The Association for Qualitative Research
The Hub of Qualitative Thinking

So after the sizzle, where is the sausage?

Wendy Gordon, a Partner at Acacia Avenue, looks at how Behavioural Economics can be applied in qualitative research.

So far, as in the preceding article, there has been a great deal of talk about what Behavioural Economics (BE) IS, the context in which it can be used, and the benefits – but little practical help with 'so what do I do differently?' I'll attempt to help with question marks around application, and offer two examples of BE in action.Here are four areas of application to think about:

  1. When should I use BE thinking and/or key concepts?
  2. How does it alter the way I write a brief or design a qualitative project?
  3. What is the best way of managing internal stakeholders once the project is complete and the learning valuable?
  4. What inhibits BE ideas from diffusion within a company or brand team?

Remember, people are very poor witnesses to their own motivations. Think of an iceberg. What you see above the water line are post-rationalisations for behaviour. What lies below is the mass and shape that influence the iceberg's behaviour.

When to use BE as a model as a model of thinking

BE thinking applies in situations when people are unsure, confused, unclear or overwhelmed by information. The context of confusion could be social, emotional, physical or financial and could be relevant to a current 'now' decision or a future one. This is when cognitive biases come into their own – they are strategies to minimise stress and effort:

Choice decisions – a decision about an outcome in the present:

  • Shall I buy Brand X or Brand Y today?
  • What do I feel like for breakfast out – a McMuffin, a croissant, Prêt porridge or a proper cooked breakfast at the local caff?

Judgments of risk – a decision about an outcome in the future:

  • Shall we go to Cornwall or the South of France this year – what are the chances of the weather being miserable?
  • Is it worth changing my bank account for this offer or shall I stay?

The choice of words and phrases used in client-originated briefs signpost the relevance of BE, e.g. 'reactions to anticipated price changes', 'understanding consideration sets', 'describe purchase intentions', 'likelihood of switching', 'understanding brand choice', 'customer journeys', 'decision-making criteria', 'communicating value or price', 'communicating relevant product features', etc.

These very specific – and very familiar – phrases may be thought of as 'language leaks' and quite clearly indicate whether or not the heart of the issue is about understanding decisions in the future or choices today.

The important point is to differentiate between the BIG why questions asked by marketing and insight teams from the little why questions asked directly of research participants.

  1. Generate hypotheses about the uncertainty contexts and bias factors that might be influencing customers/consumers e.g. heuristics, loss aversion, anchoring, etc. Invite a BE expert (or someone who knows more than you) to facilitate the session.

  2. After the hypotheses have been generated, discuss the available interventions that the organisation/brand could make in response. It is important to clarify the desired behavioural change and possible strategies to achieve this before embarking on the research study.

How does BE alter the way I write a brief or design a qualitative report?

Applying BE to a project requires a total rethink from start to finish. It is not a magic bullet, nor particularly easy to apply.

  1. Recruitment
    Behaviour is a critical dimension of BE studies. Recruit by proven behavioural differences rather than claimed future or past behaviour or indeed by attitudes or opinions.

  2. Sample design
    Be more scientific. Create experimental conditions to contrast behaviour in different contexts. This means letting go of the default use of demographics to differentiate sample cells.

  3. Forcing behaviour change.
    This could be a pre-task prior to the interviewing stage or could be part of the research process itself. Forced behaviour change is a way of vividly revealing the underlying biases at work.

  4. Methodology
    Triangulate and use different approaches within one study. Focus groups alone will not cut the mustard.

  5. Discussion guide
    Write a discussion guide without asking 'why'. Ask instead Who? When? What? How? Where?

  6. Analysis
    Use a BE 'ear' during fieldwork. Listen hard during the interviews/group sessions for evidence of biases and distortions. Language 'leaking' is hugely important and therefore reading transcripts or having a co-researcher to note down language choice is essential (See examples below).

  7. Core team engagement
    Once the research fieldwork and analysis has been completed, ask for an interim meeting with the core client team to figure what behavioural interventions are possible in response to the cognitive, emotional or social biases discovered.

  8. Double check with client research manager
    When preparing the final debrief, take client guidance as to how to talk about BE concepts to the audience otherwise you run the risk of it being seen as academic but not actionable.

    Managing internal stakeholders

    In our experience, it is virtually impossible to apply BE effectively unless the project owner is on board and committed to the journey – however it might end. This means that either the end user is sufficiently intrigued and partially informed about the potential of BE thinking or more usually the Insight Manager responsible for the project itself has recognised its role and is prepared to act as guide for the organisation.

    Here are some top tips we have picked up from two Acacia Avenue clients who encouraged the use of BE thinking on their projects:

    • Don't use BE language such as heuristics, endowment effect or anchoring. It communicates 'clever clogs' or 'academic, not useful' and can act as an inhibitor to acceptance of the recommendations.

    • Remember there are two completely different stakeholders in a company:

      a) The coalface who simply want clear guidance as to what to DO. They are happy for the reasons behind this to be understood by others, but are not interested in BE diagnostics. Don't be like the car mechanic who insists on telling you exactly which widget wasn't working, when all you want to know is: 'Have you fixed it, can I drive it?'

      b) Those whose remit is strategic rather than operational who will spend some effort to understand the value of BE thinking, provided the concepts are explained simply and clearly illustrated by the research.

    Organisatonal distortions and biases at work

    The main difficulty in applying BE is that individuals and organisations are subject to distortions too, just as 'consumers' or 'customers' are. We are all human beings united by the array of cognitive, emotional and social biases that define the human condition.

    Heuristics determine much decision making when it comes to planning, commissioning and conducting research. Think about the default to focus groups or geographical sample splits.

    Loss aversion is in play when it comes to doing things differently. Think about how much effort is required to persuade teams of the advantage of a bricolage approach or using a more sensitive tracking methodology.

    Social norming ensures that decision makers and researchers do not step too far out of line. How often do you hear the excuse that 'the senior directors' will not be comfortable with this approach or with this new thinking?

    Framing results in the team seeing the project in only one way rather than plural ways; a confirmatory bias will result in a decision maker or team favouring and selecting information that confirms their theories and hypotheses and so on. What all human beings can do is to become more aware of the distortions and biases that determine their own decisions and behaviour and seek ways to challenge their own judgments and choices particularly at times of uncertainty whether these be financial, social or organisational.

    Example 1: Switching brands

    Imagine being involved in a very competitive sector. Brands are constantly attempting to win customers from each other by developing attractive offers. The BIG why question is: How likely are people to switch to a competitor in a particular scenario?

    Key Quotes

    • "You never know if the company you've just changed to is going to put up their prices too"
    • "They are all much of a much-ness – might as well stay with the one you have a track record with. That might be an advantage"
    • "They all follow each other – what one does the others will do soon after"
    • "It's too much hassle – I've got better things to do"
    • "They are all in cahoots anyway – what's the point?"

    These language leaks reveal the underlying evaluation dynamics at work.

    Customers are using a similarity heuristic (a comparison rule of thumb) rather than a difference heuristic. This makes differentiation difficult for a start. Clearly, there is anxiety about what will be lost in the future if customers were to switch – a BE concept called loss aversion. Loss aversion is often connected to another bias called the endowment effect. When something is owned it has more value than something not yet owned – in this case the existing relationship however imperfect.

    Qualitative research (without using the BE lens) indicates that the category is a low interest one and other than deal seekers, the majority of people can hardly keep awake during the research interviews! The recommendation is to work harder (at product development) to find a proof point of difference in the offer.

    Using a BE lens leads to a different set of insights. People fear loss far more than they crave gain so they look for reasons to reinforce their current position, rather than looking for reasons to change. If there is not a discernable sense of loss (emotional, financial, social) associated with not changing, people are more inclined to stay where they are. Furthermore allowing people to default to a similarity focus reduces the impact of the offer.

    So the recommendation is very different. This is an emotionally risky decision not a low interest one. The advice is as follows:

    To avoid switching:
    Do relatively little – cognitive, social, emotional biases will work for you to alleviate emotional anxiety.

    To encourage switching:
    Communicate the potential emotional loss in staying with current supplier.

    Example 2: Market Dynamics

    Imagine being involved in a sector where the category has some unusual gender related dynamics.

    1. Why is it that British women don't drink as much beer as their European and US counterparts?
    2. What are the key drivers of choice both within and outside the beer category?

    There are many hypotheses as to the underlying drivers of choice but little understanding of what is really important to women in different venues and occasion contexts. Qualitative research revealed significant product and social barriers that women talk about once the beer category is raised. A traditional recommendation would be to develop a lighter tasting beer for women and to address image issues through communications.

    Key Quotes

    • "I don't like the taste of beer"
    • "It makes me feel bloated and fat"
    • "Too gassy"
    • "None of my girl friends drink it"
    • "I didn't grow up with it in my home"
    • "I prefer a sweeter drink – like a cocktail"
    • "I think it's a masculine drink – a pint glass isn't sophisticated!"

    Looking through a BE lens identifies two highly relevant distortions at the heart of the problem.

    Firstly heuristics
    These are experience-based strategies that operate well below consciousness and help people make quick and effective decisions. People tend to resort to heuristics when they find themselves in situations that are uncertain. Many women find beer a difficult category to negotiate in social situations – which type, which brand, how to drink it, etc?

    Secondly, framing effects
    Women who don't drink beer associate the category with loss of control – lying drunk in the street, macho and anti-social behaviour and loss of femininity. Women who do drink beer associate beer with country pubs, summertime, barbecues and holidays.

    Human beings tend to believe that they are independent agents of their own choices. But more often than not choice is determined by factors outside of our control. BE highlights the extent to which the choice not to drink beer is influenced by social, emotional and situational factors – a complex web of cultural influences.

    Research focused on behavioral levers rather than product innovation or above-the-line communications. Experimenting with different glass shapes, bottle shapes, labeling, bar demos and in-store merchandising, it became clear that these factors could change women's perceptions of beer and their willingness to buy/drink it, far more rapidly than developing a new brand, product or campaign.

    References

    Gordon, W, Two part introduction to BE, YouTube
    Gordon, W, (2011) Behavioural economics and qualitative research – a marriage made in heaven? IJMR, volume 53 Issue 2 pages 171 – 186
    Lehrer, J (2009) How we decide, Houghton Mifflin Harcourt
    Thaler, R and Sunstein, C (2008) Nudge: improving decisions about health, wealth and happiness, Yale University Press
    Ariely, D (2008) Predictably irrational: the hidden forces that shape our decisions, HarperCollins Publishers
    Kay, J (2010) Obliquity: why our goals are best achieved indirectly, Profile Books
    Khaneman, D (2003) A psychological perspective on economics, American Economic Review, Vol 93 No2, pp 162 – 168
    IPA (October 2009) Behavioural Economics: Red Hot or Red Herring?

     

    Wendy Gordon
    Copyright © Association for Qualitative Research, 2011