Looks [of your software] Do Matter

As an experiment, I once posted a form on my previous company’s EMR. My intent was to see how my colleagues would react to the form, and I intended to use this information for further design of the EMR. This goes back quite a number of years when I was still on shift in the clinic. In a totally anecdotal way, I simply watched doctors reaction as they came upon this form. The form covered the entire interface of the EMR and had one button on it that said “press here”. Most of the physicians actually took a few moments to look around the screen to see if there was some other available option rather than to press the button. A couple of physicians stood back and were effectively frozen until I stepped in and showed them that all they need do is click on the button.

This mean little experiment was intended to test a theory of mine that an EMR interface must be intuitive. That is not the same as saying that it must be simple, because nothing is a simple as a single button that says “press me”. An intuitive interface is one that must have a visual flow to it, such that users’ eyes naturally follow a path across the screen. There needs to be sufficient explanation to make it clear to the user what is to be done, but not too much as to overwhelm the user [especially if the size of the text is exceedingly small].

Over the years, I have read a great deal about interfaces for all types of software. I have asked many users of different kinds of software what they most like or dislike about a program’s particular interface. One user once told me that he found a program’s interface to be very depressing. The entire program, which consisted of hundreds of different screens, all had a background of gray. At a certain point, this user said that it felt as if all the buttons [which were also gray in color] just melded into the background. At a certain point, especially when the user was tired, it became far too easy to click the wrong button and then spend a significant time moving backwards to correct the error. It was actually due to this single discussion that I personally decided to use multiple bright colors in my EMR. I received multiple complements about this choice of color, which solidified my opinion.

By the way, the idea of going back and correcting errors is also a major issue in interface design. I know of one user who told me that it was easier to delete a couple of minutes of data entry rather than try to go back and correct the single field that was inappropriately entered. I heard from another user that the back button would take the user to a webpage that was not previously viewed. The assumption of the developers was that clicking the back button indicated a desire to move to a different section of the software, rather than revisit the previous page.

Another issue which I have heard a great deal about is the number of clicks it takes to reach a particular destination within the software. One physician noted that in order to write a prescription, it took close to a minute to work through a series of screens which finally presented an option for writing the prescription. On this prescription page, there was little in the way of assistance, such as a drop-down of available medications or suggestions for appropriate doses or calculators for doses geared towards children. The physician using this prescription tool often had to rely on other websites or printed material to know what to prescribe. After all of this, the doctor had to click on a series of screens to return to the main interface of the program. Needless to say, doctors were less than thrilled any time they needed to write a prescription, especially if it was an ad hoc prescription that was being written while not in the middle of recording the entire medical record for the same patient.

There are books written about creating effective interfaces for all different types of software. At times “hard-core” programmers do not sufficiently respect the work of interface designers, who must use the various available functionality to back up the buttons and checkboxes that they design. It should not be a surprise that certain software will fail, not for lack of functionality, but for lack of ease in its usage. I have often read reviews of high-end graphic design software that on one hand is highly regarded for its amazing functionality but on the other hand is rebuffed for the complexity of the interface. The “learning curve” to master this excellent software was so steep, that only people truly dedicated to becoming expert in computerized design, would take the time to master the interface. I imagine that in some ways this was a self-selection process which proved to the client, that the designer using this software was in fact very dedicated and of a very high skill level. From the software company’s point of view, I can’t imagine that such a convoluted interface was very good for sales.

The movement of software to the mobile phone has forced developers to squeeze in a great deal of functionality into a very small space. While some interfaces really do try to use the smallest possible font in order to write more on a mobile phone screen, most developers effectively retrain themselves to design interfaces in a wholly different way. Also, mobile users tend to have even less patience than desktop users. As such, if it takes more than two to a maximum of three clicks to get the desired information or perform the desired action, users will tend to abandon the software, even if they have paid a few dollars for its use.

I am hearing more and more an old adage that says that the best interface is no interface. The idea is that the software being used should be intuitive enough such that it can understand the user’s desires without requiring typed letters or even hand gestures. As computerized systems become more ubiquitous and more aware of our habits, we will find that a great deal is done for us without any action on our part. For example, if we receive a notification requesting a copy of our driver’s license, our computerized home may simply say that it is about to send the copy unless we express the desire not to. So, simply ignoring the notification will automatically perform the act that we wished to happen. So much of what we do is habit, and recurring events, such that we will find that we must do less and less to obtain more and more. I will not discuss the philosophical overtures of living in such a way, at this point. But there is definitely an audience for such systems. And that audience is most likely to be far bigger than we would first imagine.

Thanks for listening.