Jonny Longden, Conversion Director at Journey Further, shares his expertise regarding consumer behaviour and how to maximise your ecommerce strategy.
Customer insight and research aim to help businesses understand behaviours so that they can make decisions about how best to serve those customers with products or experiences. Surveys, focus groups, behavioural data – there are many different ways of trying to understand how customers think, feel and act and, of course, what they want and how to provide it. But can we ever really know what makes customers tick?
I’ve been running controlled experiments on websites for nearly 14 years now, and almost all those ideas come from customer insight and research. Whilst this insight certainly makes it more likely that these ideas will be successful, it’s a marginal gain rather than a certainty. The reality is that most businesses execute ideas from insight without ever testing them, so they would never actually know the real impact.
The fact is this: no matter how good your insights are and how obvious your ideas seem, there is an incredibly good chance you’re wrong and they won’t work. The world of experimentation, where all ideas are tested, tells us that your chance of being right is actually only, at best, around 20-30%, no matter where your idea came from.
Why is this?
Ultimately it is because ‘understanding customers’, which is what we’re trying to do, is impossible. There are 3 main reasons for this:
-
Customer behaviour is not conscious and, therefore, is not rational
Customer insight is a process of creating very rational and logical reasons for behaviour. We create a picture of a customer’s context and motivations and then infer how they would make choices and decisions based on this situational analysis.
The issue with this is that it is a) a hideous over-simplification of both human life and consciousness and b) it is rational where, in reality, customer behaviour is unconscious and irrational.
It’s fairly well established that up to 95% of all brain activity is unconscious. The idea that we have complete conscious control and volition is almost certainly an illusion. Furthermore, a lot of customer behaviour doesn’t even pretend to be conscious. How often do you do things on ‘autopilot’ without really even thinking about them at all?
Customers don’t know why they do what they do, so how could you possibly know?
-
Your customer insight is YOUR customer insight
All data analysis is entirely subjective on the part of the analyst, as well as their audience. Data is merely a stimulus; literally, sometimes, a load of 1s and 0s. What you make of that data is your own interpretation and cognitive bias.
And just as your customers are making their choices based on subconscious drives and motivations, so are you when you analyse data. The whole context of your life, experiences, biases and more, is brought to play when you believe you are being objective.
So when you believe you understand customers, in a way you are only understanding yourself.
-
There is no way of ‘knowing’ a mass of people
The only way you can ‘understand your customers’ is to generalise them into a single person. You’re trying to think of the way someone would behave because it isn’t possible to think of how multiple different people would behave at the same time.
Even if you build some kind of segmentation, you are still massively generalising within each segment.
In reality, there is no way of generalising a mass of people and their behaviour. Therefore, any meaningful insight about behaviour cannot be anything other than a very myopic view of a tiny proportion of those customers.
What can you do?
It’s very simple: experiment. Test everything.
Scientists create hypotheses about things (such as how to build rockets or create medicines) based on observations and theories, but they don’t just build them and hope for the best. They experiment, fail, learn and iterate, over and over again until they achieve the goal.
Businesses shouldn’t be any different! Customer insight is useful and valuable, but you must test your findings.
