The IDEO shopping cart. Users didn't know they wanted this till they saw it. Most “users” don’t know what they want until you show ...
|The IDEO shopping cart.|
Users didn't know they wanted this till they saw it.
No one really knew they wanted an iPhone till Steve Jobs showed us an iPhone. Ditto for the Boeing 747. And likewise for myriad other products out there. I was reminded of this recently by this article, "What do you want your phone to do that it can’t?” Among things it claims users want: "they want thicker phones with larger batteries; better cameras with actual zooms; huge screens with no bezels; wireless charging that picks up signal from anywhere; speakers that don't sound like a tin can mess; and dozens of other things that probably shouldn't be shoehorned into a tiny computer that fits in your pocket. They want flexible phones for some reason.”
No evidence is provided for any of this except, of course, that all these functions and features have been showcased in one phone or another in recent times. Even better: read the comments. They all (as far as I can tell as of this writing) refer to features that have been talked about, advertised, or even implemented already.
It’s not just the case with phones either. I can’t think of a single case of a user ever having declared a want (or even a need) that hadn’t already been mentioned somewhere, if not implemented to one degree or another. (Of course, I exclude from the group of “users” any experts in design, requirements elicitation, want/need analysis, advanced systems modelling, etc. These experts constitute a very small fraction of the general population.)
This is not to denigrate users. Asking a generic member of the public what they want in a phone is in many ways like asking a heart surgeon to discuss which research avenues the physicists at the Large Hadron Collider should pursue next; there’s just no way for the surgeon to have a cogent answer, unless they’ve serendipitously read or heard something relating to the LHC. When designers ask users what they want, they’re unfairly expecting those users to know as much as they themselves know about problem analysis.
So, I’m forced to ask: why do we bother with user wants at all? It’s a serious question. I think it’s all a question of marketing; we need to keep the general public convinced that we (designers) are actually working in their best interests, and it would seem quite insulting for us to tell them what they want. Maybe I’m wrong, but it’s a theory that explains the facts and I’ll stick to it till I find (or am informed) of something better.
Of course, that hasn’t stopped companies like Apple and Google from achieving global success by doing just that.
Indeed, I would submit that Apple and Google (and IDEO and many other companies who use similar approaches) are doing it right. What they’re doing is observing users and intuiting from their studies where they can insert a new function or product that will facilitate whatever goals the users are pursuing. If it sounds a bit like anthropology, that’s cuz it is: it’s design anthropology.
This kind of study is particularly useful because it can be used to reveal what a user actually wants, filtering out the influence of the exposure that the user has had to what others have told them they want. Sometimes, this can lead to surprising and even problematic results. A classic and well-documented case is IDEO’s redesign of the shopping cart. In that case, IDEO researchers interviewed and studied supermarket shoppers and employees to discover (a) how they shopped, and (b) what they thought was wrong with existing shopping carts. From that study, they came up with quite a novel design that was both very surprising to everyone but also appealed to most stakeholders.
I think item (b) is particularly important: while users might not know what they (really) want, they are literally experts in what’s wrong with what they’ve already got.
This connects back to Herbert Simon’s definition of designing: "To design is to devise courses of action aimed at changing existing situations into preferred ones.” The users are experts in the shortcomings of “existing situations” by virtue of being immersed in them. They carry information that designers lack about those situations, but that designers need.
Designers, on the other hand, are the experts in finding ways to achieve preferred situations.
For instance, say a survey found that drivers uniformly complain about the high price of gasoline (not an unreasonable hypothetical situation, I think). The immediate (and naive) inference here is that they want gas to cost less. A closer study of the actual situations in which these drivers find themselves could (and usually does) reveal a variety of other factors that all contribute to that situation. Some of these could be:
- They live too far from where they work.
- They spend too much time in heavy traffic.
- They drive cars that consume too much gas.
- They don’t use navigation apps that could find them more fuel-efficient routes.
- They have fallen prey to Jevons Paradox.
Clearly, the list of potential solutions will depend on actual reason for the user’s expressed want. Get the actual reason wrong, and you’re very likely to choose the wrong solution.
One other thing: notice how interrelated many of the factors are, to each other and to other factors like lifestyle choices, size of family, income, cultural preferences, environmental concerns, availability of public transit, and so on. What we’ve really got here is a large and complex systems problem that’s nowhere near as simple as “gas costs too much.” Even design anthropology isn’t yet up to the challenge of modelling this system sufficiently well to give designers the understanding they really need to find truly preferred situations.
We really need more research on how to build system models of actual situations such that they can inform design decisions.
And we really need to stop focusing on “user wants."