Research Now SSI: Scale Matters, Accuracy Matters, Confidence Matters
Track Opinion
Download the insights industry's #1 strategic planning tool

Busting The 3 Myths About Behavioral Economics That Are Holding MR Back

There are many factors that may be preventing some from using behavioral economics and others from realizing its full potential. But busting these 3 myths in 2016 would be a good start.



By Ian Murray

My favorite business book of 2015 is ‘Misbehaving: The Making of Behavioral Economics’ by Richard Thaler. It’s a memoir of 40 years of ground breaking academic endeavor in which Thaler concludes that behavioral economics has gone mainstream:

‘Behavioral economics is no longer a fringe operation and writing an economic paper in which people behave like Humans is no longer considered misbehaving, at least by most economists under the age of 50.’ (p. 347)

Back in the summer, Thaler was the star attraction at Nudgestock 3, positioned by Rory Sutherland of #ogilvychange as a ‘coming of age’ for the application of behavioral science in marketing and advertising.

Contrast this with the latest GRIT report which shows just 17% of research clients and 23% of research suppliers are using behavioral economics frameworks as part of their toolkit.

So why is market research lagging behind? I think there are a number of persistent myths that are holding us back:

Myth 1: Behavioral Economics tells us that ‘all market research is wrong’

“The trouble with market research is that people don’t think how they feel, they don’t say what they think and they don’t do what they say.” (David Ogilvy)

The advertising industry’s relationship with MR has always been mixed. In the era of Mad Men they were pioneers of consumer research and psychology. But more recent generations are more likely to blame MR for ‘killing creativity’ and being just plain wrong. So the idea that market research has been completely undermined by behavioral economics is particularly seductive to MR’s critics.  It also reinforces some of the deepest insecurities of market research professionals.  What if skeptics like Faris Yakob are right? What if ‘all market research is wrong’?!!!! Rather than rise to the challenge it seems that many in MR simply prefer to avoid the debate and stick with business as usual.

Faris is right- to a point.  Market research, like most other business disciplines including advertising and marketing, has clung to the standard rational model of economics for far too long. We may not be changing fast enough but the MR skeptics overstate their case when they argue that MR has no value.

There are many innovative approaches (implicit research, prediction markets, etc.) that are enhancing the ability of MR to capture real world decisions and behavior.  But BE shouldn’t be seen as a niche offering. In fact BE is a collection of frameworks that help researchers build real world hypotheses and identify appropriate methods for any brief.

BE is not a ‘technology’ and you don’t need to be at the bleeding edge to apply it. In fact the MR skeptics seriously underplay the role of ‘traditional methods’ (i.e. asking people explicit questions and recording their answers) in many of the seminal experiments of behavioral economics.

How did Kahneman and Tversky develop their Nobel Prize winning prospect theory? They framed the same question in different ways and modeled people’s answers. How did Thaler and Kahneman design their classic work on fairness and the endowment effect? Among other things, they constructed a telephone survey, presented lots of different scenarios to people and asked them which ones were ‘fair’.

So all MR methods have their place in the BE toolbox. But to get the methodological balance right market researchers need to have a comprehensive knowledge of behavioral economics principles and how they can be applied.  The persistence/resilience of 2 further myths indicate that there is some work to do.

Myth 2- Behavioral Economics tells us that people are ‘irrational’

‘To exchange one orthodoxy for another is not necessarily an advance’ (George Orwell)

In market research the discourse around behavioral economics typically sets up as revolutionary struggle between two opposing paradigms (System 1 v System 2). System 1 is the new orthodoxy that threatens to sweep System 2 into the dustbin of history.

But System 1 has also become synonymous with ‘irrational’ behavior.  And, of course, ‘Irrationality’ is bound up with a multitude of negative connotations. The collective psychology of market researchers is important here. BE has a real ‘do they mean us?’ problem. Researchers are real people too and it is simply very difficult for people to reject a hard wired (normative) rational paradigm in favor of an ‘irrational’ one that seems to focus only on flawed or stupid behavior.

The pejorative tone of much of the coverage of BE in market research inevitably creates barriers to its adoption. We need to make BE more palatable for people and easier to live with.

Thankfully, all that is required is a slightly more careful reading of the BE classics. This quickly reveals that the binary choice between System 1 and System 2 paradigms is spurious- we really don’t have to look too hard for this- the clue is in the title of Kahneman’s book- it’s about Thinking Fast and Slow…..

Kahneman is also concerned that System 1 has gained a ‘bad reputation as the source of errors and biases’. Again, a closer reading of the conclusion to ‘Thinking Fast and Slow’ reveals a more balanced and positive view of System 1 decision making:

‘System 1 is indeed the origin of much that we do wrong, but it is also the origin of most of what we do right-which is most of what we do’ (p.416).

It is important to keep things simple, but not simplistic. Advocates of BE in market research would do well to add a little more nuance to the story.

Myth 3- Behavioral Economics is a ‘qual thing’

Ray Poynter addressed this one back in 2013 noting that ‘there seem to be a remarkable number of market researchers who seem to assert that BE is mostly a qual thing, or that the main implications of BE for market research will be a focusing on qual’. Ray forcibly and eloquently argues that BE is ‘mostly a quant thing’

Parochial infighting between quant and qual must be one of the most debilitating things in market research. Of course we should be designing integrated approaches with no a priori preference for quant or qual……But until that day comes there is no escaping the fact that quantitative models are the currency of BE.

The standard economic model has proved remarkably resilient in all areas of business and policy making because it generates precise, falsifiable predictions from mathematical functions based on quantifiable parameters of choice such as expected utility and probability of outcomes.

The pioneers of behavioral economics understood that merely observing or describing anomalies in the standard model would not be sufficient. To challenge its normative status behavioral economics had to offer new quantitative models that made superior predictions about people’s real world behavior. Thus Kahneman & Tversky’s prospect theory provides an alternative mathematical function that makes predictions that have become the cornerstone of much of the latest thinking in business and marketing (e.g. reference dependence, framing, overweighting of small probabilities etc.). The ability to make accurate and quantifiable predictions is what gives BE this power.

So the template for promoting the value of behavioral economics in the boardroom has already been established by BE pioneers like Kahneman, Tversky and Thaler.  And it is quantitative not qualitative.

Thankfully advocates of behavioral economics in market research do not need to concern themselves with developing new value and utility functions! But they do need to take their lead from the BE pioneers by focusing on providing the kind of robust quantitative evidence that will convince the boardroom of anomalies in their existing models and drive adoption of the new framework.

Of course, I’ve barely scratched the surface. There are many factors that may be preventing some from using behavioral economics and others from realizing its full potential.  But busting these 3 myths in 2016 would be a good start.

Please share...

35 responses to “Busting The 3 Myths About Behavioral Economics That Are Holding MR Back

  1. A lot of good points, Ian. I’m especially interested in your second point- irrationality. BE doesn’t say people are irrational, they say they are economically irrational. That is, people don’t always make a decision that will benefit them economically. This shouldn’t shock us as marketers or market researchers. Our whole reason for being is that we can create brands and influence shopping behavior.

    Where BE has fallen down, I think, is that it has become like cognitive dissonance in the 1960s – it can explain anything through the operation of one bias or another. In so doing, it becomes untestable – if a prediction doesn’t come true, you just invoke another bias to explain it away.

    1. Hi Steve. Totally in line with. The article is very interesting and useful but saying that System 1 is synonymous with ‘irrational’ behavior is a shortcut : system 2 is also irrational. A lot of involving financial decisions which activate system 2 are somewhat irrational: just think about the way you decided to buy the house you live in rather than a more economically rational one. And another point: irrational does not mean stupid. It can lead to smart decisions without spending much efforts, which is not stupid.

  2. Enjoyed reading and made me think :

    Isn’t behavioural Economics continuing to reduce and fragment the understanding of human behaviour…what is the list of biases up to these days?

    It’s almost like humans are becoming a bag of biases.

    At some point I’d like to see a new synthesis, maybe something related to a constructivist viewpoint that recognises that people are complex dynamic systems and their behaviour emerges from how these mental systems are constructed individually and socially.

    BE seems to me to be very narrow and short term in its approach (helps with experiments). It takes no account of longer term learning processes or experience. Mind you this time-horizon fits in well with the digital marketing short term measurement. I suspect BE has difficulty predicting how behaviour changes over the longer term. Are there any BE examples of this?

    I get feeling BE in MR is mostly useful for the short term, tactical aspects of marketing…less so for longer term marketing strategic decision making.

  3. I personally see BE as enhancing our understanding of consumer behavior rather than a revolutionary breakthrough. Much nonsense has been written about it (not in this column, though) and this has held back its acceptance by marketing researchers. Who has really taken the backboard economics we learned as college freshman as The Truth?

    Moreover, Kahneman and Tversky’s work has been in the public domain for decades now and it not universally accepted by academics – Gerd Gigerenzer , for example, has some other viewpoints I feel are worth a look. K&T’s work has also been criticized on methodological grounds, e.g., lack of replication and, more fundamentally, that their experiments are really tests of IQ and math aptitude – thus MIT students often perform better than students at other less prestigious universities. Another criticism is that subjects in some cases actually know the correct answer but think they are being asked a trick question, and give a different answer.

  4. Great article Ian, thank you. I believe we are on the way to mainstream BE adoption in any case. However – and subject to industry, corporate culture and other factors – we should not underestimate the time it sometimes takes us to onboard our internal stakeholders on new methodologies, latest sciences, consumer trends or tools, especially if they are used to direct questioning and focus groups and/or if bonus relevant KPI’s are depending on conventional approaches. We’ll get there eventually though..

  5. Thanks Steve, some strands of BE certainly DO say that people are generally irrational rather than specifically economically irrational. Much of Tversky and Kahneman’s work on the conjunction fallacy, which they use to challenge classical probability theory as a model of ‘rational’ decision making, is based on experiments like the ‘Linda’ problem which have no economic references at all (the experiments show that people rank the probability of an individual exhibiting certain characteristics in a way that contradicts set theory i.e. even subjects who have a sophisticated understanding of statistics consistently asses the conjunction that Linda is a bank teller AND a feminist to have a higher probability than Linda is a bank teller). Similarly people think that the conjunction ‘Swedish people have blond hair AND blue eyes’ has a higher probability than ‘Swedish people have blue eyes’ and experts such as doctors, when assessing patients, think that the conjunction ‘has had one or more heart attacks AND is older than 55’ has a higher probability than ‘has had one or more heart attacks. So ‘irrational behaviour’ (i.e. compared to classical probability theory) has been shown across a range of decision making domains.

    Of course, classical probability theory is only one model of rationality (by the way classic experiments like the Wason selection task show that logicism doesn’t stand up as an adequate model of decision making either). That gets to Kevin’s point about Gigerenzer who critiques Kahneman’s heuristics and biases project by advocating a model of ecological rationality and highly adaptive fast and frugal heuristics. I have a lot of time for Gigerenzer. But that is another story….

    I have some sympathy with your other point which is essentially that BE lacks normative status. The big list of behavioural biases can be used to justify a lot of stuff POST HOC. But I think aiming to provide an accurate description of the world (no matter how messy) is a better plan than hanging on to an outmoded model just because it offers some straightforward (and often inaccurate) axioms for decision making.

  6. Hi Martin, I don’t think BE is deliberately trying to fragment out understanding. It is just trying to tell it like it is. There are plenty of people working in the field that are trying to synthesise.

    I also disagree that BE has a short term focus. There is a whole field devoted to inter temporal decision making. Thaler, who I reference in my blog, has a very accessible section in his book on the pitfalls of ‘present bias’. A lot of BE is devoted to developing strategies to combat this bias and help us to think and behave in ways that are in our long term interests.

  7. Hi Kevin, completely agree on Gigerenzer. We launched house51 with a blog about Gigerenzers ‘Risk Savvy’. I find his critique of Kahneman et al quite convincing. Although the model of fast and frugal heuristics and ecological rationality has challenges of it’s own. I focused on Kahneman as it is the popular entry point into BE for most people. But I agree that MR needs to broaden it’s range of BE reference points.

  8. Hi Christian, I agree we are on the way to mainstream adoption. And I take your point about the sensitivity and understanding that is required in on boarding stakeholders.I founded house51 to help us get there! We have been working hard to develop ways to integrate BE with conventional approaches to make the whole thing more actionable and accessible.

  9. HI Ian,

    I like your article. I think the issue is that you can’t determine how an individual is going to respond/act, but you can tell how a group on average is going to respond. That’s why market research doesn’t just ask one person, we ask a statistically valid sample quantity. But we also build in Kahneman System 1 and 2 into our consumer models (using agent based modeling – It works very effectively.

  10. Not a bad overivew of why folks should explore how to to integrate BE into market research. It would have been useful to have a brief discussion of tools and approaches for doing this.

  11. Hi Ian, Fragmentation may not be deliberate, however, I have not found any synthesis to show how the list or biases relate to each other.

    The longer term focus is not evident to me (where should I look) -maybe the dynamic modelling Guy does can show how an individual bias has a longer term effect. I was meaning that its application is in short term marketing application areas.

    I’d like to see it applied to marketing and business decision makers (and market researchers) too- to reveal the bias they have to be wary of 🙂

    1. Hi Martin,

      I think I can help you with that 🙂 Would you mind if I emailed you with some examples of frameworks we’ve used and also developed at the Irrational Agency? We’ve written several articles on this as well as an ebooklet which I’d be happy to send you.



  12. Hi Martin, everyone,

    We use ABM and model both short and long term effects. Not sure of your question about bias.

    Will also reach out offline.


  13. Sorry Guy, I mean by bias how in our own decisions about our brands and intepreting data we have the same types of potential BE bias patterns as customers.

    Which ones are we most susceptible too. BE is not just for “customers”.

  14. BE will need to apply BE principles more effectively if it ever is to become more widely-accepted by MR practitioners and users. Many seem increasingly overwhelmed by all the choices they now have (though gratified too), and I find myself having to spend more and more time correcting basic misunderstandings of the basics. In some ways marketing researchers are like physicians who have skipped the first year of medical school. BE should have an intrinsic advantage over many competing innovations in breaking through the clutter, I’d think.

    1. Hi Kevin

      One of the challenges of applying BE in market research is that it takes a much deeper understanding of and immersion in behavioural science than can be easily gleaned from the boys that are readily available and what most market researchers are willing to commit to. A list of biases are a nice fair dust to sprinkle on a project and certainly a minor improvement, but a more fundamental reform and innovation of methods requires looking deeper into the wider *processes* behind human decision making. Unfortunately that’s often complicated – even fully understanding dual process models is quite a lot of work! (Contrary to what many people think, the debate on S1/S2 is far from over and you need to understand the nuances of that debate to be able to build accurate quantity tool in MR – for further info on the debate look up In Two Minds).


  15. Interesting post and dialogue. Can anyone direct me to a current list of BE heuristics that are commonly accepted? And, any companies or methods you see that are good demonstrations of BE applications?

    I don’t personally see the connection between predictive markets and BE – or System 1 for that matter – as they do not measure an individual’s non-conscious attitudes or reactions (like IAT, facial recognition, or EEG) and really boil down to a System 2 account by respondents view of what they think other people will think about an idea (often including many people without knowledge of the category).

  16. Hi Folks,

    Sorry, been busy so just picking the conversation up again. Thanks for all your responses. I hope you don’t mind if I respond to the key points in one message..So, here goes in no particular order…

    @Martin- I do agree that fragmentation is a problem. But the ‘bag of biases’ narrative is a good example of the type of misinterpretation of BE that needs to be challenged. A lot more work needs to be done in MR to synthesise the diverse data that BE produces. And there is a massive commercial opportunity here. Indeed this is exactly why I founded house51. We make BE accessible and actionable by building frameworks that codify clients strategy and marketing activity according to BE principles. But rather than ask them to get their heads around a ‘bag of biases’ our frameworks connect to the relevant literature via no more than 4 or 5 categories or dimensions of response.

    @Martin (again). agree that we should use BE to examine the practice of market research. I think you will recall that I have written about this elsewhere (i.e. ‘Turning the Behavioural Lens on Ourselves BE should not be seen as something we simply ‘DO’ to other people. The principles would help us to make better decisions too and confront persistent biases that blight MR and wider business decision making- confirmation bias, present bias etc.

    @Jeff- Google ‘big list of behavioural biases’ to see a reasonably comprehensive list. However, as I have suggested above, I think this is the wrong approach. There is much more value in the synthesising approach that Martin suggests. And this is the route we have adopted at house51. We have only been going for 1 year but so far our clients seem to appreciate it!

    @Jeff (again) As I emphasises in my original blog, one of the myths that is holding back the application of BE in MR is that it is all about System 1 -this is the point I am making with the Orwell quote about the risk of replacing one orthodoxy with another. BE has a lot to say about System 1 and System 2 processes and alternative models of ‘bounded rationality’ such as Gigerenzer are based on the argument that humans are socially and ecological rational (i.e. fast and frugal heuristic processing based on adaptive social intelligence is one of our most effective and reliable forms of decision making). Predictive markets are certainly compatible with these BE frameworks

  17. BE is not new – Amos Tversky assed away nearly 20 years ago – why has this line of thinking caught so many marketing researchers by surprise?

    The Austrian School of Economics (e.g., F.A. Hayek) has been a counterweight to standard economic thinking for many years. They essentially view value as subjective (I’m simplifying a bit) and, in that line of thinking, BE and classical Econ 101 is suspect. For example, someone may buy a high-priced item even if they believe to be of lower quality than cheaper alternatives. This is not necessarily irrational, however, depending on the context (e.g., maintaining peer-group status).

  18. Couple of questions:

    Does BE link to any overall model of human and consumer behaviour? Like the Sheth Model? If not would it help to create one, to place bias in a wider context.

    Has any work been done on how BE heuristics and biases developed (what causes and shapes them). Are any new ones likely to develop through behaviour?

    BE originated as an alternative to economic modelling. Why do we think it is suitable for understanding consumer behaviour

    1. Hi Martin

      Yes, there is a vast literature on how the heuristics have developed – unfortunately most people who talk about BE in public (not Ian though!) have a limited knowledge of the science as they mainly base it on the pop science books without fully immersing themselves in the literature or academic discipline. Again, I’d be happy to send you examples of this – a bit rushed this morning to be able to write a full response here as it would take a while to catalogue the research!

  19. Hi Elina

    When I did my marketing education in the 1980s, we looked in detail at different consumer decision models and assessed their strengths and weaknesses – it was a multi perspective approach. These acted as useful thinking frameworks – we recognised that they were models and so were never completely “right”.

    These recognised both “rational” and “irrational” decision making perspectives and processes – the social, emotional and functional, and how these linked back to other disciplines fields.

    This seems to have disappeared with BE…it just comes over as a ragbag of ideas.(head ducks back down behind parapet)

    I have not found a way to link together the BE findings into something deeper or fundamental…a foundation.

    System 1 and System 2 seem to be simply ways of describing, in a vague way, that we process information in our environment consciously and sub consciously.

    The “system” in both – their elements and their relationships are not made explicit.

    Even the empirical studies I have read (in popular books) used to describe/ justify/explain BE findings feel almost anecdotal sets of statistics.

    I wonder if at the bottom of this, business decision makers are pragmatists and do not really care about the truth or explanations anyway, so it does not really matter as it does to the more purists.

    Business decision makers just care about whether “it works”…and working means, does it help me improve my performance.

    Look forward to the examples – I am open to being convinced


  20. Hi Martin

    Part of the reason for how BE has been portrayed is because it first became a phenomenon in the public policy sector where it really has been (and continues to be) a radical innovation.

    By the way, there’s a lot of confusion about what bounded rationality actually means: it *doesn’t* mean people are irrational, but instead that we make the best decisions possible within the biological limitations of our minds. Additionally, a more comprehensive, rounded view of BE takes evolutionary psychology into account but that doesn’t come up a lot in the research by the so-called Heuristics & Biases school of thought that most of the pop science books (and consequently most of the conversation in MR) are based on. It’s important to remember that it’s not the *only* school of thought, and as such not The Truth. Some good insights into this topic can be found in the book “The Heuristics Debate”.

    And you’re right, one of the reasons there isn’t much talk about the fundamentals is that business decision makers aren’t really interested in bigger theories which makes it harder for them to fully appreciate what decision making science (which is a better description for the discipline than BE in my view) could offer them. We see this when we talk about that stuff to clients all the time. Taking the bag of biases approach plays well in conference talks to pick counterintuitive findings/studies as it makes for a more interesting story. Unfortunately in this case storytelling and effective applications are often mutually exclusive.

    In an attempt to try to persuade you, here are some things we’ve written that may be a good starting point – we’ve been talking about this need for a more cohesive framework for a couple of years now and feel like we’ve made some progress:

    And also our article in the 2014 BE guide – unfortunately you’ll need to download it first:

    We’ve also written a couple of ESOMAR papers that attempt to build that bridge to a more comprehensive theory that shows more of the theory than these blog posts – email me and I’ll send them to you 🙂


  21. Hi Martin, by ‘foundation’ do you mean a normative model with a series of simple axioms that presents an ideal about how people should behave? If so, I am afraid you are looking in the wrongs place with BE. It focuses on descriptive models of how the world really is (e.g. Prospect theory is a descriptive model that originates from observing the numerous anomalies that arise by contrasting observed behaviour with the predictions the standard model of economics and develops new principles to explain these anomalies). This is inevitably more complicated and messy story. But it is also more accurate. You are right that business decision makers want something simple. But how can you claim that these simplified models ‘work’ if they can’t capture real life? Most proponents of BE are not purists, they are pragmatic realists (By the way, in case you haven’t guessed I count myself as a realist, and if Elina doesn’t mind, I count her as one too!).

  22. Hi Ian

    My interpretation of Martin’s question was that he’s looking for something that would tie the bag of biases together to understand the underlying dynamics a bit better instead of just scattered findings in various experiments. Martin – could you clarify what you mean please?

    I do agree that reality is more messy than many business stakeholders would like, and it’s a tough job to communicate that sometimes 🙂

    And yes, I’d be happy to be classed as a pragmatist!

  23. Hi Ian

    By foundation I do mean some kind of axioms, however these do not need to be based on how people should behave. Simply a set of explicit assumptions/beliefs about humans from which a theory is built.

    So, what does prospect theory accept is true?

    This is my list from

    – People framing of an outcome determines the utility of the outcomes
    – People’s context of a decision affects the utility of outcomes
    – People decide between alternatives based on their assessment of the potential value of loss and gain between the options
    – people will use learned concious or unconscious habitual short-cuts to manage the amount of effort used in a decision
    – people will miss-judge the real probability of outcomes
    – people use reference points from which to make judgements between options
    – people will miss-judge the effect of a decision and its utility the further away the outcome will be realised
    – people will omit social and emotional considerations from their judgements
    – people will not learn from there experiences and modify their choices (no feedback loops)

    How did I do?

  24. Hi Elina

    Philosophically, I think I am a mixture of social/personal constructivist and system dynamist (if that exists). People construct an internal model of the world and this is what frames their decisions, and they modify this model in a learning process, fed by the results of outcomes from their decisions. This means placing emphasis on how these models are formed and how, how one can recognise them in use, and how to influence them.

    This might be why I am struggling with the BE perspective.

    From my study of BE/Prospect theory (which is not as deep as yourself and Ian) BE seems to be more about solving an Economics problem ie an alternative to normative economic modelling, not an understanding of human decision making in the area of marketing (ie something broader than the final choice of how people chose and act)

    Mind you I have no prospect of getting a Nobel prize, and I simply want to understand what I don’t know about BE, learn from those who do 🙂


  25. Applying Behavioural economics in market research requires much more deeper understanding of how the respondent would behave in a particular situation or to a particular question, which might be irrational or rational. Every Individual cannot be expected to be rational in all situations. There are two main reasons why the market research would continue to have a bias due to BE . a) Respondents will tend to have a bounded and simple rationality which cannot be eliminated completely.b) The survey design methods often involve a complex process which optimizes time, simplicity and information. Minimizing the irrational impact due to BE will require a tradeoff with time and information.


  26. Great comments about the MR on digital. I use digital to spread the gospel, and that is how twenty-first-century ministers will have to become more digitized to be effective. Since almost everyone has a tablet and a laptop, as well as smartphone, this is the most connected
    generation. This is what sells products as well as promotes the gospel.
    We as ministers can learn a lot about being fishers of men by reading this Blog. I am grateful to have the time to do so and I hope others will as well.

Join the conversation