User Centered Design is Dead
Midday on Sunday I stopped in to listen as Jared Spool led us on a Journey to the Center of Design. Okay, I admit it--I didn't plan on going. Truth is I was too lazy to walk a mile across the convention center to another panel. Well, not only did I save my soles, I got the better end of the bargain. I'll go ahead and say that this was the best presentation I've attended thus far. I Twittered my way through the panel, but I'll provide you a summary here.
1. There's no evidence that UCD has ever worked. We've seen marginal success, but nothing overwhelming.
2. Apple has all but closed their user experience division. Microsoft runs over 15,000 usability tests per year. Who has the better experience? Yeah, interpret this one for yourself.
3. Process is how we do things. Methodology is a repeatable process. Dogma is something we believe without any logical reason.
4. Every successful organization that Spool has studied doesn't have an official corporate methodology for UCD. Except One. That one has a VP whose job it is to release people from the official methodology.
5. Process includes techniques and tricks. Tricks are things we do when the preferred technique is too lengthy or expensive.
6. UCD is a dogma and it's time we change it.
7. The placebo effect works 60% of the time. UCD work is placebic.
8. It goes without saying that the point of user research is to inform design.
9. Most usability tests end the same way. Developer says "If we'd known that two years ago, we would have done this differently."
10. Analytics can be interpreted in conflicting ways. Eye tracking is voodoo.
There are three core user experience attributes that make a difference: Vision, Feedback and Culture.
Vision.
Can everyone on the team articulate the experience of using your design in five years?
Feedback.
In the last six weeks have you spent more than two hours watching someone use your design or your competitors designs?
Culture.
In the last six weeks have you rewarded a team member for creating a major design failure? This indicates a culture that learns from failure.
The UCD dogma is dead. Informed design is here. Let's go.
1. There's no evidence that UCD has ever worked. We've seen marginal success, but nothing overwhelming.
2. Apple has all but closed their user experience division. Microsoft runs over 15,000 usability tests per year. Who has the better experience? Yeah, interpret this one for yourself.
3. Process is how we do things. Methodology is a repeatable process. Dogma is something we believe without any logical reason.
4. Every successful organization that Spool has studied doesn't have an official corporate methodology for UCD. Except One. That one has a VP whose job it is to release people from the official methodology.
5. Process includes techniques and tricks. Tricks are things we do when the preferred technique is too lengthy or expensive.
6. UCD is a dogma and it's time we change it.
7. The placebo effect works 60% of the time. UCD work is placebic.
8. It goes without saying that the point of user research is to inform design.
9. Most usability tests end the same way. Developer says "If we'd known that two years ago, we would have done this differently."
10. Analytics can be interpreted in conflicting ways. Eye tracking is voodoo.
There are three core user experience attributes that make a difference: Vision, Feedback and Culture.
Vision.
Can everyone on the team articulate the experience of using your design in five years?
Feedback.
In the last six weeks have you spent more than two hours watching someone use your design or your competitors designs?
Culture.
In the last six weeks have you rewarded a team member for creating a major design failure? This indicates a culture that learns from failure.
The UCD dogma is dead. Informed design is here. Let's go.
All Articles
© 2010 Capstrat
Check out Spool's talk on Slideshare. He does a much better job of explaining it all. http://www.slideshare.net/jmspool/journey-to-the-center-of-design
The point of the Microsoft/Apple example is that UCD isn't proven, but you're exactly right. Apple and 37 Signals understand their users well and they are using those insights to inform their design.
Also, Spool wasn't saying that user research was placebic, he was challenging the notion of UCD as a dogma that we practice. We should move to a process where design is informed by multiple inputs including but not limited to the users.
I'm a huge fan of Jared Spool and agree with many of his points. Personally, I'm a bigger fan of activity-centered design because it's often less fluffy that UCD.
I have one issue with this however. Apple has a small ecosystem of products, designed for people ostensibly like them. Basecamp is much of the same--a product designed for people who are quite like the designers. Microsoft, on the other hand, has a huge set of products for a highly varied set of users. They invest in user research because they feel they benefit from the outside input.
For me the question is "Do the designers know the customers well enough to make informed conclusions without their input? Sometimes you do, sometimes you don't.
Finally, if "informed design" is here to stay, what is it informed by? If user research is placebic, eye tracking is voodoo, analytics are interpretable, and so on, do we just run on gut instinct and the client's word?
Post a comment
We look forward to hearing what you have to say. Before joining the conversation, please take a moment to review our comment policy.