It wouldn’t be unreasonable to say that in the UK, we are surrounded by an abundance of choice, and yet this choice is something we often feel somewhat ambivalent about. We feel overwhelmed and overloaded, perhaps even ending up buying less if we’re faced with too many options. But we should recognize that we are privileged to experience choice abundance and that things might feel worse if we faced a choice deficit. A recent study with over 7,000 participants from six countries found that choice deprivation—a feeling of not having enough to choose from—not choice overload is the most common consumer experience in both trivial and highly consequential domains. And that choice deprivation isn’t just more common—it’s also more harmful to choice satisfaction than overload.
So faced with choice abundance we need help and that’s where thinking about design of choice architecture is important. There’s a whole slew of books and articles covering the topic, including the one I’ve read recently – The Elements of Choice by Eric Johnson.
People have got rather clever at the whole choice design thing and apparently some organisations use our increased understanding in this space to do more harm than good. I’m thinking of ‘dark patterns’ – design elements that make people select options they did not mean to – for example opting in to receiving marketing comms inadvertently. Or how about the now infamous ‘sludge’, sometimes used deliberately, and which ends up making it harder for the user to do what would be in their best interest, such as unsubscribe from a service for instance.
At the very least, we should be aware of how different decision architectures can influence the options we go for. Let’s begin by imagining that it’s a lovely sunny day and we go to get an ice cream. At the shop, there’s a list of the ten different flavours on offer, all priced the same, but with details on the inspiration for the ice cream recipe and the origin of the carefully sourced ingredients. You look down through the list, reading carefully through the details for the first couple of flavours. You’re likely to remember how tempting the crunchy caramelised pecans sounded in the ‘Vanilla with a Twist’ option at the start but after that, it becomes hard to focus on what’s in the later options. It wouldn’t be surprising if you ended up just going for that first option – the special vanilla. This is an example of the primacy effect, when the option taking first position in a list has an advantage.
But it doesn’t always work like that. Imagine working through a set of options differently. This time, you’re deciding with a friend what film to go and see. The friend talks through all that’s showing; it’s a small multiplex, so perhaps there’s only five screens. You listen, but by the time the friend has listed all the films and a bit of background behind them, you can’t really remember what the first was about and have latched onto the fact the one mentioned last is based on a book you loved, so it just seems obvious to go for that last film.
On this occasion, it’s the recency effect that’s operating, where being last (or most recent) is best. This scenario is also an example of a sequential presentation – the friend walked through describing film one after the other – and with sequential presentations, how long the list of options is and where they are in the order can make a big difference. An experiment was carried out asking people to choose their favorite drink off a list. In the list with two options, the first drink was chosen 70% of the time, effectively meaning it was picked twice as often as the second and much more than we would expect by chance. That’s the primacy effect in action – what’s first has an advantage.
But what happens when the list is longer? The last drink in the sequence starts to do better – the recency effect kicks in action. This is all worth bearing in mind – with sequential presentations, it suggests that if the list is short, it’s best to be at the start; if the list gets longer, the advantage of being last grows. Bear that in mind the next time you’re trying to agree what to watch on Netflix.
Defaults are another important ‘tool’ in the world of choice architecture. They can represent one of the best ‘helping hands’ for people trying to work out what they should go for –they’re effectively the option that’s selected if the user does nothing. An obvious example is in auto-enrolled pensions, where, in general, people are automatically set up to contribution 4% of their salary, even though they could ask to have it set at any level they want. But how does the ‘system’ know what the default should be?
Perhaps there are some clever people who have worked what seems like a good starting point, which is sort of what was the case for auto-enrolled pensions. (However, though 4% contribution was actually thought of as the minimum and that people would raise it as appropriate. No one actually does – and that’s the endorsement effect for you: “The government set it at this level – must be right,” goes the thinking.)
But what if you have to design a system to put in a default, should it be based on the commercial imperatives of the organisation? Perhaps pre-fill the choice of type of seat with first class? Not sure that would go down well. But what if the ‘system’ can be smart, and look at all the other choices that have been made earlier, could it put the default as the most popular choice so far? Or even better, what if that the ‘system’ had invited the user to register, and therefore could use its knowledge of the user to make a smart suggestion for the default? I think this is happening in my online grocery shopping: I can start my weekly shop with my ‘favourites’ I’ve ordered before. Yes, I guess that’s quite helpful, but actually I would rather be inspired out of the rut of ‘the usual’!
When we think about how we approach making decisions, there’s sometimes actually a bit of a paradox. We have two conflicting beliefs about our free will. On the one hand, we sometimes think of ourselves as subject to the influence of external forces – with all those algorithms, cookies and other sneaky tactics that make us see things in a particular way, we’ve got no chance of deciding for ourselves. For example, we can’t really even trust the order of a list without wondering why something is at the top. On the other hand, we like to think that we are masters of our own destiny and can decide what we want for ourselves. We can see through all those techniques: we’ve learnt to zone out the annoying repetitious adverts for toasters because you looked for toasters last time and we know exactly what we’re doing and why…..
Perhaps the most telling thing is that research has shown that even when the presence and intent of choice architecture is disclosed, it doesn’t really change how people made their decisions. It’s not just choice overwhelm we experience – we’re also rather lazy!
