Undo Events as Indicators of Usability Problems
April 20th | 2009 By
One of the presentations that I had the opportunity to see during CHI suggested that Usability practitioners may be able to use undo and erase events as indicators of usability problems (a quick summary of the presentation is in this blog post). I’ve been giving this a lot of thought lately. Wouldn’t it be great if the software we design (including Web pages) reported Usability problems back to the designers?
These days, it is common practice to have software report problems back to the designers. There is not a doubt in my mind that every Windows user has recently pressed the “Don’t Send” button on the error reporting window that pops up when a program crashes. Additionally, when installing some newer software, you have probably seen an option to send usage statistics or other information back to the designer. So, sending reliability information back to the designers of software is nothing new. Yet, I have not heard any chatter about extending this functionality to improve the usability of software.
I’m not saying this is will be easy. Every software package is different and will need various heuristics to determine when a user encounters a usability problem. In the aforementioned presentation, the notion that undo and erase events can signal usability problems was specifically limited to object creation software. For this type of software, undo and erase are important functions that will be used when there is an actual user error, or a change of heart for the direction of the design. Thus, filters must be used to determine if an undo/erase event signals a usability problem or not.
So my challenge to you, the software designers and usability practitioners of the world: determine what events in your software may signal a usability problem, embed heuristics, and have the software report back to you. This may not only save money on usability studies, but may also provide valuable data from the field for the next iteration of your software.