Whenever I spend time reviewing the messaging of one of our provider clients, they often ask about how to review their Web site. While lots of methodologies and approaches exist, my own preference is to use a Scenario based approach.
Why? It puts everything into the context of a visitor/buyer goal.
You can ask anyone if they like your content, your layout, your navigation, or even your “Web Information Architecture” (doesn’t that sound fancy), but it is all irrelevant if your visitor can not accomplish their goal when visiting your site.
Your site is not about you. Its about what your visitor wants to accomplish and how you help them get there. Now, if that visitor has no interest or value to your business, then you don’t really need to care about helping them accomplish their goals. But for prospects and customers (or other key audiences), you had better care.
One simple, affordable way to do this is using a service called UserTesting.com (Disclaimer: I do not, in my current Gartner role, evaluate Web usability testing tools or vendors. I happen to know UserTesting.com because of past experiences. An unvetted (by Gartner) list of other testing tools can be found here. Feel free to check some of them out. If you’ve used another service or approach and gotten value, I encourage you to share the info in the comments. But, please don’t promote your own service.)
Here is what I like about them:
- Their approach is based upon defining a scenario that you want to test. This could be “Please look for information on our content management product and show me what you would look for if you were in the early stages of evaluation.” or “Please try to find three Gartner research reports by Hank Barnes on messaging and try to buy them.”
- You then define who you would like to conduct the test. You can have one test (one person) or multiple. You provide the demographics for the type of people you want. Those people have NO vested interest in saying nice things (like your agency might or even internal staff who might speak no evil, see no evil, hear no evil).
They are just people like your visitors–they may know alot or nothing about you.
- The test is conducted almost immediately, by folks from a testing pool that fit the profile you defined. You get a recorded video (with synced audio) within 1 hour of the tester’s experience trying to complete the scenario. You see where their mouse goes while hearing them tell you why they tried what they did. You also get the tester’s responses to a short set of followup questions that you define.
- The cost per test is very low (pricing changes sometimes but it is currently $49 per test on their site).
The testing pool is really interesting. It is crowdsourced. UserTesting recruits testers who have to “pass a test” to prove they can talk and mouse at the same time. Once that happens, they are in the pool. They get paid a portion of the fee for every test they conduct. They get ratings from the test requestors. The ratings are used to continually refine the pool. (Additional revelation: I was an early member of the testing pool. I did it for fun in the evenings when I had an incredibly boring un-fulfilling job. My ratings were pretty high and I enjoyed the experience.)
With services like this (and the others you can find above), there is no excuse not to do some usability testing for your Web site. But before you do, start thinking about scenarios. What do visitors, that matter to you, want to accomplish? What would you like to guide them to accomplish? Thinking about adding some addition gating to your content (to get more registrations), test it and see what the reaction is.
The ability to think in scenarios starts to shift your whole approach to being much more buyer/visitor-centric. You’ll be surprised at the little things on your site, that may have been there for years, that are getting in the way of success. You’ll also find that it helps you think about other things from the buyer perspective v. your own.
And usability testing closes the loop by actually letting you see and hear what buyers really want and expect. I strongly believe no new Web site (or major change) should be launched without it.