Ask a room of managers what a brand is and you'll hear about logos, promises, reputations and gut feelings. The research literature is more precise. Drawing on associative network models of memory (Collins and Loftus, 1975; Anderson, 1983), Keller (1993) treats a brand as a node in memory to which a web of associations is linked. Brand knowledge then has two components: brand awareness, whether the node can be retrieved through recall or recognition, and brand image, the strength, favourability and uniqueness of the associations attached to it. A name, logo or strapline is a cue that activates this network; it is not the brand itself. The uncomfortable consequence for marketers is that the brand lives in customers' heads, shaped by product experience, word of mouth, price and retail placement as much as by advertising.

Brand equity is the value these memories add. Aaker (1991) framed it as a set of assets and liabilities linked to a brand's name and symbol — brand loyalty, name awareness, perceived quality and brand associations, plus other proprietary assets such as patents, trademarks and channel relationships — that add to or subtract from the value a product provides. Keller (1993) narrowed the lens to customer-based brand equity: the differential effect that brand knowledge has on consumer response to the marketing of the brand. You see it when people pay more, search less or forgive more for the branded offer than for an identical unbranded one. Financial brand equity, by contrast, is the value of the brand to its owner, expressed as incremental cash flows (Simon and Sullivan, 1993). The two are linked, since customer-based equity is the source of financial equity, but they are measured differently, owned by different functions and can diverge in the short run.

Two limits deserve attention from the outset. The concept is contested: reviews of the literature find little agreement on its dimensions or how to measure them (Christodoulides and de Chernatony, 2010). And the associative view has a behavioural challenger. Sharp (2010) argues that brands grow mainly by being easily noticed and recalled by large numbers of light buyers, rather than through deep or differentiated meaning — a debate you'll pick up in Module 2.

Keller's (2001) customer-based brand equity model turns the definition into a diagnostic. It arranges six building blocks in a pyramid that a brand climbs in four steps, each answering a question customers implicitly ask. Identity ("who are you?") rests on salience: how easily and often the brand comes to mind across purchase and consumption situations, a matter of breadth as well as depth of awareness. Meaning ("what are you?") has two routes: performance, the functional associations of reliability, durability, design and price; and imagery, the more abstract associations of user profiles, usage situations, personality, heritage and values. Response ("what about you?") splits into judgements of quality, credibility, consideration and superiority, and feelings such as warmth, excitement, security, social approval and self-respect. At the apex, relationships ("what about you and me?") becomes resonance: behavioural loyalty, attitudinal attachment, a sense of community and active engagement (Keller, 2001; Keller and Swaminathan, 2020).

The left-hand side of the pyramid, performance and judgements, is a rational route; the right-hand side, imagery and feelings, an emotional one. The model's logic is sequential: resonance is hard to build for a brand that isn't salient or whose performance is doubted. That makes it a useful diagnostic. A brand with high salience but weak meaning is known but not understood. One with strong performance but thin imagery and feelings may win side-by-side comparisons, yet it's easily substituted when a rival matches its specification.

Use the model with care. It is a framework rather than a tested causal law, and the strict sequence is an assumption more than an established finding. Romaniuk and Sharp (2004) argue that salience should mean the propensity to be noticed or thought of in buying situations, not simply top-of-mind awareness, which changes what a strong identity looks like. Resonance may also be an unrealistic goal in low-involvement categories where most buyers are light, repertoire buyers (Sharp, 2010). Fournier (1998) shows that consumer–brand relationships are real and varied, but not every customer wants one. Treat the pyramid as a structured set of questions, weight the building blocks for your category, and support every score with evidence.

Once a business sells more than one product, it has to decide how its brands relate to one another: which names appear, how prominently, and what each is asked to do. Aaker and Joachimsthaler (2000) organise the options in a brand relationship spectrum. At one end is the branded house, where a single master brand spans every offer with descriptive names; Virgin is their best-known example. At the other end is the house of brands, a stable of stand-alone brands with little visible link to the parent, as in Procter & Gamble's portfolio. Between them sit endorsed brands, where a parent lends credibility to a distinct brand through a "by" or "from" device, and sub-brands, which modify the master brand's associations while leaning heavily on its equity.

Each option trades efficiency against focus. A branded house concentrates investment, makes every product a signal for every other and simplifies new launches. But it also spreads risk across the whole portfolio, since a failure or scandal anywhere touches everything, and it limits how far the brand can stretch before its meaning blurs. A house of brands allows sharp positioning for different segments, keeps channel conflicts apart and quarantines risk, but it multiplies marketing costs and gives up shared equity. The evidence is suggestive rather than decisive. In a study of 113 US firms, Rao, Agarwal and Dahlhoff (2004) found that corporate branding was associated with higher Tobin's q and mixed branding with lower Tobin's q. As a correlational study, it points to a pattern rather than a rule.

Architecture also means giving brands roles in the portfolio. Strategic brands carry future growth, flanker brands defend against low-price entrants, cash cows fund the rest, and silver bullets are brands or sub-brands that positively influence the image of another brand, often the master brand (Aaker, 2004). For any new offer, the practical tests are whether the master brand's associations would help it, whether it would help or harm the master brand, and whether it has a compelling reason to stand alone. Few portfolios are designed from scratch. Most reflect acquisitions and history, so a good recommendation sets out a migration path as well as an end state.

Financial brand equity matters because brands are bought, licensed, taxed, litigated and reported. Under UK-adopted international accounting standards (IAS 38), and similarly under the UK's FRS 102, an acquired brand can be recognised on the balance sheet, but an internally generated brand cannot, so the most valuable brands a company builds itself are often invisible in its accounts. Valuation fills that gap. ISO 10668:2010 sets requirements for monetary brand valuation and recognises three approaches: the market approach (what comparable brands have sold or licensed for), the cost approach (what it would cost to create or replace the brand) and the income approach (the present value of future economic benefits attributable to the brand). The income approach dominates practice, often through the relief-from-royalty method, which asks what the owner would have to pay to license the brand if it didn't own it (Salinas, 2009). ISO 10668 is being revised, so check the current edition before relying on it. A companion standard, ISO 20671-1:2021, which replaced ISO 20671:2019, covers brand evaluation more broadly, linking inputs such as brand resources and management to outputs that include customer and financial measures.

Valuations are highly sensitive to their assumptions: the royalty rate or the brand's share of earnings, the discount rate, forecast growth and the horizon. The published league tables use different methods and often disagree about the same brand's value, so treat any single figure as an estimate and always quote its source and year. The market approach is limited by how few comparable brand transactions are disclosed; the cost approach measures what was spent, not what the brand earns, so it tends to understate strong brands and overstate weak ones.

For strategy, valuation is most useful as a direction. It links the CBBE diagnosis to money, justifies long-term investment and informs architecture and licensing decisions. A brand audit combines an inventory of everything the brand currently does with an exploratory study of what customers think (Keller and Swaminathan, 2020); the tracking side of this will be developed in Module 7. AI tools now speed up the exploratory part: social listening and automated text analysis can map associations across thousands of reviews and posts (Netzer et al., 2012; Humphreys and Wang, 2018). Their limits are real, though. People who post online are not a representative sample, sentiment classifiers misread sarcasm and category jargon, and generative AI summaries can invent themes. Validate what they find against a properly sampled survey and a sample of the raw text.