Why do I like pillows?

How will I know if I like something that I have never tried before? With technology offering a wide range of product and services, this is becoming an increasingly important question for brands. We are arguably in the age of disruptive rather than incremental innovation. So when there is no precedent how can we gauge what consumers want?

Market research industry has a range of sophisticated tools to help – from ethnography to conjoint. Despite the predictive validity of many of these approaches there is, nevertheless, an increasing question mark. Has behavioural science not indicated that our preferences are constructed by context rather than our needs? If this is the case, how will asking people questions surface the issue?

That’s the question psychologist Itamar Simonson posed himself about pillows. Rather unusually he had happily slept without the aid of a pillow until mid-life and as such he did not really see the need for one. However, when Simonson was in a shop one day, the promotional offer on pillows ‘nudged’ him into buying one. Upon using the said pillow, he found he greatly enjoyed the experience.  

This simple issue goes to the heart of an issue vexing the market research industry. Many of the ‘traditional’ approaches of the industry (e.g. surveys and focus groups) broadly rely on consumers reporting their intentions, so reflecting the assumption that we have stable preferences that shape our behaviour. 

But of course Daniel Kahneman’s Nobel Prize reflected his work on demonstrating how non-conscious ‘system one’ thinking was a significant driver of behaviour – that is how the environment we are in ‘nudges’ our choices. This view of the world suggests that our preferences are not stable and consistent but can easily be manipulated.

So how does this work for Simonson and his pillow? First, he agrees that in many instances our preferences are largely determined by the task characteristics, the choice context, and the description of options.  These were the factors that persuaded him to try a pillow in the first place. 

But in his paper he makes the point that to attribute all our behaviour to this, makes a significant over-claim. In his view we also have more stable inherent preferences that are not always determined by context. 

Simonson argues that much of the time we actually have stable explicit preferences that we are conscious of – such as our liking for heavy metal, action films, sitting forwards on a moving train, cucumber sandwiches and such like.  But some of the time we also have stable implicit preferences that are non-conscious, that often only emerge upon experience. 

So by resting his head on a pillow at night he could start to experience the feeling of softness, the levitation of his head and so on. It would be hard to anticipate whether you liked or disliked these if you had never tried a pillow before.

So to Simonson’s point, we can start to see the qualifications and boundary conditions to situational influences on preferences.  Specifically:

• We can have explicit stable preferences that are resistant to situational effects. If I like heavy metal and hate jazz, then no amount of ‘nudging’ will persuade me to start liking John Coltrane.
• For those products/services where people are not able to anticipate their likely experience, we are likely to be more influenced by situational factors – we will use the information in front of us to make decisions.
• Where products and services are low in experience then we are likely to remain influenced by situational factors. For most people at least, toilet cleaner is not particularly experiential so our enduring use of a particular product is likely to be influenced by habit rather than preference.
• For more ‘experiential’ product or services, our continued use may well be determined by our implicit preferences. 

So what are the implications of this for understanding consumer behaviour?  I think there are a number of points that we can usefully make: 

• Much of the time, it continues to be reasonable to ask people questions about their preferences. There are a broad range of activities where we have insight into our own preferences and are able (given sufficient skill on the part of the survey design) to anticipate the degree to which the item or experience will meet our desires. 
• For new products and services where consumers are unlikely to be have encountered similar situations (and therefore not be able to anticipate their preferences) we ideally would offer consumers a trial version which allows their preferences to emerge.
• More work needs to be done to measure implicit preferences – my view is that technology has the potential to rehabilitate projective techniques.
• We can start to define the conditions under which situational factors drive choice and as such we can focus on identifying and measuring their impact.

Understanding consumer behaviour requires an integrative approach. An effective practitioner needs to draw on behavioural science as well as more traditional ‘self-report’ techniques to understand consumer behaviour. It’s worth making the point that our implicit preferences will interact with situational cues – it’s not necessarily one or the other.

Market research is undergoing a renaissance as new techniques enter our repertoire – the challenge and opportunity is in the understanding what each offers and where the boundaries are.  

Colin Strong is global head of behavioural science at Ipsos