It seems that for the last few years I have been involved with usability testing in some shape or form, although I suppose it could be debated if there were any “usability tests” at all. These “tests” were more heuristic, always informal in nature and continue to this day – in no way did they involve the scientific measurement of users and their interactions with the system in question. There are a few explanations why the testing has been so informal…
First, you cannot force end-users to evaluate the system unless they are paid, bribed, or are already convinced of the benefits of the system. Sure, ideally it would be great to enlist a bunch of excited users to test the system and provide feedback, but often there are constraints. In my experience, the constraint has always been money, time, and prestige. There’s little money devoted to formal testing with end users. End users do not have the time for testing. A majority of the primary users of web-based systems I supported were scientists or Ph.D.s who were not open to testing the system. Still, even when end-users were kind enough to participate in beta testing, the focus was often on what the paid software testers reviewed – does the system work, are there bugs, etc. There was no systematic review concerning how usable the system was and how long it took for end users to accomplish needed tasks without giving up. In addition, the beta testers often did not fully represent the types of users of the system.
Concepts such as “information architecture” and “user experience” have been all but void in my work environments. Places that I’ve worked at have never even heard of these concepts, much less valued the user experience enough to hire an employee to meet these needs. Traditional “testing” usually involved making sure the software worked as advertised with as few bugs as possible. The usability of a system was never a consideration. Just pay a contractor to provide some training sessions and throw in a user’s guide.
I believe as software development processes have matured, more consideration has taken place to the user experience, particularly as the trend has moved from client/server-orientated solutions to web-based solutions. After all, a heightened user experience and the greater usability of a website translates into a larger customer base, with a greater possibility for increased revenue. A number of people (such as here, here, and here) have discussed the power of a well-designed web presence from a ROI perspective. In fact, Jakob Nielsen indicates that,
“Development projects should spend 10% of their budget on usability. Following a usability redesign, websites increase desired metrics by 135% on average; intranets improve slightly less.”
Which brings us back to the need for usability testing. What would the Internet World look like if all software development teams included information architects, user experience professionals, and the like? Well, for one, there would be much more usable and user-friendly sites.
A wise client would award an incentive-based contract to a contracting team. Can you imagine a contract that paid for ROI milestones? Project managers would need to do a lot of up-front work with their clients, not only to determine the nitty gritty requirements, but to determine the strategic objectives for the design of any solution. It is only by determining an organization’s strategic objectives that software development teams can build a solution that can meet the needs of their clients and all stakeholders. So to the bigwigs out there with the purse strings, demand that any software project plan include sufficient funding for proper business analysts, information architects, and user experience professionals, including funds for usability testing. A few bucks now will save you money, it will increase brand loyalty, and it will increase satisfaction among your end users.