying situation [8].

Small brands are given loyalty KPIs. Double jeopardy says a 10% penetration brand will show lower loyalty than a 40% brand for no reason other than size [6][9]. A retention target set at the leader’s number is one the law forbids. The loyalty-programme test in this literature is Sharp and Sharp (1997), which benchmarked a programme’s repeat-purchase patterns against Dirichlet norms [26].

The 80/20 slide. Kim, Singh and Winer’s mean is 0.73 over six years; Sharp and Graham’s is around 0.40 in a quarter [12][13]. A plan that concentrates budget on the top fifth rests on a number the panel data does not produce.

Our own observation. In one of the supply chain group’s established markets the brand had gone years without consistent presence and had slipped out of the consideration set, with the operation intact and the alternatives long-standing. Nothing about the service had changed. Its availability in buyers’ memory had. Rebuilding it was a matter of presence, a weekly publishing rhythm per market, and the clearest lesson of the year was that consistency of spend mattered more than size of spend. Fragmented allocation stopped campaigns before they could compound. The industry survey closest to that experience is Edelman and LinkedIn’s 2024 study of 3,484 executives in seven countries: 75% said a piece of thought leadership had sent them to research a product or service they had not been considering, and 86% of decision-makers said they would invite a company that consistently publishes good thought leadership into an RFP [29]. Presence gets a brand into the room; nothing else in the survey does.

Would invite a consistent publisher into an RFPBase: Decision-makers86%

Researched a product they were not considering because of a pieceBase: Decision-makers and C-suite75%

Questioned whether to keep a current supplierBase: C-suite70%

More willing to pay a premium to the organisation that publishesBase: Decision-makers and C-suite60%

Began buying from the producerBase: Those it sent to research a product23%

The survey measures what gets an unfamiliar supplier into consideration among buyers who were not looking; the drop from 75% researching to 23% buying is the gap between being thought of and being chosen, and the first has to happen before the second can.

Source: Edelman and LinkedIn, 2024 B2B Thought Leadership Impact Report, n=3,484 executives in seven countries, fielded 30 November to 14 December 2023, pages 9, 13, 14 and 17. Redrawn by Brave New Media.

Unit: % of respondents

Show the data

Outcome	% of respondents	Base
Would invite a consistent publisher into an RFP	86	Decision-makers
Researched a product they were not considering because of a piece	75	Decision-makers and C-suite
Questioned whether to keep a current supplier	70	C-suite
More willing to pay a premium to the organisation that publishes	60	Decision-makers and C-suite
Began buying from the producer	23	Those it sent to research a product

What to do with it

1. Put penetration, the share of category buyers who bought you in the period, on the first line of every brand dashboard, with loyalty metrics beneath it [6][9].
2. Before setting a retention or share-of-wallet target, tabulate the category by penetration as Romaniuk, Dawes and Faghidno describe, and set the target where the double jeopardy line predicts for your size [9].
3. Replace the single-cue awareness question with a battery of buying situations, ask which brands come to mind for each, and report the share of situations in which you are named [8].
4. Classify your category as repertoire or subscription before choosing metrics; use purchase frequency and share of requirements in the first, defection and tenure in the second [11].
5. Test the Pareto ratio on your own customer file at one quarter, one year and three years, and stop using any figure you did not compute [12][13].
6. List every place and moment a buyer could look for you, mark each yes or no, and fix the noes before spending on anything else in this article.