If at first you don’t succeed, goes the old saying, try, try again. Good advice, up to a point. But let me offer a modification: even when you do succeed, try, try again. Tempting as it is to declare victory and move on, in many endeavours there is much to be said for rethinking an apparently satisfactory formula.
Consider the advice for job interviewers in Talent, a new book by economist Tyler Cowen and venture capitalist Daniel Gross. They suggest asking a routine question, such as “give me an example of when you resolved a difficult challenge at work”. Then ask for another example. And another.
The pat answers will be exhausted quickly, and the candidate will have to start improvising, digging deep — or perhaps admit to being stumped. “If the candidate really does have 17 significant different work triumphs,” write Cowen and Gross, “maybe you do want to hear about what number 17 looks like.”
One way to describe this tactic is that the interviewer is asking for answers in parallel rather than answers in series. Instead of stringing together a logical sequence of 17 questions, the interviewer is asking for 17 different answers to the same question.
It is counterintuitive advice, but the logic is clear enough and it seems simple to execute. So why don’t we do it? First, we feel uncomfortable. Second, demanding 17 different answers to the same question may seem silly or fruitless.
While the approach is unconventional in job interviews, it is common practice among designers. They will often produce several distinct attempts to meet a given brief, rather than immediately focusing on what seems to be the best idea. In doing so, the designers force themselves to explore the full range of possibilities, to avoid the risk of committing too early to a concept that seems attractive but which may eventually be a dead end.
Researchers Steven Dow, Alana Glassco and others at Stanford University explored this idea by asking experiment participants to use simple software to design a web advert for a magazine. Half of the participants worked in series: they sketched five prototype adverts, receiving feedback after each one. The other half worked in parallel: they sketched three prototypes, then received feedback on all three and then sketched two more before receiving feedback.
Dow and his colleagues asked experts to rate the quality of the final adverts and tested them on the internet, measuring click-through rates. They rated the diversity of the adverts and they also asked the participants for their confidence after finishing the process. In every respect, the parallel process ads were superior: the final designs looked better and earned more clicks; the initial sketches covered a greater range of ideas; and the budding amateur designers gained in confidence as a result of prototyping in parallel.
A striking example of parallel design is the creation of the Windows 95 start-up sound. Microsoft was looking for an opportunity to show off the burgeoning audio capabilities of the computers of the day, so somewhat implausibly it commissioned Brian Eno, whose previous collaborators included David Bowie, Talking Heads, U2, Devo and Roxy Music.
Eno recalls receiving a brief, requesting music that was “inspirational, sexy, driving, provocative, nostalgic, sentimental . . . there were about 150 adjectives. And then at the bottom it said, ‘and not more than 3.8 seconds long’.”
Eno describes himself as being “completely bereft of ideas” at the time. He found the brief both hilarious and inspiring. In the end he composed more than 80 tiny pieces of music. The final result was a musical signature that has stood the test of time, and a liberated Eno. “It really broke a logjam in my own work,” he told the San Francisco Chronicle.
No doubt that was partly a response to the tight constraint of a 3.8-second piece, but surely it was also the creative response to trying to produce the 83rd composition. When it would have been easy to put his nose to the mixing desk, obsessing over the tiniest variations of tone and timing, Eno forced himself to explore the possibilities.
Bill Burnett and Dave Evans, in their delightful book Designing Your Life, suggest an exercise in which you sketch out a vision for the next five years of your life. What will you be doing? Where will you live and with whom? Are you hoping to run a marathon? Start a business? Write a novel? Get married?
This is often a straightforward act of imagination, but what makes the exercise excruciating is what comes next: Burnett and Evans want you to do it again, only this time, it’s different — the idea at the heart of the plan is completely forbidden. Back to the drawing board. And then a third time.
I’ve tried this myself and seen others try it. People squirm. They protest. Sometimes they cry. And then, sooner or later, the ideas start pouring out. We contain multitudes, all of us. But we don’t always let them see the light of day. Perhaps we should try producing answers in parallel more often. Even when you do succeed, try, try again.
Written for and first published in the Financial Times on 3 June 2022.
The paperback of The Data Detective was published on 1 February in the US and Canada. Title elsewhere: How To Make The World Add Up.
I’ve set up a storefront on Bookshop in the United States and the United Kingdom. Links to Bookshop and Amazon may generate referral fees.
