Reject How Brands Grow and a lot of marketers would probably look at you like you were denying gravity.
While someone will occasionally query whether Bryon Sharp’s laws of growth apply in specific environments, the basic premise of his book is unassailable, as far as many people in the industry are concerned.
But Felipe Thomaz, an associate professor of marketing at Oxford University’s Saïd Business School, is not one of those people.
Thomaz, who has published numerous papers on marketing effectiveness in major academic journals, argues that Sharp’s model ‘breaks branding’ by ignoring differentiation and asserts that most marketing scholars have never even heard of physical or mental availability.
We contacted Thomaz last week to discuss a paper he’d co-written on cross-media synergies. At a conference last month, Sharp had dismissed the phenomenon as a ‘bit of a myth’ and we wanted to know what Thomaz thought about his comments. But it turned out that he had a lot more to say, about the complexity of brands and the ‘beautiful game’ that marketers play.
What’s your view on Byron Sharp saying cross-media synergies are ‘a bit of a myth’?
That view is inconsistent with the marketing literature and marketing science. We've known that these things interact powerfully for a very long time. Holding the idea that it doesn't is weird to me. Somebody that knows how marketing works would know that these things exist, at varying levels of strength for different conditions.
Is there a chance that you and Sharp are at cross purposes over definitional differences rather than substantive ones?
This is not a difference in understanding. If we're to pick words apart, I don't understand what one means by ‘bit of a myth’. Which bit? It's unclear. And that's, again, weird. Maybe it's related to the audience he was speaking to, but it's not an academic treatment. From everything that he said, the only thing I can get out of it is that he's ignoring about 60 years of published work, if not more. You can disagree with it, but it's published, so you have to actually publish a counter. That's the science part.
The beauty and the complexity of marketing is respecting the fact that humanity and social systems and markets are dynamic systems. They change over time. Effect sizes change, interactions change. We know these dynamics take place and that gives you room to challenge existing knowledge, but to do so you have to offer proof.
To say ‘this is a myth’ and ‘do something simpler’, to me, is really, really dangerous. It's an oversimplification, and oversimplification in a competitive scenario is a very dangerous ploy.
Felipe Thomaz, Oxford University, Saïd Business School
You talk about marketing being dynamic, but are there fundamental principles – Sharp might refer to mental and physical availability, others might refer to brand building and activation – that stay the same?
Those are very, very dangerous assumptions and simplifications in a dynamic social system,
I'm lucky to interact with managers daily and they all want it simpler, ‘What's the guide, is it 60/40 is it 70/20/10?’ Everybody wants that simpler thing, as if there was a singular answer for every company, in every market in every vertical, regardless of growth stage.
Is my startup strategy the same as Unilever, is that what you're telling me? These are not static games, these are competitive games. The second that you say everybody should do X, you have just destroyed everybody's strategy, because everybody will do the same thing. I think it's as simple as that. As soon as you say, ‘everybody must follow this rule’, the incentive is for me to do something else, because then I'll win.
You’re talking about marketing spend but does what you’re saying about laws apply to something as fundamental as the concepts of mental and physical availability?
Most marketing scholars won't even know what you're talking about when you say those words.
Those ideas don't exist in marketing science. They are Byron Sharp’s popular presentation that essentially most academics have never heard of, especially if they have less interaction with practitioners. But in published work in the top journals, those ideas have never existed. Most scholars won't be aware that those things exist, and that's where confusion starts to come in. Because they're like, ‘Does that just mean salience?’
There's grains of truth in what he's saying. But in my reading of his book, he's selected carefully around existing frameworks and left out important details.
To me it looks like he uses [Kevin Lane] Keller's brand framework from 1993 but removing differentiation. But by removing differentiation, he's broken branding.
[How Brands Grow] is built on Andrew Ehrenberg’s work from the 1960s, which is published and peer reviewed but deals with static market shares, which is the opposite of growth. You cannot grow in a static market.
Andrew Ehrenberg did good work, but you can't use those ideas to explain how you grow. If you read Ehrenberg's work, his laws aren’t about how you do marketing and how you grow a brand. It's about how a steady, stable system works. For instance, why people choose products that have more facings on a shelf.
And all those things make sense, but if you take those ideas, and you try to apply it to something that breaks those assumptions, like growth, and then you remove differentiation, which explains growth in brands, you end up with Byron Sharp’s ideas, which unfortunately don't work.
Could you explain a bit more about why you don’t think they work?
Ehrenberg had to make a few assumptions/simplifications about the world in order for his formulas like the Dirichlet to work, and to generate his results like the double jeopardy. First, that brand market shares do not change. So, in this simplified view of the world, brands are locked in a static, stable market. Second, he assumes that brands are undifferentiated, and do not serve different segments.
Simplifications like these are common and needed in academic work, attempting to bring rigor and definitive answers to complex social environments. However, the results of these models are only useable if you agree with these conditions. By extension, the laws, double jeopardy, etc, are useful if, and only if, you agree that your brand’s market share will not change, and all brands in your market are undifferentiated.
Here's where the logic breaks down. How can you use results of a model that requires no growth in order to explain and offer guidance on growth? You can’t, because the model intentionally ignores changes in market share. How can you recommend that brands should be undifferentiated, based on a model that doesn’t take differentiation as a possibility? Again, you can’t. It’s not an emergent strategy that comes from data, it’s a restriction placed on the analysis.
Could a firm replicate Sharp’s and Ehrenberg’s results? Sure! Provided they are already large, operating in a static marketplace with other large competitors, and everyone is largely identical. But for everyone else, this will break down. Altogether, the work highlights the advantages of being large, but cannot address how to grow.
Interestingly, Andrew Ehrenberg himself made this very clear in his article published in Nature in 1993. In that text he describes other market influences, such as differentiation, advertising expenditure, price, distribution, and beyond as forces that explain differences in market share. Similarly, in his book on repeat buying he cautions that these results explain general in-market behaviors but are not guidance in the execution of marketing. Those warnings from the foundational work seem to have been lost.
So the consensus among the academic marketing community is that differentiation is vital to growth?
Yeah, the rest of academia has published saying that differentiation is how you actually grow.
The consequence of the opposite is the slow decay and destruction of the brand over time because if you work through the logic, what you get is commoditisation.
That's what the evidence has shown over 100 years. Growth is a function of perceived differentiation. Distinctiveness doesn't work. There's work with Google on YouTube looking at thousands of ads, and distinctiveness of assets doesn't break into the top 10 characteristics that drive performance.
And again, is this not a case of being at cross purposes and describing the same levers of growth with different terms?
Not at all. The one asterisk that I'll use is that, because How Brands Grow is not an academic text, the definitions and word use is not very precise. There's not a clear measurement of availability. Somebody could just say, ‘It's pure salience’. Is it? I don't know. It's using mental structures and some of that architecture that we know works, but it's unspecific enough that we could all have different ideas of what it means. And that on its own is dangerous because what you end up with isn't science. If I have to go back to the author to get validation that what I'm doing is correct, that's guru work. If the owner of the idea has to continuously explain how the idea works, the idea doesn't stand on its own.
If it was an academic text, it would be very precise, and I’d be able to either attack or defend the ideas better, but they’re so unspecific that I could say, I am doing mental and physical availability and you could say you're doing the same thing, and we would be doing completely different things.
Inside of it are grains of truth because he's using aspects of known results, but to me, once you look at the whole, it just ignores all modern knowledge of marketing.
Do you know why How Brands Grow is popular? One reason is that it’s simple, but also because he says it's different. So differentiation is why he has grown.
Felipe Thomaz, Oxford University, Saïd Business School
How has it felt to see How Brands Grow become so widely referenced among practitioners? Frustrating?
It's not frustrating. It's just a weird quirk of how true marketing science has evolved versus practitioners.
Over here, we're using very complex and novel mathematics to deal with the questions that we have from Google, Meta, the United Nations. And on the other side of the world it’s like, ‘But this one simple trick – buying more reach solves all of your marketing problems.’
If you had simple laws that were persistent and universal, do you need a marketing manager?
If all you need to do is get more X…I can do that in my sleep. If the world is not complex, I have no need for most of the labour in this industry.
Is there a set of principles about how marketing works that marketing scholars have agreed upon?
Yeah, there is. It's not like marketing is this amorphous beast that no longer makes sense. I would think of it as a medical textbook. Your doctor is dealing with the very complex environment that is your body. He doesn't prescribe vitamins for everything that goes wrong with you. It's not one-size-fits-all. But we know how the pieces fit together, how the interactions happen. There are a few unanswered questions, but very largely we know how branding works. We’ve known the components of branding and the structure of it since about 1993.
And we know how diffusion works, largely: the process of adoption of new products or ideas. That was from Frank Bass’ work.
We know how influence works. We know how to manage consumer journeys. We know how decision making very often works, although that's not straightforward. It's not as simple as an AIDA model. We know how competition works. We know how market structures work.
All of those very large constructs, we have a very significant understanding of how they function. Scientists pick apart minutiae inside of those larger constructs, but I don't think those very large things are going away, even in a dynamic system.
Does diffusion and product adoption go away? Does new product development? No, those are too big of an idea, but how they function differs. You know, how does that look in a high-inflation world? How does that look in a world that’s moving towards deglobalization?
Those dynamics keep things really sexy and interesting, because you have to deal with change. But you're still thinking, ‘Okay, how does diffusion work with this?’ My favourite model was published in 1969, that was pre-internet – by quite a margin. How does social influence change the diffusion of ideas now that you have a highly interconnected world? You’ve probably heard of six degrees of separation. That's how word of mouth would have worked. Our world today is at about three. Suddenly you're in a very different environment.
And we’re trying to run a marketplace on top of that social system. I think you're fine if nobody else is respecting it, either. If everybody's dumb, it's fine. But the second that one person says, ‘if I do this better than you, I’ll take market share’, then they will do that thing. If everybody's playing the same game, the incentive is to do something else.
Felipe Thomaz, Oxford University, Saïd Business School
So how does branding work in your view?
We think of brands in a branching model. Brand is going to split into two initial components. One is awareness – is this a well-known brand? – and the other one is image – what is your opinion of that brand?
And each one of the components will branch further into their different components of image and different components of the awareness.
Awareness is not as simple as ‘do you know this brand?’ You can have recognition and recall. Recognition is super important in Old-World marketing when you're walking down the supermarket aisle and you have visual cues. However, if you're staring at a Google search bar, recognition is not going to help you in the least. You need to recall the brand without help.
And then image branches into aspects of associations the brand has: types, favourability, strength, and differentiation. These association types can branch out further into product related, emotion related, location related, experiences, symbols, and so on.
You can see how it forms a tree, right? So, in my reading, this is the model that gave rise to Sharp’s mental availability. Except that Sharp dismisses the need for differentiation – where it lives inside of brand image – and bundles all of the other moving parts into a single item: mental availability. This coarse bundle is the combination of all of these branches, back to its original root – the brand. If you pick that apart, when you say a brand needs more mental availability, you’re really saying a brand needs more brand, without explaining how the internal components are organised, nor how to achieve it – and leaving out a key ingredient of brand growth.
This oversimplification, in my reading of it, leads to a system where you're saying. ‘If you want to grow, just sell more’. It's a definitional statement that is inherently true – the brand is in fact a brand – but dangerously incomplete and offering no guidance to marketers trying to do marketing. It sells well to people with no understanding of marketing, who can diminish the science to the acquisition of this one undefined ‘availability’. And that's the only thing that bothers me on that whole concept. But in my experience, most high-end marketers have moved well past it, despite the noise around it.
And what’s the thing that they have moved towards? The more complex systems that you describe?
Yeah, it’s managing complex systems, and respecting and exploiting those. There's a lot of money on the table in that complexity. I think what they’ve found is, ‘If I do this differently then I make more’. And if you pull on that string long enough, you're very far away from a simple world.
It doesn't become a boiled ocean kind of problem, which is, ‘Oh, it's so complex, I can't handle it’. It's small, incremental things over time. So, it's not daunting once you're working on it.
How does creativity as an advantage sit within the complex system? Is it part of the discussion?
It's a huge discussion. I've been working on creative measurement and inclusion of creative quality and creativity in marketing measurements for a number of years now. It's complicated because if you go into any creative agency and ask what creativity is and how you measure it, you go very quickly to this magical-thinking world where creativity is this thing that cannot be contained. And yet the same team and your agency does it every single time. You know who the talent is. So there is a system.
There's a new paper in the Journal of Marketing this year that shows an increase in elasticity of advertising because of creative strategy, so that's really good initial evidence.
One of the problems is, if you went to Cannes over the years, you would have heard everything, from creative is 20% of your returns to 80% of your returns. And when you have those ranges, all it means is nobody knows.
Creativity is essentially a strategic asset because creativity involves business models, market changes. It's not just the execution of advertising, there's all sorts of services in the marketing domain that are creative in that sense.
We know creativity is important. If you're looking at the historic performance of advertising, we know we have a declining line over the past 100 years and also an increasing line of advertising spend. We're spending more, we're getting less. We know that the long term is more powerful than concurrent [activations]. So, the branding and brand building is showing up in that consequential aspect. We also know that it's not just doing more that's helping us, it's how we're doing it. Because we have evidence that creative is driving 30% to 60% of additional lift. That's massive.
And on top of that is – back to our initial discussion – the allocation game. If I have the same amount of advertising but I'm picking the right mix [of media channels] for the right outcome that I'm interested in, then I'm going to get higher lifts.
So, the game is not to buy more ads, buy more reach. It's, ‘Alright, I have my budget and I am going to drive awareness in this campaign. So, I have a very precise combination of channels that will drive awareness, and I have a specific creative that will give an extra lift to that.’
So, you get abnormal and disproportionate returns by that combination of activities, rather than just throwing more money into the bucket and saying, ‘Just go buy me more eyeballs’. The evidence suggests that that doesn't work because as [ad]spend is growing, efficiency is decreasing.
I like that we're entering this world where we can actually properly value creative labour because that's one area where we still have some dominion over computers.
Any final thoughts that you’d like to share with marketers reading this?
Just look who you pay attention to when you're looking at these high-level studies. My role in this universe is to talk about theories and things that are more universal. So, I don't rely on my opinion. Me and my colleagues and all these other professors rely on peer review work that has had people like me try to put every single flaw and bullet into it to disprove it. So, it's survived a pretty rigorous attempt by people highly motivated to prove me wrong. And I think there's value in that process, so we don’t just follow speculation and end up with all the issues of lack of confidence that we have in the industry that shouldn't exist.
It's a beautiful game that we play. It's a phenomenal industry. We're more complex than finance. We're more interesting than finance. But we’ve let finance control the strategy because they have clear decision science aspects.
I love [finance]. Part of my training is in finance and most of my research touches on financial returns. But where does the money that finance plays with come from? These cash flows come from customers, and it's our job to make sure that that comes in. And there's a beauty to that. There’s a beauty to the competitive aspect that I will only succeed if I'm smarter than you.
So, is it the competitiveness and the complexity that attracts you to marketing?
Yeah, it is. I have yet to see a science that's more powerful at explaining, understanding and guiding decision making. The power in it is immense, and once it connects with the other parts of business, and we respect that connection, we also immediately get that respect back because they cannot function without us. They simply cannot. If a company does not manage its marketing, marketing will still happen to it, and they will have no agency over that trajectory.
Want more of the same? /
We don’t just write about best-in-class campaigns, interviews and trends. Our Members also receive access to briefings, online training, webinars, live events and much more.
Related Articles /
23 May 2023
‘Behavioural science is not a straitjacket for creatives and planners’ /
Richard Shotton is an expert in the practical application of behavioural science. He talks to Contagious about why friction should come first, why agencies should apply biases laterally rather than literally, and more.