The Limits of User Advocacy or Why Everyone Should Be Investing in Usability Testing

The field of “user experience” is in a phase right now where everybody seems to acknowledge that it’s important, but no one wants to actually invest the time and money to make users the centerpiece of their web strategy.  Even when there is a testing phase baked into projects, it’s usually one week (or less) spent asking four or five volunteers to complete a remedial set of tasks in the company of a test-giver who can conveniently prompt certain behavior.  Ultimately, we often end up gambling on the expertise of the individuals on the web team rather than getting some robust data from actual site users.

This isn’t to say my and my colleagues’ expertise is not vast!  Advocating for users is an important part of all our  jobs — from information architecture, to user experience to strategy, to front-end development, to back-end integration.  But a bunch of people who make the Internet for a living have got to have a somewhat skewed perspective on how said Internet is actually being used.  At some point, if the project is going to be a long-term success, the actual user needs to be included in the planning.

To be fair, not all web endeavors require a robust usability testing phase.  The less app-like a website is — that is, the less complex a user’s interaction with the site will be — the less user testing it likely requires.  I even made a (terrifically complex) chart to illustrate this:

(Where 0% appiness is equivalent to an ad campaign site that does not contain a single user interaction more complex than a hyperlink, and where “infinite” weeks spent on usability testing means “on-going”.)

However, at this point in the web’s development, we’d be hard-pressed to find a site that is truly 0% appy.  Even a site that is ultimately just an ad campaign usually attempts to get information from its users via a form of some kind.  Furthermore, in my experience, the more brochure-like a site is, the more likely it is that there are some innovative design leaps being taken by the creative minds involved in its production — which recommends more usability testing.

Even furthermore, we seem to have finally gotten to a place where everyone recognizes that wholesale redesigns every few years is suboptimal.  Instead, web-makers are opting instead to continually tweak and update (often based on user feedback! Ahem).

So, really, that chart up there should look like this:

That’s right. I pulled out an asymptote.  Well, almost an asymptote.

Clearly, as I am arguing in this very post, you should not take my word for it.  There’s actual research backing up my mad claims!  Check it:

Wikipedia’s “References” section for its article on Usability Testing is a great place to start, though it focuses on user testing generally, not specifically for websites.

Unsurprisingly, there are whole blogs devoted to the science of usability research.  Though one should be skeptical of the data on a site promoting a service it provides, there’s a lot of externally-funded stuff on there.

If you’ve got an extra $499 laying around, this report from Forrester Research looks promising: Need To Cut Costs? Improve the Web Site Experience.

I’m sure there’s even a kit out there to help you transform user advocacy into cold, hard usability testing. Don’t make me Google this stuff for you.





The Limits of User Advocacy or Why Everyone Should Be Investing in Usability Testing

2 thoughts on “The Limits of User Advocacy or Why Everyone Should Be Investing in Usability Testing

  1. Perhaps usability testing (or “evaluation” as we say in the social sciences) could determine whether or not a business website has managed to list the hours it is open. This appears to be a perennial problem in Boston, according to my own research.

  2. I should think any website for a local business which doesn’t make its hours known would fail not just any usability tests conducted on it, but also just plain fail.

Leave a Reply

Your email address will not be published. Required fields are marked *