Designing the User Experience: The Perils of Skipping User Tests

It’s “Bike to Work Week” here in Vancouver. A local cycling organization (the Vancouver Area Cycling Coalition, or VACC) has once again added a “trip log” to their website, so cyclists can form/join teams and record the miles they ride to work.

It’s such a wonderful idea, I feel bad about criticizing their implementation. But since the organizers actually had to send out an email explaining how to use their “new and improved” website, it’s pretty clear they’ve been getting complaints!

From a quick look around, several problems are obvious:

Signing up is quite a process, requiring your full name, phone number and mailing address, plus double entry of both password and email address:

Problem: It’s a lot of work, and people may well wonder why they’re being asked all this stuff. If there’s a good reason – for example, so you can mail me a prize – then tell me!

Specific password formatting requirements are imposed, including a very narrow 6-8 character min/max length:

Problem: Many people have an easy-to-remember “default” password they use whenever possible. If you insist on specific formatting, users will have to invent a new password… and it’ll be harder to remember. This isn’t a banking application, folks. Why all the fuss over security?

After filling out the lengthy form, you have to notice and click the little “Bike to Work” box near the bottom of the page:

Problems: It’s hidden amongst other items like the membership signup. And it’s not pre-checked.

You must use their clickable map to compute your mileage – even if you already know how far your ride is:

Problems: First, it’s not self-evident how to use the map. Second, it doesn’t work in some browsers.

Overall, right from the Home page, the entire “Bike to Work” signup process is downplayed too much. It’s relegated to inline links that appear in random, inconsistent locations. The critical function of logging trips doesn’t even appear in the navigation until you’re all set up and logged in. (Problems: People can’t find it… and the organizers will have to send out an email explaining how to use their site!)

All these problems are so obvious, I’ll bet my bottom dollar the developers didn’t conduct any user tests (except perhaps with people already familiar with the process). Had they held even a handful of informal user tests – which would have cost nothing and not taken very long – they’d have uncovered these obvious sources of confusion.

Bike to Work week is VAAC’s biggest function of the year, their main opportunity to raise their profile and recruit new members. It’s a shame the sign-up process is so clumsy.