User centered design doesn’t work for innovation

written by Javier on 4/02/2010

Ariel Guersenzvaig, who knows me well and understands my take on user-centered design, refered me to Apple’s Secret? It Tells Us What We Should Love, an article that questions UCD as a tool for radical innovation. I’ve been moving from true believer in UCD to these positions over time and, although I think it’s easy to use Apple as an example, I consider this article by Roberto Verganti full of true statements:

User-centered innovation is perfect to drive incremental innovation, but hardly generates breakthroughs. In fact, it does not question existing needs, but rather reinforces them, thanks to its powerful methods.

Firms that create radical innovations make proposals. They put forward a vision. In doing that, of course, they take greater risks.

Thanks to this process these companies are serial radical innovators. Their non-user-centered proposals are not dreams without a foundation. Sometimes they fail. But when they work, people love them even more than products that have been developed by scrutinizing their needs.

User centered-design (observation, interviews, user testing, etc. ) is for those who want to improve something existing, not for those who want to create something new. Those need to understand human nature but don’t need to microscope every little behavior and take it as a starting point.

There are 11 comments to this article:

  1. 4/02/2010Ariel Guers says:

    I knew I was spot on! ;)

    Donald Norman is surely one of the major apostates:

    “Once a product direction has been established, research with customers can enhance and improve it. Beforehand? Leave it to the technologists.”

    http://jnd.org/dn.mss/technology_first_needs_last.html

  2. 4/02/2010Javier says:

    Well, was that before or after he was “steved” from Apple?

  3. 4/02/2010Ariel Guers says:

    The quote comes from a very recent article, the dude abides.

  4. 5/02/2010UCD, UCI y la creciente locura del design thinking « Interacciones.org says:

    [...] Javier Cañada publicaba su elocuente opinión al respecto. En el post Javier cita a Roberto Verganti que en el blog del HBR escribre: Insights do not move [...]

  5. 5/02/2010Humberto says:

    This is a very interesting debate that has been around for quite sometime and that Donald Norman made popular with his article a few weeks ago.
    I’ve been struggling with it myself for the last 3 years without a very clear conclusion.
    While i see some good points on the position that Javi states here and in Norman´s article, there are also some very important things that haven’t been taken into consideration.

    First of all is the concepts of disruptive and incremental. Are those valid categories? My experience is that no one starts something know deciding if they want to be disruptive or incremental. Most of the times people try to solve problems or to improve the way things are done currently and then they come up with ideas that might end up breaking a market or simply making things slightly better. But that’s something you don’t usually know at the beginning, very small changes in a product can create huge changes in the market or in the social behaviour.
    An example i use in my classes to reflect on this is the plastic syringe. This was an spanish “invention” by Manuel Jalon, the same person that patented the “fregona”. While creating the spanish mop what he did was find a great way of injecting plastic into molds. Once we had that he started thinking what other uses could he give to this new method and he conducted a huge international research, that included user research (altohugh not called like that) and he found out that the glass syringe had a problem he could solve. Glass syringe were imperfect (not two were exactly identical) therefore nurses couldn’t determine the response of the pump to their pressure, sometimes was too hard sometimes was too soft. By making them with his new industrial process he could make every single syringe identical solving the problem the nurses had using it. He did so and he became a very rich man (until he opened a factory in Rusia, but that’s another story).
    When i show a picture of a plastic syringe and a glass syringe all the students claims it is an incremental innovation, after all he only changed the material. But, if you follow Christensen definition, this is a disruptive innovation because it wiped out all the competitors and created a whole new category.

    The second point is about the goal you are after when including users in an innovation process. Simona Masschi, former Ivrea and now the head of CIID, teaches that when working with users in an innovation process, you should look for inspiration more than understanding. If you are looking for something to trigger the new idea observing and talking to people is a great source of inspiration. I’ve seen things they’ve done for hospitals in Denmark after an ethnography and they are quite amazing, they were challenged to reduce the recovery time after programmed surgery working only on non medical aspects of the hospital experience and they were successful, they’ve also used it for Philips with very good results. (Innovation is not always about coming up with the next ipod)

    Lead user innovation, as Von Hippel understands it, is also a way of user centered innovation. Von Hippel claims that most of the innovations come from users who have the passion, the ability and the means to improve something already existing in the market (this would be totally incremental but has open huge new markets as the Kite surfing or the mountain bike).

    I guess where Donald Norman is completely right is in implying that design doesn´t equal innovation and therefore, user centered design might not be the best discipline to find new ideas. But i would say that user centered innovation, using many of the design tools with different purposes, not only exists but it has a lot of success stories in the market.

    But, as i said at the beginning, this is far from being a conclusion. There are some days that i totally agree with you. :)

  6. 5/02/2010Humberto says:

    I wish i could edit, i can see a lot of misspelling.

  7. 5/02/2010Ariel Guers says:

    Many designers, and this is evident, manage to know what people need without having to resort to formal observation or participatory techniques, this way of knowing does not guarantee success though. Most designers fail miserably and deliver poor results. Others (great designers) excel at delivering great products time after time and without the help of user research.

    To me, the UCI/UCD methods are just a way of informing our intuition to help us build better hypotheses about whether something would work or not. Needless to say that these methods are not a guarantee of success either.

  8. 6/02/2010Tzek says:

    Interesante. A mis alumnos siempre les planteo el dilema de diseñar para lo posible o para lo probable. En varias ocasiones hemos concluido sobre la innovación y cuál carácter debe tomar tu proceso de diseño.

    De repente, me vino a la mente ese sentimiento de reflexionar y discutir sobre el diseño vs. las ingenierías y ciencias. Sentí algo así sobre el UCD visto ahora como un ente rígido y que la innovación se da en forma más libre.

    Es claro que la innovación requiere un acercamiento al problema de forma no convencional. Aunque creo que UCD está contenido dentro de un proceso de innovación, de cualquier forma es “input”.

    Saludos.

  9. 6/02/2010César Astudillo says:

    I’ve been drawn to this interesting discussion from dnx’s internal mailing list, and some coworkers (not Humberto) have asked me to come into it. So if I’m being too verbose, at least I’ve got an honorable excuse :-)

    Methods, processes, models, paradigms… when it comes to design, they’re useful both for helping seasoned designers explain what they do, and for providing all designers with action schemes, heuristics and thinking frameworks on which to act… or not. Depending on the kind of problem to solve, constraints, current knowledge about these constraints, and the designer’s skills and preferences.

    Design is not physics. It hasn’t laws, but a bunch of quite fuzzy principles that may apply or not, depending on the situation. Design is not Six Sigma manufacturing. It hasn’t a procedure, but a handful of possible processes you might or might not follow, and even if you choose a process, you might well decide it’s for the best to skip a stage or twho, or do some extra task. Design is not chess. There is no such thing as a set of rules that objectively define when the game is over, who has won and who has lost, or whether a move is “legal” or not, but a series of heuristics that, when followed, don’t guarantee success, and when ignored or broken, don’t guarantee failure. This is the inconvenient truth about design theory.

    But but but, knowledge about models, processes, paradigms… is a tool that can amplify the design force you are capable of exerting, and sometimes you’d be a fool not to know and use them. Just when you feel they help. And hunches and creativity and self-expression and freedom and empathy and brilliance are, well, just not enough. This is the inconvenient truth about design practice.

    So when I hear a discussion over whether a specific process, model, paradigm… “works” or not to solve a particular genre of design problem, I find it misleading, because it might encourage some listeners to think of design as if it was physics, Six Sigma manufacturing, or chess. Not only design principles are always fuzzy, principles about which principles work and which don’t is even fuzzier. Which is not surprising if you think about it :-)

    Myself (and I suspect many of the very competent, experienced designers who write in this blog and read it), I like to find out as much as possible about different processes, paradigms, and the like (many of them from outside the field of design, by the way); I let them ferment in my head, and then I like to decide which pieces of them I’m going to use for each particular design problem.

    That said, about UCD: I, too, think the design community is mature enough to begin to question many of its dogmas (I never considered them dogmas in the first place). But I believe, if we can afford questioning the utility of UCD for innovation, it is only because we’ve learnt enough about it to begin to understand its limitations. I do still think the principles of UCD should be present in every designer’s toolbox (from where they can be picked up or not). And, honestly, I can’t think of a single design project I’ve done where it was counterproductive to do some user research. Maybe it’s true that “User-Centered Design” in its most strict interpretation does prevent you from thinking outside the box. Maybe we should do something we might well call “User-Aware Design” instead. Or maybe one should start by making some “User-Deliberately-Agnostic Design” first, and then letting to users to decide which design they condone with adoption and appropiation. It depends. Overall, I think of UCD a little as I think of Christianism. The fact I’m an atheist today does not stop me from recognizing that some Christian values have shaped my worldview and my belief system in very positive ways. Even as I consider Immaculate Conception as an obsolete construct.

    What I do believe is that, historically, really disruptive innovations are not explainable from the perspective of this or other method or process followed by an individual innovator, inventor, or designer. They are more explainable from the perspective of the whole market/society following a particular emergent, collective process: evolution. Evolution, much in the Darwinian sense, as the sum of two mechanisms: a mechanism that generates diversity, and a mechanism that generates selection. The mechanism that generates diversity is diverse in itself: we humans, at the scale of population, are quite proficient at generating diversity. Scientific curiosity, serendipity, greed, fear, war, myth, famine, drought, lust, injustice… everything that is human can generate diversity at the scale of population. And then, the mechanism that generates selection… you might well call it “supply and demand”, or “culture”, or whatever. One way or the other, ultimately, some thinks work, and some don’t. And selection mechanisms are sociohistorically situated: what doesn’t work today may well work tomorrow. When it comes to innovation, I believe evolution theory is more consistent with the facts than “intelligent design” by a single god-like designer (I hope it is still safe to say this when there are Steve-Jobs-worshipping zealots in the room).

    I honestly don’t know of a single good method to generate innovation in the individual scale of the designer or the design studio. But I suspect it must have these two elements embedded, it must involve generating a micro-evolutionary environment at home. If you want to innovate, first of all you must be able to generate an exhuberant, overwhelming shower of diversity, as fast and cheaply as possible. And then, you must be able to exert some selection inside the four walls of your studio, if only because you don’t have infinite resources to push every silly idea to the market. What role can user knowledge play in the process? I’m sure it can help not only in the selection mechanism (something about I’m pretty sure very few would disagree), but also in the diversity mechanism (this one is not so easy to agree upon). Yes, I believe knowledge about people can help the designer generate diversity… as long as we throw away some obsolete, limiting constructs such as, for example, “needs”, and start using knowledge about people in less restrictive ways.

    (sorry for the speech)

  10. 6/02/2010Ariel Guers says:

    Very interesting outlook, César.
    While I agree with you in several points, I profoundly disagree on evolution being an explanatory frame for innovation. Innovation as well as design have always an extent of intent (I’m not talking about succeeding or failing). The “mechanism that generates diversity” is in this case not random, or at least not always random.
    On the contrary, evolution is by its very nature always random. We are here in this planet by chance, not by necessity. Maybe after all you are more christian than you think ;) (I’m only kidding)

  11. 6/02/2010César Astudillo says:

    You’re right, Ariel. Of course, the diversity mechanism in innovation is not random. There are one or two examples of innovations driven by chance or serendipity, but most of the times, the innovator knew reasonably well what she was doing and/or why. Innovation in human-made things, and innovation in the gene pool, have a number of differences, and the one you’ve pointed out is not the only one. But I keep on finding the resemblances interesting.

    What I like about comparing innovation to evolution is that this resemblance reinforces one belief I’m quite fond of: the belief that design problems are not single-solution problems, and that the linear vision of innovation as unidimensional progress along a single, technology-driven path is terribly limited. It is the nature of things that the probability of any event that has already happened is always 1, no matter how unlikely the event was. One of the properties good design often has is to seem inevitable. But this is just because of our natural tendency to give things theleological explanations. And that may be misleading. This is why trying to learn about innovation by examining success stories is a little bit (not entirely, just a little bit) like trying to understand how lottery works just by interviewing lottery winners.
    When I was learning about my infertility, the doctor told me that, even after how difficult is to get an egg fertilized, the chance of a fertilized egg to end up causing a detectable pregnancy in a healthy, fertile woman is just ten percent. “Gosh!” –i said. “Given that, how do so many people get to have children?”. “Heh, because we find it funny to try it a lot” — was the answer. I guess, ultimately, that “trying it a lot” is definitely the single ingredient that I would always put in a recipe for innovation. To have an intent (and, hopefully, some skills) affords us to churn the diversity machine not as many times as evolution does by pure chance. But we have to churn anyway. If every innovation worked out at the first try, innovation would be a commodity.

Write your comment: