I'm almost positive that was a song on Rush's "Different Stages". Running prototype usability testing is about watching what users do when you ask them to attempt realistic tasks using a prototype. It's not brain surgery. Here's the best advice I can offer on how to run usability testing.
(1) Agree on the "things" we want users to do. Do we want them to: buy books? Enroll in a loyalty program? Switch to paperless billing? In an agency situation, the "things" we're interested in save money or make money for our clients. Simple right?
(2) Realize we can't test everything. This is where things get ugly. We all want to ensure the entire site is of a certain quality. The cold reality is that we usually just don't have time or it's simply too expensive to test it all. So we get ruthless. What does the site absolutely need to do well? What's less important? My experience tells me to make my recommendations and make sure I can back them up. The rest is up to the client.
(3) Test. Everyone loves watching usability testing. You get to sit in a dark room, munch on local pastries, and genuinely feel like you understand a little more about how people use websites at the end of the day. What's not to like?
(4) Revise. This is where I can relax. After compiling and sharing the results of usability testing (although I like to invite the design team on the actual day so they can see themselves), the results speak volumes. If users complained about B2B jargon, we have to revise the copy. If users shook their fists at the mouse-type on the prototype, we have our marching orders to make the text larger. Anyone who disputes what five to seven perfect strangers unanimously say about a site's design is in it for the money, and that my friend, is wack.
