Most design failure is decision failure
Human decision-making is influenced by many factors we are oblivious to. Awareness of them leads to better decisions and better design.
Design is the practice of making deliberate decisions about how a system behaves.
That sounds pretty straightforward until we consider the mountain of ways in which we can inadvertently make suboptimal decisions. Suboptimal is one of those awful management-speak words but it reflects the reality that few decisions are binary good or bad, right or wrong, but rather that different decisions carry different risks, rewards, trade-offs and outcomes. Designers constantly need to make black and white decisions in a sea of shades of grey.
Sure, we can make decisions – we have opinions on everything – but deliberate ones? Well, that is another matter.
It’s not that we don’t recognise the importance of deliberation, but rather that we make so many of our decisions unthinkingly and arbitrarily based on personal taste and mood. This often doesn’t matter. Will I have Corn Flakes or porridge for breakfast? Will I wear jeans or slacks to the restaurant? Will I watch Breaking Bad or Succession?
The difficulty is that the brain processes which get you a tasty breakfast or an enjoyable evening watching Netflix will build you a horrible product.
The liberating open secret about decision-making is that no one knows if their decisions are right because no one knows the future.
When a group of people in a professional environment collaborate to make a decision, all sorts of unseen and unspoken influences can distract them from good outcomes:
Consistency is not correctness – everyone around a table can have a consistent opinion but that doesn’t mean that it is correct. Teams armed with the wrong data, irrelevant insights, strong personal preferences or no sight of the customer can quickly make decisions in no-one’s best interests.
Bias and noise mean we don’t see the world as it really is – bias is systemic error, where understanding is removed from the truth in a particular direction, whereas noise is random error, representing outcomes arising from chance, context or inconsistency. A biased gambler might always overestimate their favourite team's ability whereas a noisy gambler just gets the score wrong.
Social bias – corporate culture and (most) Western societal cultural norms dictate that the view of older people (up to a maximum threshold age), or higher paid people, or more senior people gets more air time and is afforded more gravitas than others. This means that while your boss may be right more often than wrong because they have more experience, it also means that if they are talking nonsense, or you disagree with them, it can be difficult to let them know that.
We get wedded to our opinions – one of the great bugs in the human brain’s operating system is that we are often more wedded to our unevidenced opinions than we are to our evidenced ones. We might get angry at our friend because of their views on religion or American politics, but if they have a different view to us on the laws of physics or how mathematics works, we just think they’re a bit crazy.
Bertrand Russell summarised it brilliantly in his 1943 tome An Outline of Intellectual Rubbish: “If an opinion contrary to your own makes you angry, that is a sign that you are subconsciously aware of having no good reason for thinking as you do. If someone maintains that two and two are five, or that Iceland is on the equator, you should feel pity rather than anger, unless you know so little of arithmetic or geography that his opinion shakes your own contrary conviction. The most savage controversies are those about matters as to which there is no good evidence either way. Persecution is used in theology, not in arithmetic, because in arithmetic there is knowledge, but in theology there is only opinion. So whenever you find yourself getting angry about a difference of opinion, be on your guard; you will probably find, on examination, that your belief requires more evidence.”
The only means by which designers can avoid these wrong turns is to identify them, and be deliberate about mitigating against them. Here’s how:
Different decisions carry different risks, rewards, trade-offs and outcomes. Designers constantly need to make black and white decisions in a sea of shades of grey.
1. Focus on the job in hand
We daily remind ourselves of the problems which we are there to solve, which nearly always fall into the following two categories:
i/ The business has a series of goals it hopes to achieve, which the product or service you are designing should contribute to
ii/ The customers of the business have a range of needs to be met and behaviours to be accommodated, through the product or service you are designing
Clarity on those problems and a plan about how to solve them keeps project teams focused on the job in hand, avoiding distractions and getting hung up on vanity sideshows.
A half design of a good problem is better than a full design of a bad problem.
2. Be deliberate about decision-making
It seems obvious that this would be the thrust of this piece, but just making project teams aware of the arbitrariness trap is the first step in helping them view decisions differently. Many people just don’t notice social factors, noise and bias in decision-making but can recognise them when they are described.
Engineering has adopted the concept of decision intelligence, which is baked into its processes for complex civil, hydro, mechanical and chemical projects. Faced with the overwhelming complexity of the challenges, project teams often ask “before we start to solve this problem, let’s agree how we’re going to solve it.” Recognising all of the factors which can lead them away from the best solution, they consciously decide at the start of any programme of work how decisions will be made and tested.
3. Treat all decisions as working hypotheses until they are tested in the wild
The liberating open secret about decision-making is that no one knows if their decisions are right because no one knows the future.
It makes sense therefore, that decisions are tested as quickly and inexpensively as possible. Build wrongness into your processes and your culture. Make it OK to be wrong. Celebrate when wrongness is caught early and inexpensively.
Design processes can never be perfect, or ‘right’, but they can reduce the chances of being systematically wrong. We can never remove uncertainty entirely, but we can narrow the range of bad outcomes.
Design like you’re right and test like you’re wrong.
4. User behaviour is the only real measurement of design decisions
Design is the optimisation of a system towards a desired set of effects.
All design influences behaviour and the best design encourages the best and most desirable user behaviour. Aesthetic design decisions should be made in the pursuit of desired behaviour. Functional product decisions should be made in the pursuit of desired behaviour.
Improve the design, improve the behaviour – the user wins.
Improve the behaviour, improve the outcome – the business wins.
User opinion is notoriously unreliable when it comes to assessing performance in the real world. Consider a well-known comedy show which was trying to break into the US. During focus groups, which were set up to explore if the UK show would be suitable for an American audience the show was ranked joint last, alongside women’s bowls.
Audience opinion – this show is unwatchable, as boring as bowls.
The show, The Office, went on to be most streamed TV show in America, with 40% of Americans having watched at least one show. It received 42 Emmy nominations and won 5, it was named the best TV series by the American Film Institute in 2006 and 2008, won two Screen Actors Guild Awards for Outstanding Performance by an Ensemble in a Comedy Series in 2006 and 2007 and won a Peabody Award in 2006. Steve Carell won a Golden Globe for Best Actor in 2006.
Audience behaviour – we are going to devote 57 billion minutes of our lives in 2020 alone to watch this.
User behaviour eats user opinion for breakfast.
Design doesn’t fail because designers aren’t smart enough, creative enough or talented enough. It fails because teams make decisions unthinkingly within systems which don’t test them soon enough and punish wrongness down the line. Design’s role is to slow the big decisions down just enough to ensure they are focused around the right problem, deliberate about how conclusions are reached, and tested by real users as part of the process.
