Aaker or Keller: which brand equity model answers which question

Origin

Before 1991 a brand was mostly a design problem. Aaker made it a balance sheet problem.

His definition is deliberately double edged. Brand equity is the set of assets “and liabilities linked to a brand” that add to or subtract from the value a product delivers to customers and to the firm [1]. The liabilities half gets dropped in almost every retelling, which is the first mistake. A brand can carry negative equity, and plenty do. A name that makes a buyer hesitate is not a neutral asset sitting at zero. It is a cost.

Keller arrived two years later with a sharper instrument and a narrower claim. Customer-based brand equity is the differential effect of brand knowledge on consumer response to the marketing of the brand [2, p. 2]. In plain terms: do people respond better to this offer because your name is on it than they would to the identical offer with no name? That difference is the equity. It is measurable, and it can be negative.

Brand equity

• Brand loyaltyValue of the existing base, and how exposed it is
• Brand awarenessRecognised or recalled where buying happens
• Perceived qualityThe judgement that supports a price premium
• Brand associationsAttributes, benefits, feelings, people, occasions
• Other proprietary assetsTrademarks, patents, channel relationships

Four dimensions are perceptual. The fifth, proprietary assets, is legal, which is why it is greyed. The model is an asset map, not a scorecard.

Source: Illustrative. After Aaker, D. A., Managing Brand Equity, Free Press, 1991. Brave New Media, 2026.

1. Resonance4 · What about you and me?
2. Judgements+ Feelings3 · What about you?
3. Performance+ Imagery2 · What are you?
4. Salience1 · Who are you?

The order is the point. Each level is the foundation of the next, so a brand cannot buy resonance while it is still failing salience.

Source: Illustrative. After Keller, K. L., Building Customer-Based Brand Equity, MSI Report 01-107, 2001, and Strategic Brand Management. Brave New Media, 2026.

What the evidence shows

Start with what each model actually is, because this is where most summaries go wrong.

Aaker’s five dimensions are brand loyalty, brand awareness, perceived quality, brand associations, and other proprietary assets such as trademarks, patents and channel relationships [1]. It is an asset map, not a scorecard. The strategic question is never whether all five are high. It is which of them creates value in your category, and which you are paying to maintain for no return. The fifth dimension is the odd one out and should be treated that way: proprietary assets are legal, not perceptual.

Keller’s 1993 paper sets out the definition and the structure of brand knowledge, which he splits into brand awareness and brand image [2]. Note what is not in it. The four-level pyramid that everybody attributes to “Keller 1993” is not in the 1993 paper. It appears later, in the 2001 Marketing Science Institute report and in Strategic Brand Management [3]. The pyramid has four levels, each a question the customer is implicitly asking, built on six blocks: salience, performance, imagery, judgements, feelings, resonance.

Who are you. What are you. What about you. What about you and me.

That misattribution is not pedantry. People cite the 1993 paper to justify a pyramid exercise and never read either source, which is how a diagnostic tool becomes a workshop template.

The order in the pyramid is the part people skip. Each level is the foundation of the next, which means a brand cannot buy resonance while it is still failing salience. Most rebranding projects are an attempt to do exactly that: fix the top because the bottom is slow and expensive.

Aaker

1991 · five dimensions

• Brand loyalty
• Brand awareness
• Perceived quality
• Brand associations
• Proprietary assets

What is the brand worth to the firm?

Keller

1993 · four levels, six blocks

• Salience
• Performance, Imagery
• Judgements, Feelings
• Resonance

What is the brand doing in the buyer's head?

Side by side, ending on the question each model actually answers. Pick by the question, not by the model you met first.

Source: Illustrative. Brave New Media, 2026.

The two men were not rivals and worked together. Aaker and Keller published jointly on brand extensions in the Journal of Marketing in 1990, a year before Managing Brand Equity appeared [4]. Keller and Lehmann mapped what the field had settled by 2006, which is the best single place to see how the two traditions sat together once the argument had cooled [5].

The challenge

Both models share a weakness, and it is worth stating before the practical section rather than after it.

Neither has a measurement standard. Aaker gives you five categories and no instrument. Keller gives you six blocks and no scale. What exists in practice are proprietary trackers, each with its own questions, each incomparable with the next. A brand equity score from one agency and a brand equity score from another are not the same number, and frequently not the same construct.

That is a real limitation and it explains the persistent gap between brand equity as a concept and brand equity as a line anyone will defend in a budget meeting. The financial valuation tradition, Interbrand and the ISO 10668 standard, sits outside both models and answers a different question again: what would somebody pay for this name. Do not mix the three and call it measurement.

The Ehrenberg-Bass position goes further. If growth comes mostly from penetration and availability, then much of what these models measure is downstream of brand size rather than a cause of it. That argument is covered in Double Jeopardy and in the Principles paper on mental and physical availability. It does not demolish either model. It does mean a high loyalty score is weaker evidence than it looks.

Where I see this go wrong

They run the pyramid as a workshop instead of a diagnostic. The pyramid is a sequence with a dependency. Used properly you score the six blocks, find the lowest, and start there. Used badly, a room of people agree the brand should have more resonance, and the output is a tagline.

They treat awareness and equity as the same thing. Awareness is one of five dimensions in Aaker and one half of brand knowledge in Keller. A brand with high awareness and bad associations has high awareness and negative equity. This is the single most common error and it is usually baked into the tracker before anyone asks a question.

They pick the model that fits the meeting rather than the question. If you are in front of a finance director, Keller’s six blocks will not help you, because the question is what the asset is worth. If you are trying to explain eighteen months of flat growth, Aaker’s five dimensions will not help you, because the question is where in the sequence it broke.

I teach this badly more often than I would like. When I take marketing students through brand equity at the Bucharest University of Economic Studies, the temptation is to present the two models as a sequence, Aaker then Keller, as though one leads to the other. They do not. They are two instruments pointed at two different questions, and teaching them in order is what produces a generation of marketers who reach for whichever one they remember.

The honest version takes longer and lands worse in a lecture. It is still the right one.

What to do with it

1. Decide which question you are asking before you open either model. Write it down in one sentence. If the sentence contains the word “and”, you have two questions and you need both models.
2. For a budget argument, use Aaker. Produce the five dimensions as a map, and name the liabilities explicitly. A brand audit with no liabilities in it has not been done honestly.
3. For a growth diagnosis, use Keller. Score the six blocks, identify the lowest one, and allocate against that rather than against the one your agency finds most interesting.
4. Separate awareness from equity in every tracker you run. If your dashboard reports one number called brand health, it is reporting nothing.
5. Never present a brand equity score next to a brand valuation figure without saying they measure different things. One is perception, one is money, and the slide that puts them side by side will be quoted out of context.
6. Before any rebrand, check salience. If the bottom of the pyramid is weak, a new identity will not fix it and will cost you the recognition you had. See Distinctive brand assets.
7. Re-read the original of whichever model you use. Both are short. Most of what circulates as Aaker and Keller is third hand.

If you run the company rather than the marketing

You do not need either model. You need to know which one your team is using and why, because the two produce different answers and only one of them fits the question you are asking.

If you are deciding whether brand spend is worth defending to a board or a buyer, that is an asset question and it is Aaker’s. Ask for the five dimensions written out, liabilities included. A brand audit that comes back with no liabilities in it has not been done honestly, and you should send it back.

If you are asking why a category-leading product is not growing, that is a perception question and it is Keller’s. Ask where in the sequence it breaks. If the answer is salience, no amount of new identity work will fix it, and anyone proposing a rebrand as the remedy is selling you the expensive version of the wrong solution.

One question to put to whoever owns marketing: show me awareness and equity as two separate numbers. If the dashboard reports a single figure called brand health, it is reporting nothing, and you have been approving budget against it.