It is difficult to see the picture when you are inside the frame. (Wendell Castle)

It’s really hard to believe I’m fighting this battle again. 

One of the principles of Design Thinking, though, seems to involve designers testing their own stuff. They are encouraged to do this in full-on guerilla mode – grabbing people in coffee shops and thrusting screen shots in front of them.

Now, I will admit that their heart is at least in the right place. They are getting some feedback. It's also good for the designers to interact with their actual users. And, finally, I’m all for informal methods.

This is not, however, usability testing. First, testing cannot be done just by anyone. There are some real skills involved, skills that need to be seen, learned, practiced, and critiqued. It is very easy to bias your user, to ask the wrong questions, to intervene at the wrong point, to generally muck things up. Heck, I’ve seen my share of “professional” usability engineers who couldn’t get it right.

I’m also a firm believer that it takes a certain kind of personality as well. I’ve mentored a lot of folks over the years who have expressed an interest in becoming a usability engineer. Some people were total naturals. Others (often because they were too extroverted, or impatient, or unfocused) struggled and struggled.

Second, humans are simply not very good at objectivity. There are dozens of proven cognitive biases that show this – the IKEA effect, confirmation bias, the backfire effect, belief persistence, denial … And based on the resistance I sometimes get from mere observers, I can’t imagine that having those folks leading the tests instead of me could possibly do anything other than make things worse. 

One of the stories I like to tell in this regard involves a favorite designer of mine who was asked to play the “computer” (i.e., shuffle the papers) in a test of a paper prototype. His body language was comical. At one point, he actually spun around in his chair when the user, who had been struggling mightily with a task, finally got it right. A brilliant designer who has since gone on to a very successful career, I might add. I wouldn’t ask him to test his own stuff though. Heck, it would probably never occur to him to do so anyway. 

Tell you what … Give the usability engineers the tests and we’ll let you guys do some of the upfront research. It’s a lot harder to muck that up, plus it can really help you get to know and build empathy with your user. Just stay away from your designs though. And don’t be afraid to ask us for advice.

I have to give this guy credit, as Wendell Castle was actually a designer himself (and of art furniture, no less)