Getting your users involved in requirements gathering

— Published December 18th, 2007  |  by David Moore

As some of the concepts of user-centred design gain more currency, it’s heartening to see more organisations building in some user testing as part of the development process for their website or web application.

No last-minute testing

However, we often see the user involvement starting way too late in the process. Getting a few users to test a site that’s due to launch in a couple of weeks can be almost useless.

Either it’s meant as a rubber stamp to approve the existing design, or sometimes it can be used to solve an argument over something that doesn’t matter much anyway (‘We can’t decide if the button should be blue or red. Let’s do some user testing’.)

Of course, if the testing does reveal some serious problems, it might be too late to fix them, and you’ve all spent a lot of time and money barking up the wrong tree.

Getting users involved in requirements gathering

The best time to get users involved is as early as possible. Focus groups can help establish strategic goals – they’re good for getting lots of input on quite broad topics, so if it’s a new online business project you’re launching, a focus group will show if real people understand the idea, and offer pointers to your overall direction.

But it’s at the requirements gathering stage that one-on-one interviews really pay dividends. If you’ve identified your key audience groups, try and interview people in those groups.

You’re looking for two main things – the first is to get a sense of their background: education level, experience online, time or technology constraints, how much they know about you and the sort of thing you do, and how much they trust you, for example.

The second type of information is what their goals would be in using your site – what they would like to get done, or what questions would bring them to the site.

Goals not tasks

You don’t want them to outline how they think they’d achieve these goals, just what they are. It’s your job to come up with the simplest and most elegant ways to allow people to complete their visit satisfactorily, and often the best solution will be something the user hadn’t actually thought of.

For example, a visitor’s goal might be to visit an online bookshop to buy a new thriller. They might expect to do that by clicking through to the Thrillers section and browsing the list of best-selling books in that category, hoping to find something they might like.

But if they’re regular shoppers on the site, you might be able to look at their past purchases (and those of thousands of other people) and make suggestions on the homepage, directly related to their preferences. They see a book they like, and order it – same goal accomplished, but by completely different tasks.

How many people to interview?

The more the merrier, but if you’re short on time or money, you’ll get some valuable information from just talking to one person. And if you can get a few people involved, don’t worry if they don’t exactly fit your audience group. As Steve Krug says, ‘recruit loosely, and grade on a curve’.

The value of face time

Even if you have tons of statistical data about your audiences, conducting interviews can give you a real insight into the content and functionality your site needs to succeed.

And of course, the earlier you do them, the easier it is to incorporate the findings into your development plan.