This week, the government began emailing users that initially failed to signup for an account on Healthcare.gov to come back and try again. First impressions being what they are, it remains to be seen how many people will come back. But assuming they do, and the performance and service issues are dealt with, several issues with the user experience still remain.
Lots of ways to do a few things
On the homepage, there are really only two primary use cases: learning about the healthcare plans you're eligible for and signing up for one. The very top of the screen accomplishes this making the rest of the page seemingly redundant
After the primary navigation, the secondary navigation seems to bifurcate out into role-based delineations that do have value. From there, the value seems to end. The banner indicating that healthcare.gov is improving doesn't actually improve the experience. In fact, by adding noise to the front page, the banner itself does the opposite even as it indicates other improvements having nothing to do with the front end. A system note such as this would be much more effective if given the aesthetic of an alert - no more than one line (to make it scannable) with an offsetting color that draws appropriate attention to the message.
Below this are a series of buttons, icons, and quick links that act as redundant navigation schemes which compete with the primary navigation in function. The overarching theme here is that, if you're trying to get in a house, one well-placed front door is better than four front doors.
Learn is not Learn
Users choosing learn are prompted to apply. To see plans before you apply, you actually have to select See Plans Now, a small link tucked just above fold. Conceivably on the learn page, the only thing you can do is apply. This may have been a conscious decision to drive users toward applying for a plan but it means that their navigation is a misnomer.
Mobile friendly for some, but not all, pages
Bringing up the healthcare website on a mobile device presents a user with a mobile-friendly version of the front page. But just a few clicks in, the user is required to create an online account – a process that is not mobile-friendly. This allows the user to only get as far as the first substantive task before the site doesn't accommodate mobile anymore. Normally, this may not be a huge deal-breaker, but for a site that was designed with the inherent mandate of attracting younger insurance shoppers, not optimizing the entire experience for mobile becomes a much larger issue.
Over-fixing the problem
The biggest risk is that, in crisis mode, design problems are over-corrected – or solutions are developed for things that weren't problems in the first place. All of this can be mitigated through proper user testing. The big idea is that, while system testing seems to be neglected, user testing seems to be excluded altogether. Just because performance issues cause higher profile issues, usability issues can hurt a website in the long run.
In the end, none of these issues are insurmountable for a user that is determined (or required) to use the site to sign up for insurance. Arguably, government websites have made a quantum leap in the last several years, from the perspective of user experience, thanks to good design and formal UX analysis. But usability issues must continue to be taken into account because, in the end, a live site that makes no sense to users is no better than a site that's down.