Critical Path Newsletter

But I AM the User!

One of the truisms of user experience design is that if you want to design something usable, you need to know how the user thinks and behaves. 

So, let’s say you’re creating a very technical product, like a network management system or an IDE, where the user is very technical, just like you. Or let’s say you are the product manager for a healthcare product, and it so happens you are a doctor by trade, so the user is just like you.  You should have a pretty good idea of how to design the product, right? In fact, it should be a piece of cake to design a usable product.

Wrong.

If you want to design for you, or someone like you, you need to know how you behave. The truth is, we suck at observing ourselves. How you say you behave, and how you actually behave, are quite different.  

Let me give you an example.

Suppose I were to give you a trivia quiz of 100 questions – things like “What is the capital of Arkansas?” and “Who won the 1965 World Series?” A week passes, and we meet so I can share your results. Before telling you your score, I give you the answers. Then I ask you how you think you did. You will probably say “I think I did well! I would guess around 90%”.  In fact you only scored 60% (you need to brush up on your trivia if you’re ever going to be invited on Jeopardy). If I were to repeat the experiment, but this time I didn’t give you the answers, your guess would be much more accurate. Knowing the answers biases you to drastically overestimate your score. 

Now let me give you more practical example. 

Suppose I were to ask you whether you were successful shopping for a book on an ecommerce site. The first ten times you purchased a book were successful and the experience delightful. The last two times, however, were miserable – the site was slow, or you had to re-enter your credit card information. You would likely tell me, with conviction, that the experience was rotten. You will put significantly more weight on the last few negative experiences, even if the positive experiences were much more numerous.  Psychologists call this the recency effect.

It turns out our brains don’t work like computer memory – faithfully and accurately filing away data for later retrieval; they are more like backfilling devices, filling in the blanks with assumptions based on experiences.  And by the way, it’s our logical left-brain that is responsible for making up stories to backfill our memory. 

Simply put, we’re terrible at observing our own behavior.  This imperfection is why we need to do user research and usability testing. A good user researcher observes real behavior and helps us make better design decisions. Many usability problems are obvious only after we see it in a usability test. Here’s a common example. You develop a web form that asks for a telephone number, and will only accept it in a certain format. You put an example beside the field showing the format you are asking for (e.g. 123-456-7890). You even choose what you think is a common format. I can guarantee you that in a usability test, people will type dots between the numbers, use no dashes, parantheses, or even add a country code if they are outside the US – anything to sabotage your wonderfully logical form. To you, it can’t be more obvious. I even hear you yelling through the soundproof, one-way glass at the user “it’s written right beside the field you idiot!”

Occasionally, even the best designer will incorrectly guess user behavior, even if they are designing for someone remarkably like themselves. If you want to create a truly usable product, there’s simply no substitute for user research.