Get new posts delivered straight to your inbox.
Enter your email address
Subscriber count: 2,561
Aug 1, 2007 •
I once read an essay by Joseph Epstein in which he describes the glee and suspense in watching a stranger read his essay. It has happened to many novelists too -- the coincidental (dreamed-about) situation where you encounter a stranger reading page after page of your book, without realizing you're present.
I had the opportunity to do somewhat the same today: to watch someone try to follow instructions in a guide I was writing. I learned more from the 20 minutes of watching than I would have learned checking over the guide and testing the instructions for 3 days myself.
Areas I thought were simple turned out to be a little confusing. The person hardly read the guide at all; instead, she glanced at it quickly while focusing most of her attention on the screen. When she did read, she read sloppily, in haste, sometimes reading the wrong section (she was often out of place in the sequence of steps). One time she lost track of where she was, and didn't even look at the image where the correct area was circled.
Now granted, she knew I was watching. In fact, I asked if I could observe her following the guide. For the most part, she did follow the steps correctly. I was just surprised by the haste, the glancing and scanning instead of reading.
How-to guides are not pleasure-reading documents. Of course I know that. It's not like creative writing, where you hope the reader savors the sentences, carefully and thoughtfully processes your ideas and smiles at your wit.
I learned several things from this twenty-minute observation:
Of course users and situations vary. And the fact that I was present, available for explanations should the instructions or application prove too confusing, throws askew any scientific observations about user behavior. Still, watching a user try to follow the instructions I wrote was probably the single most effective way of evaluating the clarity, helpfulness, and worth of my instructions than any other method I've used.
It doesn't take more than 20 minutes to an hour to get more feedback than you will ever have time to implement. I am now making this user-observation a standard -- even if I have to bribe someone with chocolate or proofreading favors to do it.