(This post is taken from a talk I did earlier this year at the Shanghai UI/IUX Meetup and from another blog post I wrote a couple of years ago. Lovely illustrations in the description of the poster study courtesy of Alice Colombo.)
As ever, I seem to catch up with discussions in the world of design at a glacial pace. One such example of my tardiness was that I finally sat down and watched a great talk Mike Monteiro did at Webdagene last year, entitled “This is the golden age of design! …and we’re screwed’:
As part of the talk, he lists the qualities and behaviours that good designers should have, and one of these was that “Designers have reasons”. He argued:
Designers need to able to explain every decision they make in a rational manner and tie them back to project goals. They have to be able to explain their decisions and do so convincingly. They have to sell it. I think it looks good is not a rationale; it’s a red flag… and then you do what? [FIRE THEM!]
Well this all sounds pretty reasonable… doesn’t it? Yes, being able to vocalise why you designed something the way you did – and how it helps the project – is important. We don’t want people to design things for no good reason, or just rely on their status as the rockstar designer to force others into accepting their designs. On the other hand, being able to explain your design decisions should always be less important than the ability to design itself. Otherwise, we end up training designers to use certain types of effective language in post-hoc rationalisations, and then award them based on their ability to do this. Indeed, the nature of the grift is that we often don’t attempt to examine whether the reasons designers give are good ones or bad ones, true or false, so long as hey are internally consistent and superficially convincing.
This is all fine if our goal was to win arguments with clients or stakeholders, but it isn’t – it’s to design effective things that work. Thinking things through and being able to show you have is part of this, but we don’t demonstrate that our design choices are correct and effective by telling stories. We do so when the rubber hits the road and real people use what we’ve designed.
Which is to say, we don’t want to end up rating designers more on their ability to bullshit convincingly than on their ability to design. Now bullshit is something of a technical term, ever since the philosopher Harry Frankfurt wrote his famous essay On Bullshit. He notes that both liars and truth tellers care about the truth, because liars at least recognise the truth they are trying to deceive you about. Bullshitters, however, are different. He writes:
For the bullshitter… He does not care whether the things he says describe reality correctly. He just picks them out, or makes them up, to suit his purpose.
Bullshitters, then, want to present something in a certain light, and will say anything – true or false – in order to give this impression. Typically, this might be that they are very clever, or thoughtful or – in the case of design – that a design has been created according to very rigorous (perhaps scientific) principles, by deep and thoughtful people.
If we say explaining every design decision is important we will end up bullshitting, because not every decision is consciously made. There’s an amount of implicit, unconscious knowledge at work as well. The result is that designers are encouraged to make stuff up – to bullshit – to give the appearance of having thought every last detail through. We thus end up rating designers as much on their ability to post-hoc rationalise their work as on the quality of the work.
Yes, design should be thoughtful and considered and attempting to solve a problem – but exactly why one option solves a problem better than another isn’t always obvious or clear until you test things.
Let’s take the hoary old chestnut of left hand navigation versus right hand navigation (ignoring top navigation for the sake of this argument). The common and oft repeated belief is that left hand navigation is superior (for those used to Latin script) because “people read from left to right”. Yet there is nothing inherent in reading from left to right or in human perception that makes this so. Relevant peer-reviewed studies (Faulkner and Hayton (2011), Kalbach and Bosenick (2006)) find no difference in usability, and any difference we do find may be as consequence of this being a familiar design convention instead of it being universally true (this is also true of the “F shaped” gaze plot for most webpages).
Let’s say you place your website’s navigation on the left unthinkingly, because this is the convention and it’s where you always place it. A client asks you why you placed it there, and you tell them that it’s because “people read from left to right”. The client buys it – hurrah! In the client’s eyes you are a thoughtful and rational designer – well done you! Yet now that you’ve said this you may then make other, dumber decisions further down the line, avoiding placing anything on the right hand side because “people read from left to right”. This could happen either because you’ve created an expectation for the client or because you’ve come to believe your own post-hoc bullshit.
It would be better to admit ignorance, or that you’ve designed according to personal preference or convention, than to make up some bullshit on the spot. Better yet, we need to encourage thoughtful reasons over shallow rationalisations. But why is it so bad to explain design decisions this way? Doesn’t bullshit work?
I’d argue no – you might win a few arguments during a design critique but in the long run bullshit has a corrosive effect on your decision-making processes, on your team and on your client relationship.
Your Bullshit, Conscious & Unconscious
Firstly, bullshit can undermine your own decision making process. You don’t always know why you designed something the way you did. You might have a sense – it’s about visual attention, or spacing, or the user’s memory load or somesuch – but exactly why it feels right isn’t always clear. We all like to believe we understand our preferences and have access to the reasons why we make the decisions we do. Yet this belief is false, and has a name in psychology: the Introspection Illusion. The author of You Are Not So Smart, David McCraney, describes it as follows:
You believe you know yourself and why you are the way you are. You believe this knowledge tells you how you will act in all future situations. Research shows otherwise.
A considerable body of research demonstrates that we don’t always understand our decisions and choices, because the real reasons for those things are not available to our conscious mind. Instead, we create stories using what is consciously available to our mind, but those stories are make believe.
The most famous example of this tendency comes from split-brain patients, who have had the corpus callosum connecting the two hemispheres of the brain severed (this is done in cases of severe epilepsy). Studies with these patients demonstrate that our left hemisphere is often dominant, especially when it comes to language and interpreting actions or emotional states. Gazzaniga (1995) named it the left hemisphere “interpreter” that creates a meaningful narrative behind our actions and feelings. Since the left hemisphere does not know why the right hemisphere is performing an action in split brain patients it causes them to confabulate – create a self-deceiving narrative – to fill-in the gap and make that action meaningful.
For example, Gazzaniga (1995) performed an experiment with (corpus callosectomy) Subject PS. Subject PS was presented with two stimuli images on the left and right of his visual field. On his right hand side, an image of a chicken claw was flashed, and because our eyes connect to the opposite hemisphere (via the optic chiasm, not the callosum), it was presented to his brain’s left hemisphere. In the same manner, an image of a snowy scene was presented to his right hemisphere. When asked to point to another picture that was associated with what he saw, his left hand pointed to a snow shovel and his right to a chicken. When asked about his choices, he replied that “Oh, that’s simple. The chicken claw goes with the chicken, and you need a shovel to clean out the chicken shed.”
The left hemisphere only saw the chicken claw and so pointed to the chicken. The right hemisphere only saw the snow scene and so pointed to the shovel. Yet since the left hemisphere interpreter did not perceive the snow scene it did not know why the subject had chosen to point at a shovel (since it does not have access to the right hemisphere), and so confabulated a story that was consistent with what it could perceive but that was not the real reason for the subject’s actions.
Remember, neuroscientists study brain-damaged or lesioned patients because the resulting cognitive deficits tell us how normal functioning brains and minds work. Work such as this suggests that we are confabulating all of the time, and indeed in his later popular book Gazzaniga (2011) writes:
It is the left hemisphere that engages in the human tendency to find order in chaos, that tries to fit everything into a story and put it into a context. It seems driven to hypothesize about the structure of the world even in the face of evidence that no pattern exists…This is what our brain does all day long. It takes input from other areas of our brain and from the environment and synthesizes it into a story. Facts are great but not necessary. The left brain ad-libs the rest.
As a species then, we are continually confabulating the reasons behind our actions, feelings and beliefs. Can this really apply to our design decisions though? The next two studies suggest that it can.
I think at some point we’ve all encountered stakeholders who forget their original decisions and adamantly maintain that the opposite is what they ask for. Or, times when we have not implemented some of their feedback, only for them to believe the unchanged design is what they wanted all along. Such a tendency to lose track of our decisions and to confabulate that the opposite must be what we wanted all along is known as choice blindness. Studies such as Johansson et al (2005) and Hall et al (2010) demonstrate that when participants are asked to make choices (in the first study, to decide on the more attractive of two faces and in the other study, which flavour of jam or tea tasted better) very few detect when the selection shown back to them later has been switched and is not the one that they initially chose. It’s also worth noting that this phenomenon persisted even when participants were allowed to keep their preferred jam or tea, meaning it occurs even when there are real-world consequences.
Participants would then confabulate, creating a narrative (in much the same way as did the split brain patients) to justify why they chose something that in fact they did not. Moreover, linguistic analysis (in studies such as Johansson et al, 2006) of the participants’ self report shows no difference between confabulating and participants and the control group (whose selections were not switched) suggesting that the self-deceiving narratives created by the participants were believed wholly by them and indistinguishable from any other opinion or attitude.
Not only do we create post-hoc narratives to justify our decisions, those narratives can be created for decisions we did not make and are indistinguishable from our other beliefs. The study that I would love to run would be to reverse some of the elements in a digital design, changing decisions that have been made by the designers and asking them to justify these “decisions”. I’d predict, based on the choice blindness literature, that very many designers would happily defend decisions that they did not make, believing that these were their original choices. Indeed, the phenomenon of choice blindness calls into doubt any of the justifications we make for our decisions; they may all be post-hoc rationalisations of something we would have done anyway. Whether we consciously bullshit or unconsciously confabulate, the information we have access to is not always going to be the same as the real set of reasons why we designed something. It’s certainly possible, therefore, that the language we use to justify why we designed something the way we did is really a set of confabulations, but can this lead us to make worse decisions?
Wilson et al (1989) give the example of a study they conducted into choices of posters. Student participants evaluated five posters, one of which they were allowed to take home and keep. Half of the participants were just asked to take a poster; the other half was asked to reflect on why they made their choice. Those asked to just take a poster typically chose artistic posters; those asked to reflect chose more “pop” style posters (the author’s words, not mine), such as one with a cat and the caption “Gimme a break”.
As you might suspect, those that had to reflect were significantly less satisfied with their choices than those who did not when the researchers followed up six months later. Those reasons for choosing a poster that were most available to the participants were not the subjective aesthetic factors behind their attitudes towards the posters, but rational cognitions that did not match their actual preferences. The psychologist Emily Pronin (2009) describes this as:
The former group placed too much weight on the introspections that they generated at that moment in time, and thus lost sight of their more enduring attitudes.
Overthinking our designs, then, can be as bad as not thinking them through at all. What we don’t want to do is make decisions based on what is easier to rationalise, as opposed to what works best. We won’t always be able to describe why something is a good idea, and the rationalisations we come up with can lead to bad design decisions being made. Indeed, the justifications we give for decisions may be post-hoc, unreliable and prone to a set of cognitive biases. Obviously, we can’t dismiss all and any justifications out of hand but we do need to be cautious both with our own reasoning and that of others.
Of course, you might argue that the solution is simple: that argument is a core part of design, and that bad rationalisations will be smoked out through discussion. This is panglossian for two reasons. First, that both interlocutors in an argument may have distorted post-hoc beliefs for why a certain decision should be made, meaning that who ever wins, bullshit (or at least, unintentional confabulation) reigns. Second, that all too often we are not asked to give valid or correct reasons for our designs but simply to give some, any reason. Indeed, whilst Mike Monteiro suggests that designers can “explain their decisions and do so convincingly” and that they do so in a “rational” way, he does not add that those explanations have to be good ones or somehow correct (he may well feel otherwise, but I’m focusing here on what he actually said). As mentioned earlier, designers are frequently expected to give reasons without any regards to the quality of those reasons – as long as they tick the “can justify designs with reasons” box. If the process of creating reasons occurs after a design decision is made, can be corrosive to the quality the decisions we make and is a superfluous box ticking exercise, why should we justify our design work?
As I’ll argue in the next post, we need to find a balance between accepting some decisions as implicit and instinctual and others as needing thoughtful justifications, and then be careful with the sorts of justifications we give so as to avoid bullshit. The other thing to consider is what stage we are at in our career. Jared Spool, using prior research into surgeons and doctors as a model, defines a designer’s skills growth as having four main stages. First, as neophyte designers we start out as incompetent, both unconsciously and then consciously. As our skills grow, we reach a level of conscious competence, whereby we’re good at doing something and can describe in broad terms why. As we approach mastery, we reach a level of unconscious competence, where we are so good at doing something that we cannot even describe why something works. As we become better designers, we won’t always be able to describe everything we do and why – and thus our willingness to accept decisions based on instinct should be tied to the experience of the designer. I don’t mean “accept blindly”, and we should still want to test those decisions, but we should certainly be more willing than we would be for a novice designer.
Yet even as novice designers, overthinking means we might miss more enduring truths. An example from my own experience comes from a banking website that I worked on a number of years ago. Lots of background research was conducted, and personas & mental models created. This led to a search-driven interface and a “Pinterest” style layout of search results. Asked by the client why this layout was desirable, we came up with a host of reasons about how it was a more engaging way to explore content and came to believe our own reasons.
These rationalisations meant we lost sight of a more enduring truth – that 3 columns is a very hard way to scan through content, and that most users would prefer to view complex banking information in a single column. Had we just sat down to design some search results, we wouldn’t have made this mistake. We discovered this issue through user testing, but bullshit rationalisation undermined our design decisions. We need to be just as wary of unconscious confabulations as we are of explicit bullshit.
Bullshit and your team
Bullshit can also undermine the team decision-making process. By making the need to justify decisions with a brand of loud, client facing bullshit a key trait for a designer we evaluate our team not by the merit of their design skills but through another, orthogonal skill. In doing so, we fail to support junior designers and protect introverts – we don’t want to make decisions based on the personal traits of the designer, but on the quality of the work, no matter how loudly they ballyhoo what they just did.
Bad design justifications can also become a sort of folk wisdom passed down by a team throughout a project. If a creative director justifies a decision with the empty, thought-terminating cliché that “users scroll” through pages, “users scroll” becomes a reason to justify all sorts of bad decisions, like making key calls to action hidden behind a long scroll and a false floor. “Users scroll” is just as stupid as “users don’t scroll” – users will of course scroll in the right circumstances and yes, we don’t always know the size of the first viewport, but key information generally still needs to be prominent at the top of the page. As ever, what is correct depends on context, and simple bullshit slogans that worm their way into a team’s mind-set do not capture this nuance.
Bullshit and your clients
As a designer, you need to work with clients, senior stakeholders and product owners. This needs to be a relationship built on trust and respect, and it’s important to admit when you’re wrong, haven’t thought things through or are relying on intuition. Using bullshit to cheaply win arguments is dishonest, moreover many clients will see through it and the emptiness of what you’re proposing, whether or not you believe your own nonsense.
Here’s another example from my own experience. I was working on the information design of a graph for a public health website. The (otherwise very good) creative director insisted on adding a 3D effect to the bars used in some bar charts.
I suggested to him that such an effect was superfluous and (as any Edward Tufte fans knows) made the data harder to understand. His response was to claim that the 3D effect made the data “more concrete and more real” which is of course nonsense – a bar is always an abstraction of something real, whether it’s 3D or not. The client did not buy this for a second and predictably we went with flat bars. But we lost face with the client, and if the creative director had strongly felt it looked better in 3D and he should have pushed for it on those honest grounds instead of the empty bullshit of “it makes the data more concrete”.
How to banish bullshit
So what does this all mean for our design practise? First, that when asked to give reasons for our designs it’s just as easy to deceive ourselves with confabulations as it is to bullshit clients or stakeholders. Above all then, we should be more accepting of intuition. Yes, design isn’t art. It’s problem solving, and what we create should always be trying to solve a problem. Nevertheless, sometimes it is better to be honest when we have designed something through intuition than to make up some bullshit to justify it. Being honest about intuition means being honest about ignorance – I don’t know why this makes sense but as a professional it seems to. Such knowledge is tentative and is a hypothesis.
And if you have a hypothesis, you test it! Sometimes it’s okay to see if something you’ve designed works after you’ve done it – but you need to test if it works. Let’s say you have an intuition that a button is best placed somewhere because users will notice it there. There’s no better way to justify this hunch than with solid user testing or A/B test data – did they notice it? And you’re more likely to convince stakeholders with solid evidence to back up your hunches.
This doesn’t totally save us from rationalisations – we can still create superficial narratives describing what some data means – but it does help us avoid creating plausible stories to describe decisions, stories that can lead us astray. It also means that we place the effectiveness of our design work above the ability to spout wishy-washy nonsense to clients and stakeholders. Doing so will improve your work, your team and your relationship with your clients.
We still do need to justify our work at times though – Mike Monteiro is definitely right on that count. We can’t use “intuition” and “let’s test it” as an excuse every time – and especially not “let’s test it” for questions of what is aesthetically better – and articulating our design decisions will always be an important skill. In my next post, I’ll outline what I feel are good and bad justifications for design decisions. Even if we cannot always avoid fooling ourselves, we can avoid the most pernicious forms of bullshit.
Faulkner, X. & Hayton, C. (2011). When left might not be right. Journal of Usability Studies, 6, 245-256.
Frankfurt, H. G. (2005). On bullshit. Princeton: Princeton University Press.
Gazzanigga, M. S. (1995). Principles of Human Brain Organization Derived from Split-Brain Studies. Neuron, Vol. 14, 217-228.
Gazzanigga, M. S. (2011). Who’s in Charge?: Free Will and the Science of the Brain (1st ed.). New York, NY: Ecco.
Hall, L., Johansson, P., Tärning B., Skiström, S., Deutgen, T. (2010). Magic at the marketplace: Choice blindness for the taste of jam and the smell of tea. Cognition, 117 (2010) 54–61.
Johansson, P., Hall, L., Skiström, S., Tärning B., Lind, A. (2006). How something can be said about telling more than we can know: On choice blindness and introspection. Consciousness and Cognition, 15 (2006) 673–692.
Johansson, P., Hall, L., Sikstrom, S., & Olsson, A. (2005). Failure to detect mismatches between intention and outcome in a simple decision task. Science, 310, 116–119.
Kalbach, J. & Bosenick, T. (2006). Web page layout: A comparison between left- and right-justified site navigation menus. Journal of Digital Information, 4 (1).
McRaney, David (2011). You Are Not so Smart: Why You Have Too Many Friends on Facebook, Why Your Memory Is Mostly Fiction, and 46 Other Ways You’re Deluding Yourself. New York: Gotham /Penguin Group.
Pronin, E. (2009). The introspection illusion. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology, (Vol. 41, pp. 1-67).
Wilson T.D., Dunn D, Kraft D, Lisle D (1989). Introspection, attitude change, and attitude–behavior consistency: the disruptive effects of explaining why we feel the way we do. Advances in Experimental Social Psychology, 19:123–205.