“The most common user action on a website is to flee.” -Edward Tufte
Even the most thoughtfully planned and well-crafted product is doomed to fail if people are not able to use it. Usability testing and tracking usability metrics may seem frivolous at first, but they can both lead to a huge payoff in the end. This is the story of how one of our client’s product managers is saying yes to usability and how their business is consistently gaining new Fortune 500 clients.
Over the past two years, we have been helping our client build a suitable front-end for a large and complex web application. Throughout that time, they had been effectively utilizing the agile development methodology in all except one crucial arena: usability testing. The agile methodology promotes continuous improvement and rapid response to changing requirements. It allows the optimum software to be produced with minimal work. Unfortunately, usability testing often gets left out of this cycle of “produce, evaluate, redirect” that makes agile software development so successful. Without testing with real users, how can we ever be sure we’re going in the best direction?
We started seeing improvement as a new product manager stepped up to the role of user interface manager. She believes in usability testing, and knows that a piece of software is not “ready” until initial ideas about usability have either been confirmed or proven false by real users (outside of the development and management team). Her influence was strongly felt in the development of an important new feature for our web application. As the team designed, developed, and iterated, we kept users involved at many steps in the process. What we came up with was a new feature screen that helped put our client’s platform at the leading edge of their industry. Our client was able to use this work and work from countless others to pitch and win new business with several new clients just this year.
After watching her success, and reading a little on my own from the Nielsen Norman Group, I learned that anyone can begin incorporating usability testing into their product development process and there are several counter-intuitive truths amid the myths about usability.
It Pays More to Test Less
One such certainly counter-intuitive truth is that when it comes to usability testing, it pays more to test less. What this means is that instead of hosting day-long, multi-participant usability study sessions, it turns out that you get more value by conducting shorter, small-scale test sessions fairly frequently throughout the development process. Research from the Nielson Norman Group shows that the information you learn from the first few test subjects in a usability test is both unique and incredibly useful, while conducting more than five or six tests at a time yields mostly repeat information.
Setting – We were building a complex drag-and-drop interface that presented some difficulty to novice users.
Challenge – We watched three people bumble through the UI, and we noticed something dangerous—people could not figure out how to start. Each user required precious time to figure out where to begin.
Conclusion – We had no need for a large sample size or extensive study; we simply watched a few users hit the same stumbling blocks, and we realized what needed to change. After clearing up the first steps, we enabled users to perform desired actions in half the time.
Remember to conduct these small sessions frequently. Testing three to five users once a month, or once every few sprints, enables adaptive product design and therefore more effective development. Additionally, when testing small groups, some issues may fall through the cracks. By testing often, this knowledge gap can be mitigated. In practice, the first testing group will often help discover the major roadblocks to functionality and instruction, with subsequent test groups finding deeper issues with application architecture and data storage/presentation.
Utilize the “Thinking Aloud” Method
“Thinking aloud may be the single most valuable usability engineering method.”
–Jakob Nielson (Usability Engineering 1993)
The “Thinking Aloud” method involves test participants verbalizing their thoughts, impressions, and pain points as they try to accomplish tasks during usability tests. Hearing what users are thinking as they first encounter your application and attempt to use it gives valuable insight into the usability of the product.
Thinking aloud is flexible and efficient. There’s no special equipment needed or statistical analysis to be done. Moderators can conduct a thinking aloud session using only a pen and paper! This makes it ideal for remote testing. With the addition of technologies such as simple screen sharing and a conference call, the involved parties can be located all around the globe. The online collaboration tool join.me is a great option for conducting remote user tests.
Thinking aloud helps reveal misunderstandings of system status. This is a common usability failure when creating web applications. Users may think their actions have completed successfully when they have actually failed; they may think they have made a grave mistake, when in reality they can simply undo their action. Addressing issues in this area helps users utilize the full extent of functionality without the “state-confusion” associated with complex, multi-step operations.
Thinking aloud helps inspire future functionality requirements. The user may be interested in different functionality or data than previously assumed. Thinking aloud during a user test allows the moderators to know what the user wants to do. Often a user interface may restrict the user, and hearing their thoughts reveals their true dissatisfaction. When you understand the user’s thought processes, you can more effectively organize their data, show potential next actions, or help them visualize task flow. This data is vital in user-centered design and will lead to better-targeted products.
I recently participated in a user test in which the moderator, the test participant, and the application screen were connected across three different states.
Prioritize and Respond Quickly to Test Results
The final step to making the most out of simple user tests applies after testing is complete. What should you do with the information you have collected? At this point, you should not worry about timeline or fitting work in to an already packed schedule—just take a few minutes and figure out what could make the product better based on your findings. If time were not a factor, what would really make the application more usable? Once you have a goal in mind, then look at the level of effort involved with making each of those changes and create a prioritized action plan for moving forward.
Sometimes, the solution may involve slipping a few bug fixes into the upcoming sprints; other times it could mean taking another look at your application architecture. Either way, the earlier these changes are revealed in the software development lifecycle, the more time and money saved for the project. In her article in Cost-Justifying Usability, Claire Marie Karat showed from her experience that “for every dollar spent to resolve a problem during product design, $10 would be spent on the same problem during development, and multiply to $100 or more if the problem had to be solved after the product’s release.” This “rule of 10s”, as Karat describes it, should invoke a strong sense of urgency in addressing usability issues.
Usability is often the first thing to get cut when time and resources are concerned, but if you look at the amount of investment and gain, it’s obvious that usability testing should be a high priority for any product team. The second principle behind the agile manifesto states: “Welcome changing requirements, even late in development. Agile processes harness change for the customer’s competitive advantage.” Like our client’s product manager, consider how usability could be the change that gives you a leg up on the competition.