Today I learned (relearned unfortunately) the value of digging into a design effort with nothing more than an engaged end user and a blank piece of paper. I’ve been working on a product management reporting dashboard the last few weeks and found myself yesterday morning sitting with a detailed set of requirements and a looming deadline. Not always what you want to see at 8 AM but I thought I could rise to the occasion with a bit hard work and a phone on mute.
I plowed straight into Visio designing what I thought were going to be a useful set of screens detailing the various performance and usage metrics of a new site I’m working on. After a good 15 hours invested in the effort over two days, I looked at my work and had the sinking feeling I had wasted two days designing an absolutely useless set of convoluted screens on par with a federal tax form. So I spent a few more hours rearranging the proverbial Titanic deck chairs before seeing that I was working too hard but not too smart. Not my best days.
So my next move was to scrap the whole effort and start from scratch. I grabbed a few end users and asked them a series of questions about how they wanted to consume the information and how they would be using it. The key point I took from our talk was not so much details of the exact data points as much as their need to consume the information quickly. They was looking for an easy-to-digest set of information that they could process in a few minutes and determine if there was an emerging problem that needed immediate attention. We quickly sketched out some very crude designs and within 30 minutes had a whole new direction. The project is now back on track and some of the users have given preliminary approval to much of what they’ve seen.
This is hardly a new insight but one we sometimes lose sight of on projects with tight deadlines because we are too focused on making progress on the actual deliverables. There are some good resources at A List Apart, UseIt.com, and at the website for Carolyn Snyder’s book Paper Prototyping.