REDMOND, Wash., August 31, 2000 — Alan Waters and Laura Briggs are about as average as you can be. They each have kids in school and busy lives. Alan is a working-class guy whose teenage daughter knows more about the Internet and computers than he ever will, and he’s starting to get nervous that the technology era may be leaving him behind. Laura, on the other hand, is not shy about using the computer, but is mainly interested in the Internet as a way to save time as she works to keep her family running.
Familiar as their lives may seem, there’s something unique about Alan and Laura. They aren’t real. They are archetypes — or
“personas”
— of Internet users or potential users created by the Consumer Usability Group at Microsoft and inspired by the work of Alan Cooper, author of
“The inmates are running the asylum.”
In this case, the personas were created to help developers build the new MSN Explorer, which will soon be available in a preview 2 version that offers superior integration and user experience so that people can get the most out of their time on the Web. Named and
“brought to life”
by usability engineers, personas are tools that help make customers very real to the product developers while comprising a rich collection of consumer data, field research and other studies.
MSN Explorer is software that packages the Internet, giving people a more predictable, comfortable, intuitive and productive way to take advantage of everything Internet, from Web searching and shopping to email and instant messaging. Unlike America Online (AOL), for example, MSN Explorer doesn’t create a
“walled garden”
that limits users’ content options and makes it difficult to get out. Instead, it is an online client designed to make it easy for people who are just starting to use the Internet, while also empowering more experienced users to get the most from their time online.
“The usability team has helped to create Microsoft’s most consumer-friendly product ever,”
said Bob Visse, lead product manager for MSN at Microsoft.
”
By creating personas and conducting user research, field studies and laboratory testing, the team shaped the outcome of the product based on what they found users need and want.
“MSN Explorer is more than just a browser. It’s a totally integrated Internet experience,”
Visse said.
“When you log on to MSN Explorer, you can see right there in the interface how many instant messaging buddies are online, how many email messages you have in your inbox, the weather conditions specific to your local area, and links to local news. So it’s a personalized experience that blurs the line between local software applications and Internet services and activities, offering an overall great experience for consumers.”
Champions of the People
Ensuring that
“overall great experience”
is where John Pruitt comes in. As a usability lead in the Consumer Usability Group, Pruitt heads the team responsible for user research and testing for MSN Explorer. Working closely with developers and program managers — those who design and build the software — the usability team serves as the unquenchable user advocate, instilling user needs into the development process and representing the user perspective in design meetings and functional specification reviews.
“Usability is considerably different from marketing, which focuses on selling and promoting the product,”
Pruitt said.
”
Our key objective is to provide user data for product development. We’re interested in making sure our products are both useful , so they do the things people want them to do, and usable , so people can accomplish those things easily. On top of that, because we’re working with consumer products, we’re also interested in the hedonistic side of things — whether or not it’s enjoyable.”
“
Personas
”
Anchor Developers’ Image of the Consumer
One of the most important functions of usability engineers is to communicate to the development team, as clearly as possible, who the users really are, Pruitt said.
“When designers and developers don’t share a distinct image of their user, they carry different interpretations of ‘user’ around with them. As a result, they are prone to developing a schizophrenic interface.”
Thus, personas like the above-mentioned Alan and Laura were developed by Pruitt’s team to prevent this problem from occurring throughout the development process of MSN Explorer.
Personas begin with consumer data. The team talked to product planners, and turned to internal and external studies — as many Internet, e-mail, Web-related marketing studies as possible — to understand how the universe of users is segmented. From those demographics and related data, they generated the skeletons of four personas. Then, turning to usability field data, they incorporated various end-user goals and motivations into each persona’s profile, as well as more in-depth personality attributes they found to be appropriate. The key was that every persona attribute had to reflect real data.
The usability team generated illustrations of the four personas and introduced them to everyone on the development team.
“Essentially, we told them, ‘Here are the individuals we are designing for. Understand them. Know them. Refer to them specifically as you discuss design issues. Would Laura do this? Would Alan even want to use this feature?'”
Pruitt said.
“So the personas gave people well-defined, commonly understood referents that were used as people discussed development issues. It gave the team an instant and human understanding of our end consumer.”
Building a Friendly and Functional Interface
Throughout development of MSN Explorer to date, the usability team identified several issues that it helped the product team understand and address up front:
“shallow”
integration with MSN properties, confusing jargon and an incongruous approach to certain user activities.
One of the key strengths of MSN Explorer is its integration with MSN services and online properties, such as MSN Messenger, MSN Hotmail, MSN Search and MSN eShop, with one-click access directly from the MSN Explorer interface. This integration means that users can easily glide from one MSN property to the next and stay fully immersed in MSN Explorer without logging in multiple times or losing track of where they are. Midway through product development, usability testing revealed abrupt divisions between the intuitive, user-friendly MSN Explorer experience and some of the MSN services. As a result of usability findings, the team generated and implemented solutions for the disparities and a positive side-effect of the studies has been an increased awareness of usability issues across all MSN properties, spurring plans for continuous ongoing improvement.
A second area, the choice of terminology in the product, is a typical challenge. Software naturally takes on the vocabulary of the developer, which can be meaningless to the end user. The instant messaging button changed names multiple times because of this very issue, Pruitt said.
“People didn’t know what instant messaging was. Someone came up with”
Buddies,
“which was moving in the right direction, but usability testing showed that it wasn’t enough to help people unfamiliar with instant messaging understand what it was for. So in the end we changed it to”
Online buddies,
“which may sound like a subtle change but it makes a world of difference for those first-time users who are not yet accustomed to Internet slang.”
The third type of issue is when the user comes to a task with one approach in mind and the software uses a different approach. For example, if a user wanted to move an email message into a folder that they haven’t yet created. According to Pruitt, the original way that process was created in MSN Explorer was completely different from the way users were inclined to accomplish the task, so the developers changed the way the pull-down menus worked, the items in the menus, and the way check boxes and the inbox worked. Usability testing identified and remedied this incongruity between user behavior and the design of the product. Another example is noticing that customers often try to send email to someone by typing an email address into the Web site address bar instead of into the email address bar, and then they can’t figure out why it doesn’t work. With MSN Explorer, we decided to just make it work. So now when a user does this, it will automatically do the right thing and create a new email message addressed to that person and be ready to go.
One of the key attributes of MSN Explorer is the lack of window clutter.
“As much as possible, our team has tried to keep multiple windows and dialog boxes from popping up in users’ faces,”
Pruitt said.
“That’s one thing that’s particularly bad about the AOL experience — it’s window hell. And our user studies have definitely confirmed that. Users lose track of where they’re at as windows appear on screen. And they don’t know whether they’re seeing AOL content, or something they’re trying to sell. We’ve heard from AOL users that they’ve automatically dismissed an instant message from a friend because they thought it was an ad.”
While it takes a world-class team of developers to create a product that skimps on neither form nor function, the crowning accomplishment of the usability team has been in enhancing the MSN Explorer interface in response to the way people think and work. Usability testing was behind much of the elegance and intuitive organization of the onscreen information.
“Our studies are showing how much simpler it is for people to navigate around the Internet,”
Pruitt said.
“We have an excellent design team on this product, and they have paid particularly close attention to usability. The designers come to our tests and they watch what users do. As a result, in addition to benefiting from MSN Explorer’s streamlined capabilities, users are drawn to the fact that it looks elegant and is so easy use. They definitely have an initial reaction to the interface that’s positive — and a lot of that is due to having such a user-focused design team.”
Usability Field Study Testing
In their initial field study for MSN Explorer last summer, usability engineers gathered requirements for the software through field research, to understand which features were most important to consumers. The development team visited 15 homes, representing a variety of user types and methods of getting online, to conduct
“contextual inquiries,”
an interviewing method in which participants actively show how they accomplish various tasks, as opposed to explaining what they do. This method, said Pruitt, removes the filter of participants’ perceptions by having them actually do the activity — like telling them,
“send an e-mail to your best friend”
— as opposed to just talking about the activities they engage in. According to Pruitt, this is a very powerful way to understand user behavior and to develop user modification because the data gathered is firmly grounded in the user’s environment as well as the activity itself.
Usability Lab Testing
Lab testing for MSN Explorer began with the earliest paper mock-ups produced by the designers. Using a method called paper-prototype testing, study participants interacted with the paper design, using a pen instead of a mouse. As they
“clicked”
on various parts of the mock-up, a development team member would respond with a sticky note, displaying the intended result or dialog box, for example.
“What’s really nice about doing paper-prototype work is that the users themselves can get involved in the design work. And because it’s at such a rough stage of design, they’re less reluctant to say, ‘this is all wrong! What I really want is for this thing to be there,'”
Pruitt said.
“Because we haven’t spent months coding up the interface at this point, it’s very cheap to just wad up the paper and throw it in the trash.”
As developers started to write code for MSN Explorer, iterative lab tests began. Five to seven users in the lab would run through a given feature area, which was modified based on user recommendations and then retested.
“This process gave us very quick reactions and helped us determine exactly what consumers needed in the User Interface,”
Pruitt said.
“It created a situation where users drove the entire process.”
With the Product Launch on the Horizon…
The final release of MSN Explorer is just around the corner, but that doesn’t mean usability work is winding down. Currently, the group is doing usability benchmarking on the product, measuring the ease with which users accomplish about 30 tasks that cross the entire product to generate a broad overview of the user experience in comparison with competing products. One version of the study is being conducted internally, and another by an independent external organization.
“This deep, thorough research and testing throughout the product development process is what sets MSN Explorer apart from anything else out there,”
Visse said.
“The usability lab took our vision for MSN Explorer and then systematically tested users with practically every attribute of the product,”
he said.
“It’s one thing to say you want the product to be easy, but until you watch real people in front of it, using it, you can’t be sure whether your ideas truly equate to ‘easy.’ There isn’t a single feature in MSN Explorer that wasn’t improved at some level because of this intensive real-world testing.”