What is brand equity?

What is brand equity?

It is what a brand adds to an offer beyond what the offer does. Bronnenberg et al. (2022)³ quote Farquhar’s 1989 definition of a brand (spelled Farquar in their text) as “a name, symbol, design, or mark that enhances the value of a product beyond its functional purpose” and note that the added value is often called brand equity. Their own definition takes the side of the firm: “Brand equity consists of the intangible capital that generates sustainable, incremental profitability to the firm owning the commercial rights to the brand.” They write about product brands, but add that “in many settings the two are synonymous” with corporate brands.

Marketing research describes the same thing from the buyer’s side. In their review, Bendixen et al. (2004)¹ write that “Brand equity is derived from the overall brand image created by the totality of brand associations, perceived by customers”, a point they credit to Michell, King and Reast. They report the two accounts of its sources that the field cites most: Aaker “identified four major sources of brand equity as brand loyalty, brand awareness, perceived quality, and brand associations”, and Keller “combines the sources of brand equity into brand awareness and brand image”. Among the advantages of a brand with high equity, their review lists that “a price premium can be attained”, together with higher demand, easier extensions, readier acceptance of communications, better trade leverage, larger margins and less exposure to competitors’ marketing. This guide has Aaker’s and Keller’s accounts only as Bendixen et al. report them; the books were not read for it.

In this guide, brand equity means the difference a brand makes to what buyers choose and what they will pay, compared with the same offer under a less known name; on the seller’s side, it is an intangible asset that earns profit the offer alone would not. It is distinct from brand value, which is a sum of money put on that asset; from brand awareness and brand image, which are among its sources; and from the marketing budget, which is spending that may or may not build it. Bendixen et al.¹ pass on Aaker’s point that a “price premium measure is defined with respect to a set of competitors”, and this guide applies it to every measure of brand equity: it exists only against named alternatives.

Is brand equity the same as brand value?

No. Brand value is a money estimate of the asset, and it is hard to make. Bronnenberg et al. (2022)³ write that “Measuring the full value of brand capital to a firm is notoriously difficult” and state what such a value would have to mean: “Conceptually, the value should be defined relative to the counterfactual discounted sum of future profits to the firm but-for the commercial rights to the brand and its trademark.” That raises two problems in their account: deciding what the world without the brand looks like, for instance whether the firm builds another brand or a rival takes this one, and the fact that “the but-for profit stream is seldom observed and needs to be imputed or estimated”. Commercial brand valuations are widely used, but “they all involve a mix of forecasts and judgment”, and the authors add that such measures “do depend on a number of strong assumptions”. They still read the estimates, even allowing for considerable measurement error, as showing that brand capital is a large and growing asset, and they cite other research that agrees. This guide does not use brand valuation rankings as evidence.

Nor is brand equity the advertising budget. The budget is a flow of spending; brand equity is a stock that some of that spending builds. As Bronnenberg et al. (2022)³ put it, “Not all advertising and marketing contribute per se to a persistent brand capital stock”, and their example is feature advertising of a temporary discount.

Do professional buyers pay for a brand?

The doubt is old. Bendixen et al. (2004)¹ describe the belief that “because brands are irrational, they have little significance when dealing with a corporate entity that makes buyers’ decisions on a rational basis”, against earlier work showing that “Intangible aspects such as overall supplier reputation matter even in rational and systematic decision making”. They also note, after Hague and Jackson, that “In industrial markets, the company itself is often the brand”, because branding every item in a wide range would be expensive.

To test it, they ran a conjoint analysis in one industrial market: “The setting used for the empirical study was the market for the medium-voltage indoor circuit-breaker panels.” The respondents were members of buying centres at South African mines, chemical plants, heavy manufacturers and electricity suppliers, reached through a convenience sample; sixty were approached, 57 took part and 54 were analysed. Respondents “were asked to rank-order the profiles in order of preference or intention to buy”: 16 cards, each describing a panel by five attributes. These were the brand (four makers: ABB, Reyrolle, Tamco and Relay Setting), the price (four levels from R40,000 to R60,000, which the authors give as US$4,000 to US$6,000), the delivery time (three to six months), the technology (vacuum or SF6) and the lead time for spare parts (one to three weeks). The part-worth utility of each level was estimated with MONANOVA. One of the three authors was, at publication, an engineer with ABB, the brand that came out first.

Brand mattered, but less than price and delivery. Relative importance, computed from the range of each attribute’s utilities, put delivery first at 27 per cent, then price at 24, technology at 19, brand at 16 and spare parts at 14 (Table 3, p. 375). The authors’ summary: “while brand equity has a role to play, price and delivery were more important.” They warn that these weights depend on the levels chosen, quoting a marketing research textbook by Aaker, Kumar and Day: “the magnitude of the difference in utilities is strongly influenced by the choices of extreme levels”, and the price levels spanned 50 per cent while delivery spanned 100 per cent. They still read 16 per cent as large, because “Some industrial projects are won or lost on very tight margins, often below 5%.” The comparison is loose: 16 per cent is the brand’s share of the preference ranges, not a margin or a price difference, and their own conversion below puts the leader’s premium at 7.4 per cent.

How much of the price did the brand carry?

About 7 per cent, by the authors’ conversion, above the price at which price utility is zero. Figure 1 shows the brand and price utilities from their Table 2 on one scale.

By linear interpolation between the price levels, “a price of R50,707 would have a utility of zero.” The leading brand’s utility of plus 0.528 is offset by a price utility of minus 0.528 at R54,453, so “the most preferred brand could command a price premium of R3746 or 7.4%” relative to R50,707, which the authors call the average price, setting average in quotation marks: it is the price of zero utility, not the mean of the four price levels, R50,000. The least preferred brand, at minus 0.542, “would have to discount its price by R3305 to R47,402 or 6.5%”. Between the two, “the brand leader could afford a price premium of nearly 14% over relatively unknown brands on the market”. In the authors’ words, the buying centre members, even for a standard product, “develop a sentimental relationship for which they are prepared to pay a price premium”.

The conclusions give a different first figure: the leading brand “could command a price premium of 6.8% over the average industrial brand and 14% over a new, unknown brand”. On this guide’s arithmetic, R3,746 is 7.4 per cent of R50,707 and 6.88 per cent of R54,453, so the 6.8 per cent may state the same premium as a share of the higher price, cut rather than rounded; the paper does not say, and its 14 per cent in the same sentence is on the lower base. This guide uses the 7.4 per cent of the derivation. Either way, the premium is a reading of ranked cards, with no option to choose none of them, not of prices paid. It assumes the other attributes are equal and that utility changes in a straight line between price levels, and it holds only for these four makers. The conjoint analysis essay explains why a preference measured in a designed choice is not yet demand.

Who in the buying centre gave the brand most weight?

The people who use or specify the equipment. The 54 respondents held five roles: 8 buyers, 9 gatekeepers, 12 users, 14 technical specialists and 11 decision makers (Table 1, p. 374). The brand’s share of importance was 28 per cent for users, 24 for technical specialists, 19 for decision makers, 16 for buyers and 7 for gatekeepers (Table 4, p. 375), and technical specialists ranked it level with price. The authors read the gatekeepers’ low figure as difficulty telling technical brands apart. Setting the technical specialists’ own utility for the leading brand, plus 1.230, against the price scale, they find that the leading brand “can increase its price up to R64,020” and that “the technical specialists are prepared to grant a price premium of 26% to the leading brand name”.

Two cautions apply. R64,020 is above the highest price on the cards, R60,000, and it is the figure that results from extending the whole sample’s price utilities past that point. The 26 per cent is therefore an extrapolation, not a reading inside the tested range, and it sets one group’s brand utility against all 54 respondents’ price utilities, although the specialists weighted price as heavily as brand (24 per cent each). And the authors did not find that the roles differ reliably: their conclusions state that “There is no significant difference among the role player groups” and that “Proposition 5 is not supported”, although the sentence before reads the other way. With 8 to 14 people per role, the role figures describe this sample. The buying committee guide maps the roles a B2B purchase can involve, in its own taxonomy rather than the five roles used here.

Asked to rate nine given criteria for choosing their preferred brand, respondents put quality, reliability and performance first, while “Price is third last ahead of supplier’s reputation and relationship with supplier’s personnel.” The authors conclude that “Quality is the main brand-equity-generating variable”, meaning quality as buyers perceive it, for which they quote Aaker’s definition: the “customer’s perception of the overall quality or superiority of a product or service with respect to its intended purpose, relative to alternatives”. The brand also travelled: the authors report that every role would recommend its preferred brand, with ratings per role of 4 to 5 on a five-point scale, and for a product line none of the four makers then sold, “the consideration given to new products was biased in favor of well-known brands”, with equal medians rejected at the 5 per cent level. In their Table 5 the median scores follow the brands’ utility (5, 4, 3 and 3), and the text says the ranking by summed score does too, but the table prints 252 for the least preferred brand against 234 for the leader. With a median of 3 on a five-point scale, 54 answers cannot sum to more than 214, so the 252 is probably a misprint and the summed ranking cannot be read from the table. The authors’ lesson: “Industrial marketers have to translate quality into perceived quality.”

How do brand preferences form, and how long do they last?

Through past experience, and for decades, on the strongest evidence, which comes from consumer markets. Bronnenberg et al. (2012) use the phrase brand preferences “as a shorthand for willingness to pay” and start from the observation that “Consumers appear to have high willingness to pay for particular brands, even when the alternatives are objectively similar.” They matched the life histories of “more than 38,000 households” to their “purchases in 238 consumer packaged goods product categories” and used moves between United States states to separate a household’s past from what is on offer where it now lives.

When a household moves, “approximately 60 percent of the gap in purchases between the origin and destination state closes immediately”, which the authors attribute to what is on offer in the new state. The rest is carried by the household’s past, so they assign about 40 per cent of the geographic variation in market shares to persistent brand preferences. That remaining gap closes slowly: “It takes more than 20 years for half of the gap to close, and even 50 years after moving the gap remains statistically significant.” Preferences are not fixed in childhood, since “consumers who move after age 25 still eventually converge to the consumption patterns of their new state of residence”. The conclusion is worded with care: “Our results suggest that much of consumers’ observed willingness to pay for brands may reflect the influence of past experiences.”

The consequence is a first-mover advantage: whoever was there first keeps a lead. In a simulation of two brands that are “identical in every respect except their prices”, “a head start of even a few years creates a formidable barrier”: “If A’s head start is five years, B would need to discount its price by 18 percent to reach market share parity in just more than a decade.” Brand capital weighed more in categories with heavy advertising and high social visibility, though the authors interpret neither split as causal; for advertising they write, “We cannot interpret this difference as causal”. How fast such capital wears out, and why the published rates differ so widely, is the subject of the essay Every growth budget is a gross number, and this guide does not repeat it.

How large is brand capital next to the marketing budget?

Roughly one year’s advertising, on the assumptions these authors adopt, and the assumptions decide the answer. Bronnenberg et al. (2022)³ report that “US corporations expensed $354 billion in advertising spending in 2018, or 1.7 percent of GDP”, a figure that does not include “non-advertising-related brand investments”. Extending an earlier national series, they estimate “the total 2018 US brand capital stock to be around $350 billion, more than double the $160 billion estimated (real) stock in 1995”. This is what the authors call advertising-driven brand capital, a different stock from the past consumption in the previous section: at the 55 per cent rate below, more than half of it is gone within a year, while half of the movers’ gap took more than 20 years to close.

The ratio of stock to spending rests on two numbers: the share of spending that builds capital lasting beyond the current period, which they set at 0.6 after earlier work, and the rate at which that capital decays. With the 55 per cent annual rate they adopt, “the calculated advertising capital stock is roughly the same size as current advertising spending”. On that rate, “The debate has by no means been resolved”, so the ratio is only as firm as the assumption behind it.

Two limits matter in business markets. The series counts advertising, and the authors leave aside other marketing that builds relationships with customers, noting that “salesforce costs represent an additional 5 percent of GDP”. And expertise need not discount a brand. In packaged goods, a study they cite finds “pharmacists and physicians are considerably more likely to choose store-brand headache medicines” than similar shoppers who lack the expertise to see that the products do not differ objectively. In the switchgear study the technical specialists were the only group to rank brand first, level with price, though users gave it more weight and the role differences were not significant. On this guide’s reading, expert buyers may pay for a brand where they believe its products differ and decline to where they know they do not; neither study tests that. A ratio of revenue to marketing spend in one period cannot separate what the stock earns from what that period’s spending earns; the MER guide explains what such a ratio does measure.

Which six facts make a brand premium claim checkable?

The rivals it is measured against, the other attributes held equal, the price range tested, the premium in money, the roles whose answers it rests on, and the source the brand’s equity is thought to come from. One row per claim. The fields follow Bendixen et al.¹’s conversion and its caveats, the sources of brand equity they report, and Bronnenberg et al.³²’s finding that preferences built by past experience persist.

An illustration, with no company data: a vendor of field service software wants to know whether its name supports a higher renewal price than two smaller rivals. The record names the two rivals and adds an unnamed new entrant as the unknown brand. It lists the other attributes on the cards, such as implementation time, integrations and support response, and sets price levels that run past the premium the vendor hopes to charge, so that the premium is read inside the tested range. It states the premium in money at equal utility and splits the answers by role, with the number of respondents in each: the operations staff who use the product, IT and procurement. It records a guess at the source, such as quality as users experienced it, and whether the premium is claimed for new buyers, who have no experience of the product, or for renewing customers, who do. For renewing customers it also separates the brand from the cost of switching: data migration, retraining and contract terms make the incumbent dearer to leave whatever its name, and that part of a renewal premium is not brand equity in this guide’s sense. Bronnenberg et al.³²’s evidence on past experience comes from households buying packaged goods, not from firms renewing software. These are the author’s examples and have not been tested.

What can switchgear buyers and grocery shoppers not tell a software seller?

How large a brand premium is when software is bought, and how long it lasts. Bendixen et al. (2004)¹ studied one product in one country with 54 people from a convenience sample, who ranked cards rather than placed orders; one author worked for the brand that came first; and the paper’s figures disagree in places, with 6.8 against 7.4 per cent and three delivery levels in one section against four in the design. Bronnenberg et al. (2012)² studied United States households buying packaged goods, where the buyer is usually the user and buys often. Bronnenberg et al. (2022)³ is a review, and its estimate of the stock rests on an assumed share of spending that builds capital and an assumed rate of decay.

On this guide’s reading, one thing carries over as method and two only as hypotheses. The method: state brand equity against a named set of rivals and in money at equal utility, inside the prices tested. The hypotheses: that in a B2B purchase a brand can carry a price premium even for a standard product while buyers weigh it below price and delivery, which rests on one convenience sample of 54; and that preferences built by past experience persist and favour the incumbent, which rests on households buying packaged goods. Whether a premium of the size Bendixen et al.¹ found holds for subscriptions, whether renewing customers carry preferences the way households that move do, and how a buying committee combines its members’ views of a brand, the sources do not say.

References

1. Bendixen, M., Bukasa, K. A., & Abratt, R. (2004). Brand equity in the business-to-business market. Industrial Marketing Management, 33(5), 371–380. https://doi.org/10.1016/j.indmarman.2003.10.001
2. Bronnenberg, B. J., Dubé, J.-P., & Gentzkow, M. (2012). The evolution of brand preferences: Evidence from consumer migration. American Economic Review, 102(6), 2472–2508. https://doi.org/10.1257/aer.102.6.2472
3. Bronnenberg, B. J., Dubé, J.-P., & Syverson, C. (2022). Marketing investment and intangible brand capital. Journal of Economic Perspectives, 36(3), 53–74. https://doi.org/10.1257/jep.36.3.53