It’s widely agreed in the market research industry that just asking people their opinions isn’t enough to provide insight. I could give you a list of reasons why this will only result in top-level, conscious feedback – we post-rationalise, lie to ourselves, aren’t conscious of why we behave the way we do, etc – but I’m sure you’ve heard them all before. ‘Findings’ – i.e. the things that people say (and think) they do – are just the tip of the iceberg. What lies beneath the surface is where we discover insight.
Insight goes beyond direct question and response. Below the waterline is where you start to really discover your consumer: things about the human condition, emotions, and the dynamics of what really influences behaviour and choice.
Take website usability – a ‘finding’ might be when people tell us they want a big red button to view their basket. An ‘insight’ will explain why they’re completely blind to and distracted from the big red button that’s already there, because their eyes are drawn to the striking visuals of cars, hotels, or other goods which are significantly more eye-catching than the functional aspects of the site we’re testing.
It’s difficult to pin down a single view of what an insight is because different people, including our clients, have wide-ranging definitions. These can depend on a business’s knowledge of the category, the issue they are trying to understand, or even the level of sophistication about consumers within that business. One of our clients, for example, defined insight as a “penetrating discovery about consumer motivations that can be applied to drive growth” – an interesting perspective which places importance on the application of the information as much as its depth.
So while there’s no universally agreed upon definition of an insight in research, we’d suggest ticking most, if not all, of these boxes:
- Does it explain a consumer or an aspect of human behaviour better?
- Is it just data or have you subjectively applied your researcher’s point of view to interpret it?
- Is it true and relatable, without being completely redundant and overly obvious?
- Will your client/stakeholder be able to buy into it?
- More importantly, will the organisation or brand benefit from it?
So we’ve broadly ‘defined’ our elusive insight. But now we know what we’re looking for, how do we find it? Well, of course it varies, but here’s one real client example in which we used behavioural economics (BE) as a lens for research design, analysis, and to identify insights.
Our client, a national restaurant chain, wanted to explore how consumers make decisions about what to order, in order to optimise their menu items and menu design. We turned to BE because we know that in-the-moment choice is an incredibly difficult thing to pull apart and reflect on in a post-rationalised way. BE starts with a very simple proposition which many market researchers have already instinctively bought into: that what people say they do and what they actually do can often be two very different things!
We tasked our participants with carrying out self-ethnographic restaurant visits, using a mobile app to film their experiences of navigating the menus and ordering, which we then combined with post-rationalised through depth interviews and co-creation sessions conducted with our self-ethnographers.
By creating a list of heuristics and biases which we expected we might observe in the self-ethnography, we were able to identify which ones were at play during the decision-making. This meant we could virtually observe the mental shortcuts which reduce consumer stress and effort and significantly impact decision-making, and we observed a lot of them. By speaking to our participants in detail after the self-ethnography, we were able to play back what we’d observed to them, telling them what we saw (and what they may not have been aware of), creating a pleasantly disruptive dynamic for both our interviewees and our analysis, and really unpicking what influenced them in the moment.
We determined that many customers were heavily influenced by ambiguity bias and risk aversion, where the non-traditional layout of dishes on a menu and the vocabulary used meant people tended to stick with ‘safe’ options, or dishes they knew would potentially bring a more favourable outcome. We also learnt that, in the moment, herd mentality kicks in, and many customers were influenced more by who they were dining with or what they saw others eating than by the menu itself. In addition, many customers took what they perceived to be greater risks (in this case choosing less familiar dishes) when their perception of safety increased, such as when the occasion was deemed less important.
Insights like these (e.g. that people don’t actually choose curries over ramen because of the taste, but rather because they’re deemed a safer option for reasons they can’t articulate) encouraged our client to think about their menu differently, from the design, language, and visuals they used, to how technology and in-restaurant theatre could help consumers better navigate their offer.
So there isn’t necessarily a single definition of what an insight is, and no set way of discovering one, but whatever definition you use, an insight should help explain behaviour, be relatable, and be something you or your client can benefit from. Otherwise it might be just a finding.