How exactly should we understand customer behaviour without behavioural science?

In the past 8 years, I've read a lot of articles on the topic of behavioural science and its applications - some are excellent, some are wishy-washy and forgettable, but only rare exceptions are as nonsensical as the recent Marketing Week article questioning the usefulness of science or scientific method when it comes to human behaviour. This confusing and spectacularly misinformed piece made me spill my morning coffee all over my tablet - please join me while I dissect the nonsense.

A reportage of a talk given by a Marketing Week columnist and branding expert at the Festival of Marketing at the beginning of October, the article's main argument is that "marketers are 'underestimating' the significance of the signs that offer an insight into customer behaviour because they are failing to observe them" which evolves into a strange strawman argument dismissing behavioural science - in other words, my chosen field and profession, so please excuse my sharp tone.

On behavioural science and branding

She urged marketers to rely less on science and instead pay greater attention to signs that show how customer behaviour is evolving. If you're thinking this a call for more semiotics, guess again:

“I’m not talking about brand semiotics and what the commercial world uses about reading the big signs and symbols we push out as marketers. What I’m talking about here are the small signs and clues coming back to us from people if we care to observe them”

At this point, I was utterly confused because I have always used theories from behavioural science to analyse consumer behaviour. Instead of relying on intuition, experience and common sense (all of which vary by person and are subject to biases), behavioural science is a way to add objective structure and rigour to the process of understanding consumer behaviour.

The Marketing Week columnist questioned whether marketers are right to use the word ‘science’ when discussing human behaviour, suggesting that really behavioural science is something of an oxymoron. “It just seems to me that human behaviour is too fluid, too individual to be linked with the word ‘science’,” Edwards argued.

I'm really not sure what to say to a wholesale dismissal of an entire discipline including thousands of research papers, books, and studies over the past 100 years - especially as the field of branding has been leaning on psychology literature from the mid-90s onwards.

The biggest shift came in 1993 with the consumer-based approach: until then, brand management had been just a part of marketing, and introduction of customer-based brand equity by Keller changed everything.

Keller's seminal work, Strategic Brand Management, proposed that the brand resides in the mind of the individual consumer as a cognitive construal and brand value creation takes place by moulding the brand associations held in the consumers’ minds. The underlying logic is that if the marketer feeds the ‘consumer computer’ with the most appropriate information, then the consumer will do as intended and choose the brand. Understanding of the brand in the mind of the consumer was conceptualized as ‘brand knowledge’, which is divided into ‘brand awareness’ (brand recall and brand recognition) and ‘brand image’ (the set of associations linked with the brand). All of these are based on concepts from cognitive psychology.

Since then, the discipline of branding has continued to borrow heavily from psychology, aka behavioural science. But if behavioural science is an oxymoron and human behaviour is too individual and fluid to be linked with the word science, what does it mean to be an expert in the discipline of branding if it's originally based on... behavioural science?

(For what it's worth, I believe branding requires a multidisciplinary approach in practice - but despite two degrees in marketing, I wouldn't call myself a branding expert.)

Next up, replicability critique:

She pointed to the fact that much of behavioural science is “heavily dependent on single famous experiments”, which she noted had failed to be replicated and suggested that even fans of the discipline would acknowledge it hasn’t had a good crisis.

I can forgive anyone for thinking that much of the field of behavioural science is dependent on famous experiments - it is indeed true that many of the commonly used examples are famous, but if you have read beyond the top 3 pop science books on this topic and perhaps even have a degree in it, you'll know that there is much more to it with a continuously growing knowledge base.

And yes - some areas of the psychological literature have undeniable issues with replicability. Many of the attention-grabbing experiments used as examples belong in that problematic group because the more counterintuitive something is, the less likely it is to be a robust finding. Yet, the general audience reading pop science books would be bored to death with the more technical side of the literature - light, fun and surprising stuff is more memorable.

As for describing applied behavioural science practitioners as "fans of the discipline", I will refer you to the above comic - most of us do much more than simply thinking "space is pretty". Understanding the ecological validity of the research is a key part of the applied behavioural science practitioner's skillset, which leads me to the next point.

Don't blame "behavioural science" for UK's poor response to COVID-19

According to the article, "behavioural science" famously got it wrong about behavioural fatigue - well, actually it was largely made up by the UK government and widely criticised by hundreds of behavioural scientists for having little evidence to support it*. As I suspected at the time, this early attempt at of using behavioural science as a red herring and potential scapegoat has lingered like a dark cloud over the applied behavioural science profession, casting aspersions on its usefulness and credibility. Although we can't say for certain, many have suggested that the term came from Chief Medical Officer for England, Chris Whitty, who basically made a guess based on his experience of patients failing to adhere to their medications and assumed that they ‘tire’ of doing so.

When it comes to COVID-19, the UK government has been consistent in its stubborn ignorance of advice from established behavioural scientists such as professors Susan Michie and Stephen Reicher, experts in health behaviour change and social psychology and members of the Independent SAGE committee, who have been openly critical of the UK government's approach.

In all honesty, though, there have been cases of overreach and I agree that a bit more humility would have been beneficial. Compared to something like physics, psychology is a young science and human beings are complex - it is challenging to apply theory to practice at the best of times, let alone in a situation most of us have never been in. An effective response to a complex problem such as behaviour change requires a structured approach - something like the COM-B model and rigorous analysis of the evidence base.

What I do agree with is that "nudge theory" alone is not sufficient - especially as it's not even a real thing.

Behavioural science ≠ nudge theory

"Nudge theory" is commonly used but vague and undefined term. It is often used to convey the idea that small changes ("nudges") in a choice environment can result in behaviour changes. The underlying premise is that as humans we do not always make decisions in our long-term best interest, and that subtle "nudging" can help us get on track, and that it is a useful addition to the policymaker's toolkit alongside laws, regulations, prohibitions etc.

It is often used synonymously (and erroneously) with behavioural economics, which itself is the application of psychology to economics, in order to explain why people behave as they do rather than what the standard ‘rational’ model of economic agents would predict. (As an aside, there is also more to behavioural economics than biases and fallacies - but I digress.) Broadly speaking, nudges are a solution to remedying the issues highlighted by the behavioural economics literature and originally referred to these solutions as an alternative to other kinds of policy interventions.

In fact, you can spot an inexperienced practitioner whenever you see someone singing the praises of nudging. Experienced practitioners and academics know that behavioural interventions are not always the best fix for behavioural problems and, even when they are, silver bullet solutions are much like unicorns - a wonderful fantasy. Instead, effective behaviour change requires multidisciplinary evidence, systems thinking and logic models like COM-B to design a range of interventions.

Anyway, back to the MW article!

Science, schmience

As evidence for her argument, she tells a story of how her team created a range of t-shirt mock-ups with different slogans and used them to understand how consumers were feeling in an "ethnographic study":

Armed with this information, the team created mock-ups of eight T-shirts, featuring slogans that had been popular on the street, which were rendered design agnostic. Then 1,000 people aged 18 to 40 were asked, if they were paid a small sum to wear one, which would they choose? The slogans represented different dimensions from community to individuality, freedom to restraint.

Loosely speaking, this is actually what behavioural scientists would call "an experiment": testing which one of a range of options people choose falls firmly in the domain of decision making psychology (aka behavioural science), and I suspect there were at least informal hypotheses generated before choosing the 8 candidates to be tested.

Ethnography, on the other hand, is an approach that explores cultural phenomena from the perspective of those being studied, and typically uses methods like participant observation to create a narrative account of that particular culture. Usually I link my sources but this information is so widely known that a 10 second Google search will be enough to fact check.

“It was surprising to me that the pull was towards individuality and freedom, so you might want to think about that if you’re a brand about to jump on the comms bandwagon of community and team. You might want to think about how individuality plays within that,” Edwards suggested.

I would say this result is not surprising in the slightest - not if you are familiar with the science of cultural psychology. UK is one of the countries where individualistic values are strong and independent social orientations are prevalent, and it would be very easy to predict that t-shirts with slogans aligning with values that are dominant in a given society will be popular.

In the words of the great philosopher Homer Simpson: "doh".

You cannot understand customer behaviour without behavioural science

Quite simply, it is beyond parody to suggest that we should ditch behavioural science when it comes to understanding customer behaviour - the clue is in the name! However, I do agree that we need to move away from the idea of simplistic descriptions of what behavioural science is and what it can do.

Unfortunately, applied behavioural science is a bit like marketing: everyone has an opinion, regardless of whether they are actually knowledgeable about it (see meme below). An even bigger problem is the voices of applied behavioural science are not equally distributed: the enthusiasm and confidence of the "hazard phase" often generates a lot of the content online, whereas experienced practitioners hesitate to broadcast strong opinions because they know the answer is usually "it depends" and too complex for a simplistic, attention-grabbing headline - let alone an article anyone but their peers will have the motivation to read.

In some ways, behavioural science is a victim of its own success: in the attempt to convince people of its usefulness, the early days were spent shouting about how irrational we all are and how that can be easily fixed with some nudges. But even a toothpaste isn't just the sum of its advertising campaign or branding: a lot of science goes into creating the actual product before it lands on the desk of a branding expert.

The same applies to behavioural science.

(I'm going to leave you with this excellent meme - credit to @Samuel Salzer of Habit Weekly)