I tend to be fairly astute. I like to think that I pay attention and have a knack for catching things that may often fall to the wayside. Occasionally, however, I am prone to the same user interface fails that plague humanity on a daily basis. This would not be that big a deal if I were, say, a doctor. Or a carpenter. Or a member of any profession to which design, usability, or user experience, do not factor. As a UX professional, I can comfortably say that, in most instances, the failures I do experience are often the result of poor design, or at the very least a design which violates any number of typical interaction models. The following story is not one of those instances.
In the fair city of Chicago, parking meters are a common element of daily life. These meters have evolved from the old coin-loving models of yesteryear into a centralized meter station, like this one:
So, here’s how this little bugger works. You park your car within its little area, walk up to it, swipe your credit card in the slot, select the amount of time using a series of buttons, validate your time and the cost in the display window, press the “print receipt” button, receive a paper/sticker receipt with your valid time, and finally place the paper/sticker receipt dealie on your dashboard. Done and done. This is the process, and you’ll note that it is laid out in the white instructions sticker planted firmly in the bottom-left corner of the apparatus. So it is for literally thousands of parking meters across the city.
With one slightly irritating exception.
One evening, while parking in a particular parking lot in the northern part of the city, the fail occurred. This particular lot’s automated meters are well-known to be a problem. After observing a number of individuals familiar with this lot, the consensus on the matter is this: they never work. One evening, I attempted this lot again. I followed the process described above, as always, and still nothing. I stood there for five minutes cursing the confounded machine until a gentleman behind me interjected. “You have to select your time first,” he said. “It’s not like the other ones.” I conferred with him about this for moment. Apparently, he had only found this out after calling the parking operations people who confirmed how it worked.
At first glance, this is clearly an issue of consistency of interaction. How could this one lot, with identical meters, work completely different than every other meter in the city? Why would anyone do that? I would understand if it were not a municipal lot and therefore operated by a different agency with their own rules and processes governing the operation, but it isn’t. My initial consternation at hearing the solution to the problem was, “why don’t they tell us that in the first place”?
As you can see below, however, they actually DO:
Not only does this meter overwhelm you with directions, it includes strong affordances such as a big, red sticker over the credit card reader that says “STOP – Add Time First”. Taking 30 seconds to scan the meter gives you all of the directions you could ever possibly need, but it still did not stop my concerned citizen from calling the parking services number rather than soaking up this information.
Because all of the explicit direction in the world is not enough to override habit. There was nothing in his mind (or mine for that matter) that led him to believe that this meter should operate differently than the hundreds of others he’s interacted with over the years. Convention is the prevailing norm, and, in absence of structural or foundational differences, will always trump instruction, guidance, or information. Think of how effective this model for interaction would have been were the meter a different color, shape, or size than other parking meters. Sadly, however, it isn’t.
The lesson we can draw from this: break convention at your own peril. You may think your way is the best way. You may believe that your interaction model, design, or flow is the most effective and efficient way for users to accomplish their goals, but if it does so at the expense of previously and, in this case, cemented expectation, you are asking for disaster.