As David Pye observed, every design is a failure.
Editor’s note: David Pye OBE (18 November 1914 – 1 January 1993), was Professor of Furniture Design at The Royal College of Art, from 1964 to 1974, in addition to being a respected wood turner and designer in his own right. He also wrote several notable volumes on design theory. This article was originally published as part of DTW’s Compromise theme in January 2017.
His argument rested on the idea that no design can optimise every aspect. The more complex the object the more likely this is to be the case. If we take a simple example of a knife, it’s a compromise because unavoidably the designer had to work within constraints of time and materials. The knife has to function but be affordable and attractive to enough people to make it an economically feasible proposition. The best knife can’t appeal to everyone. For some consumers the design is unacceptable – it remains unsold for reason of price or appearance.
If we turn to cars, the same applies. An Aston Martin DB9 is undoubtedly a fast and capable car. It costs more than most people can afford and it doesn’t go far because its powerful engine needs a bigger fuel tank than can be accommodated in a small enough package.
The VW Golf is a best-seller because it makes compromises to suit the largest number of people yet it’s a failure too. It’s not really fast. It’s not really big. It’s too big as well: ask the owner of a Smart ForTwo. People who live in Kensington or the Lake District want height and off-road capability.
Yet who would actually call the Golf or its peers’ failures? These vehicles are sold in prodigious numbers as they satisfy enough of the needs and wants of enough people. So, when Pye talks of the failure of designs, he is drawing our attention to tolerating the right kind of failure and focusing on the most important aspects to prioritise.
It is ironic that the most loudly lauded cars are probably the ones with the narrowest range of abilities: broadly they are very fast. They are also capable of very little else: don’t try shopping for the week with a Ferrari Scaglietti. Don’t plan a regular long commute in a Bentley Mulsanne, not unless you’re fabulously well-off and not a little bit socially unaware.
The Lotus Elise can really only go around on tracks at a good clip or maybe take a trip with a squashy bag to Fitness Universe. Speed and acceleration impose huge costs on a car if they are to be prioritized. Demanding a lot of room inside the car constrains handling though in real world conditions a Ford Transit will get as quickly from Paris to Dijon nearly as fast as an Aventador.
Broadly, compromise is a thing for the head and unswerving focus is a thing for the heart. Eventually one can greatly admire a balanced design. That’s not quite good enough in the showroom where visceral impulses are the ones that make sales. At least relatively visceral impulses: is the Golf a car for the viscerally rational?
Driven To Write 
Look back at any of those authoritative lists of ‘100 Great Designs’ and see how many of them pass the test of ages. Those original iPhones that design fetishists queued in the rain for 3 hours to get, then took home and let their fingers caress until …. well, they sit unloved at the back of a drawer somewhere.
And even if you can look at a product and say it’s perfect, nothing could look better and fulfill its function better, it is a failure because, however good, it won’t have the ability to keep the loyalty of its fickle owners. Of course VW are an object lesson in the perils of believing otherwise. Having cleverly ingrained the ‘iconic and timeless’ (as they wouldn’t have said back then) status of the Beetle in its customers’ minds, they ended up believing it themselves, and their inability to see its many failures nearly ended the company.
Eventually every design becomes a failure as needs change. Bristow´s Patented Horse Trough of 1872 won gold medals at the Paris Exposition of 1873. It was the sine qua non of troughs and Prince Albert had two of them for his prize horses. But today they are only collectors´items. I was thinking of the fact that, narrowly, a design is a failure the moment it is launched and can only be so. The more uncompromising a car is the smaller its audience (the price is too high or the ability too limited) and less compromising a car is the less it excels and a peer can be found to do any one thing better. To their credit, VW found a way to appear uncompromising: fastidious design. That is something people notice but which does not impede the compromises elsewhere. I won´t say I think it´s inspiring yet customers clearly like the proposition.
But you’ve chosen an example (horse trough) whose use became redundant. But that shouldn’t preclude us from praising Messrs Bristow’s effort. In an alternative world it’s still in production, but in carbon fibre.
In cars there were steps forward (say compound curved glass) that allowed for more sculpted design which, inevitably, made what went before look compromised. Otherwise, assuming our hypothetical party of aliens were to be presented with Golfs Mark 1 to 7, how would they order them? Would they think Golf 1 was the latest, since VW had managed to make it more compact? Or would they think it was Golf 4, since it was the most aesthetically pleasing. Unless they subscribed to the universal edict of Sensual Purity in which case they’d judge it, correctly, to be the one with the most creases and vents.
I find my iPhone 6 worse looking than its predecessor.
iPhone 6 is a good example how marketing-led iteration can be problematic in design terms. Speak to any Apple aficionado and most of them will admit that iPhone design probably peaked with the 4/4S/5. But Apple must offer visible differentiation to keep shifting units, so off they go. Rumour is that the 8 (coincidentally the 10 year anniversary of the iPhone) may revert to glass front and back with a metal bezel all round.
Good morning Richard. I wonder how one should think of designed products that evolve very slowly, like garbage bags. They have changed a bit, but fundamentally stayed the same. Toilet paper is another example.
Interesting article. From my own lowly position as a graphic designer I see these compromises being enacted every day. A graphics solution is by necessity a compromise between the client’s expectations, their budget, and sometimes most crucially, their desire to shoehorn in as much content as possible. Of course, car manufacturers are in exactly the same boat, except their budgets run into the billions, and the consequences of failure are so much keener.
Anyway, the picture at the top of the article: is that a group photo of absolutely everyone employed at Lotus now?
Good morning Richard and thank you for a thought provoking piece. As a rank amateur with no training in the subject, only a healthy interest and keen eye, I would argue that, in matters of design, there is always a tension between functionality and aesthetics. Hence, it is much ‘easier’ to design a high-performance GT coupé than an urban runabout. With the, former, compromises in functionality are expected and acceptable as long as it it looks great and has excellent dynamics and performance. With the latter one has to optimise a much wider range of requirements in a much smaller physical package. That’s why I would regard the VW Up! as a much greater achievement in design terms than an Aston Martin DB9, even if the latter is strikingly beautiful.
Completely agree on the VW Up! versus Aston Martin!
VW achieved to make a city car with optimal practicality and visual character at the same time.
I beg your pardon but I don’t think that a GT necessarily is more compromised than a You-pee-exclamation mark.
That a GT doesn’t offer room for four is not a compromise because it was not a requirement in the first place. The GT has to meet a much narrower and more clearly defined set of requirements than an everyday runabout for everybody. This doesn’t make the task of desinigning (as opposed to merely style) one much easier because expectations are higher and more clearly focused.
This gets us to the concept of customer satisfaction by meeting their requirements.
And to the fact that I don’t care about JD Power customer satisfaction rankings because my requirements are fundamentally different from those of customers buying a Lexus.
Hi Dave. I didn’t intend to imply that the DB9 was compromised per se, but simply that it was expected to be great at being a grand touring coupé and not expected to be great at many other things, whereas the Up! has a more generalist role in a much smaller package, hence a more challenging brief for a designer.
Surely a design is not a failure, however compromised it might be, if it achieves its’ goal. Compromise need not be a vice.
The DB9 might be a failure because it isn’t the best-looking Aston, or the nicest megabucks GT car, but the Elise is a success because it was the best at doing what it was meant to do.
By putting it in such stark terms, I think Pye was trying to convey that it’s about focus: since any design carries compromise – and thus “failure” – your task as a designer is to think carefully about which “failures” are acceptable and which aren’t, before even beginning any other part of the design process. At the same time thinking like this serves to make you less afraid of the notion of failure and thus better equipped, again, to focus on which failures to tolerate and which to prevent (or catch in time). The Lotus tolerates a lot of compromise in a lot of areas in order to be very good in a few while the Golf (or the Up) tries to minimise compromise over as many areas as possible, but sacrifices being exceptional in any one area in the process. As Daniel says, it’s much harder to design for the mass market than than something bespoke – at the same time, the challenge of such limitations can push designers to do their best work.
Or produce the fifth gen Escort.
I think Tom V has grasper Pye´s point. On some levels I think the Aston is much more of a failure than a Ford Mondeo Mk 2, for example. Yes it´s really fast and also pretty useless. The Mondeo will get from A-B in pretty much the same time as the Aston and do a lot more besides. As a design object the Mondeo, or any mass-market car is much more interesting than a car that costs the same as a house and will perform best on a few kilometres of a race track.r
Thanks, Richard. I think it really is a matter of which compromises or failures you are prepared to accept. Or able to: if you have the disposable income for the Aston, you can probably afford other, expensive means of solving the problems the Aston cannot solve, but the Ford can. The prospective Ford buyer can only afford the one solution. These things are, of course, all relative, as is the appreciation one has for them. Frankly, I admire both the Up and the DB9, albeit for different reasons.
A car like the Alfasud, which on the one hand tries to make as many compromises as it can, but on the other still emphasises a few qualities above others, sits in a particular (emotional) sweet spot for me. It’s well known quality issues are as much of more a result of the circumstances in which it was built as the design, as far as I understand. Which is a good a reason as any to post a picture of it: