What is behavioural science (in a nutshell)?
Behavioural science isn’t a particularly new discipline, but within the past few years it’s become increasingly more influential within market research, guiding how we ask our questions, and how we analyse the answers. At its core, behavioural science seeks to understand the many aspects – both conscious and sub-conscious – related to someone’s habits or decision-making.
A key component of behavioural science is understanding cognitive biases and heuristics (mental shortcuts), in order to decode our (often irrational) behaviour and describe why we tend to act in the way we do. An understanding of cognitive biases and heuristics is a useful tool for market research because it allows us to gain deeper, more reliable insights from the answers to our questions.
There are hundreds (maybe thousands?!) of cognitive biases and heuristics related to our behaviours. But don’t worry, we’ve reduced them down to five that we think are particularly important when designing and analysing research (with a few bonus biases thrown in for good measure!).
Five key cognitive biases to consider in consumer research
Halo Effect
Halo effect: the positive bias a consumer may feel towards a product or proposition because of their overall perceptions of the brand, or previous good experience. The halo effect increases brand loyalty, strengthens the brand image and reputation, and translates into high brand equity. But it can sometimes skew results in research – if you want a more reliable read on the appeal of competitive products or features, unbranded or ‘blind’ testing can help. Fun fact: the opposite of the halo effect is the horn effect (think devil!); when consumers have an unfavourable impression or experience, they correlate that negativity with everything associated with the brand – something every brand wants to avoid!
Availability Heuristic
Availability heuristic: a reliance on the things that we immediately think of to enable quick decisions and judgments. Day to day this is a handy mental shortcut for making the many decisions we face a little more manageable. This is why we consider unprompted awareness to be a key metric when researching brands and assessing advertising effectiveness – if something is top of mind, it’s much more likely that people will choose it than if they have to rack their brain to remember it. Humans like things to be easy! If things are easy and familiar, they tend to feel right, acting as further validation for our decision (which is yet another bias, the cognitive ease bias).
Novelty Effect
Novelty effect: used to describe a positive effect that is entirely due to fact that there is a change, a new design feature, module, or process being introduced, regardless of what the change is. This one is very important when researching new products and services and is one of the reasons we recommend downweighting interest or claimed take-up to deliver more realistic results.
Question-order Bias
Question-order bias: respondents are primed by the words and ideas presented in questions in a way that impact their thoughts, feelings and attitudes on subsequent questions. For example, if a respondent rates one product 10/10 and is then asked to rate a competitive product, they will give a rating that is relative to the 10 they just provided. The order of answer options is also important because of the anchoring bias, where we tend to rely heavily on the first piece of information provided and judge all other options relative to it. To alleviate the impact of these biases, we tend to ask more general questions first before drilling down into specifics, and randomise the order of brands, answer options and sometimes questions.
Social Desirability Bias
Social desirability bias: the tendency to over-report socially desirable behaviours and under-report socially undesirable ones. If you ask questions in a way that makes people feel judged, they will give you socially desirable answers. Good questionnaire design is the key to avoiding this bias, framing questions as neutrally as possible. Using trade-off exercises, such as max diff can also help us to avoid this bias by forcing respondents to consider all answer options in different combinations and relying on statistical analysis to uncover the most favoured one.
