Mobile UX Research: An Interview with Danielle Arvanitis
One summer, Danielle Arvanitis fell in love with a Macintosh—and a passion for user experience was born.
Since earning a master’s degree in interdisciplinary studies at the University of North Texas, she has worked as a user researcher, interaction designer, and consultant on Web and mobile projects for Nokia, Nationwide Insurance, Fannie Mae, A.G. Edwards, Domino’s Pizza, and Valtech Technologies, among others.
She currently works in user research at Good Technology in Sunnyvale, Calif., and lives in nearby Mountain View with her husband, Andy. When she isn’t working, she loves to get out on her road bike. You can follow her on Twitter at @dannyoyello.
UserZoom had the opportunity to interview Danielle for our blog readers. Here are some great insights into the mobile UX research and best practices for mobile product design.
What are some of the biggest Mobile UX research challenges that organizations face?
Danielle: The first one that comes to mind is the incredible variety of display sizes and input technologies that exist. If you’re creating apps or websites that will be used on many different devices, how many should you test on? Few companies have the resources to do testing on every device out there. Most companies will choose the most popular one or two devices for each platform or, if they have statistics, the ones that most of their customers are using. It just isn’t much of a consideration for the desktop.
Another one is that, for mobile usability testing, there’s no tool like Morae that does it all—stream the session, record the screen contents plus hands and facial expressions, allow observers to make notes on interesting moments, gather statistics, etc. There are some nice tools out there, but no silver bullet.
A third challenge is that when you’re observing users in the field, it can be difficult to see what they’re doing on the device. I heard at a recent talk by Nate Bolt that Facebook has a treadmill for simulating walking down the street. It isn’t the same as the real thing, but it would help you observe things like hand posture and one-handed use while also being able to see what the user is doing. I’d love to get one for our office.
What are your favorite mobile UX research methods and solutions? What are the pros and cons of these methods?
Danielle: When I first got back into mobile UX research, I used Reflection on iOS to mirror a device screen to my Mac; Silverback to record the audio, screen, and the user’s face; and WebEx to stream the audio and screen to remote observers. It was quick, cheap, and easy, but for remote observers there was no picture-in-picture of the user’s face and no view of their hands or where they were touching the screen. For people who are remote, that’s a quick trip to the land of nod. You should use an over-the-shoulder camera so observers can see the user’s hands, or software like delight.io that integrates with your app to show where they’re interacting with the screen. Granted, the more pieces there are, the more likely it is that something will go wrong. But you’ll hold remote observers’ interest much better.
What are your thoughts on remote mobile usability testing?
Danielle: Remote unmoderated testing is something we’re looking into. I’d love to be able to use something like UserZoom for mobile research. Because we sell to IT admins and our value proposition is mobile data security, it can be tricky to get them to let us interrupt their users and record their screen interactions. UserZoom’s Mobile VoC solution is a low-friction way to gain feedback after users have finished up some work. I really enjoy the discussions and insights that are part of moderated formative testing, but we have a small staff, so the bang for the buck of remote unmoderated would be a boon to us. We have a plan to work on that this year.
Can you give us some guidelines for selecting the right mobile UX research method for your organization?
Danielle: For me, the primary determinant is resources. You’re going to generate a lot of data that needs to be reduced to meaningful insights. If you’re primarily testing apps, you also have to spend prep time downloading, installing, and configuring builds; creating and loading test data; reloading test data between sessions; taking notes; and recording statistics. For usability testing there are few tools that will generate statistics for you, so it’s a lot of work. Anything that can help automate data collection or make it easy to generate video clips will help your researchers get more out of their time.
What are some of the best practices for mobile product design?
Danielle: There’s a great article in the Cooper Journal from August of last year titled, “The Best Interface Is No Interface.” It gives a few examples of needless interfaces and asks us to question whether we need an interface at all to solve the problem at hand. For example, car manufacturers are creating phone apps that require painful, multi-step processes to unlock a car door. They’re worse than using a key. Yet in 1999, Mercedes invented the smart fob. What could be better than just walking up to your car door and opening it? Another comparison it gives is paying for something with your phone using near-field communication and paying for it using Square Wallet. The first feels unnatural; the second feels perfectly natural. It can be difficult for UX folks to embrace this concept, but the best interface is no interface. When you don’t have that option, though, here are a few best practices.
1) Know what you’re trying to build, and for whom. These days, mobile design is very focused on visual design and the screen itself. Does your app look slick? Does it have cool animations? What gestures does it support? These things certainly have their place, and can be very important depending on what type of product you’re building. But if you haven’t solved users’ problems, they’ll ultimately abandon your site or app for one that has. Choosing which problems to solve is important for every design project, but even more so for mobile due to limited screen real estate, intermittent connectivity, and difficulty of data entry. Focus on the basics: users, tasks, and context. First, who are your users? Are they conservative, tech-avoidant people who don’t use social networks much? Are they Facebook-addicted twentysomethings who try ten new apps a week? Are they salespeople who can’t miss a meeting or deadline without losing a customer or a sale?
Second, what tasks do they absolutely need to be able to perform while mobile? What are their expectations? What are their nice-to-haves? Form factor is a big consideration here. On a mobile phone, they might focus on simply triaging their email while they’re at the airport, but on an iPad they’re typically sitting down and will expect to get some real work done.
Third, what is their context? What does “being mobile” mean for them? Are they merely away from their desk at work? Are they spending half the day in taxis, visiting clients in Manhattan? Are they hiking up a mountain, wearing gloves and a backpack? Seeing your users “in the wild” will spawn ideas for making your site or app better. To sum up, there’s always pressure to push more and more features and density onto mobile screens. Knowing users, tasks, and context lets you focus on your users’ priorities and push back on requests that will needlessly complicate your product.
2) Prevent fat-finger mistakes. Data entry is one of the most frustrating aspects of using mobile devices. Three things to focus on here are tap sizes, text sizes, and text entry. First, make sure your tap sizes are adequate. Each manufacturer has created its own standards, so be sure to follow them. There’s a nice discussion in Tapworthy by Josh Clark. Second, don’t try to squeeze too much information on a screen. Small type is difficult to read and interact with. Set a minimum font size and stick to it. Lastly, minimize the need for text entry. Even the person at our company who’s known for “doing everything on his iPad” admits that he doesn’t use it for content creation, and there are people who would rather put out an eye than compose more than two or three sentences on an iPhone. Use auto-complete. Let people select commonly used words or phrases. There are all kinds of tricks for lessening the pain.
3) Don’t assume that platform designers are doing the best work out there. People have grown accustomed to native iOS and Android controls, including date pickers. But if you watch someone try to set a date, it’s common for them to spin right past it on the first try. It often takes them four or five attempts to get it right. Sometimes the picker doesn’t include the day of the week, and the user is looking for that instead of a numeric date. These pickers have been around for almost six years on iOS, and they don’t seem to have gotten any better. Try to keep a fresh perspective and continually ask yourself what can be improved.
Note from the editor: To learn more about the mobile product design and usability, watch a recorded webinar, Critical Insights for Mobile Product Usability.