Industrial Design & Human Factors: Friendly Face
September 1, 2006
The choices seem endless. Chocolate or vanilla? Mixed nuts or chocolate chips? A combination? Dessert afficionados in shopping malls throughout suburban Boston face these same questions when they stroll up to MooBella ice cream machines to order a frozen treat. Choosing from the more than 90 possible ice cream combinations takes more than a palette for flavors; it takes an interface that is easy-to-learn and easy-to-use by consumers of all ages.
The MooBella ice cream machine is in many ways a microcosm of what designers face when creating a user interface: the users are varied, the skill sets unknown; the environment for choosing can be frenetic, the frustration factor a distinct possibility.
Wiklund Research and Design, Concord, Mass., redesigned the user interface on the MooBella machines. Michael Wiklund, a president of the company, says that the manu-facturer initially found that consumers were having trouble discerning how to place their order.
The redesign process began by conducting a human factors' study, which examines how people actually interact with a product both physically and psychologically. The interface solution was to lead the consumer through the process in a simple, straightforward manner using a touchscreen menu system, and, at the same time, making it an enjoyable process. An onscreen graphic on the interface represents the process as it happens: ice cream fills into cup, mix-ins added. "We boiled it down to the simplest possible process, while making sure that it was still fun," says Wiklund.
The product's faceUser interfaces are often the "face" of a product. For customers, they are the link between utility, or usability in human factors parlance, and futility. They are a key factor in satisfying a customer and perhaps bringing the consumer back to a brand. Alternatively, a negative experience may drive the user into the arms of another manufacturer.
Just about every technology-based product on the market has a user interface of one sort or another. They may include physical interfaces such as buttons on a cell phone keypad or an electronic interface such as a touchscreen menu system on an automatic teller machine. A bad interface might be cell phone buttons that are too close together for the audience, or a menu system in which a user gets lost after a few screens.
Of all the products on the market that use an interface, more than 50 percent of those products' interfaces could be designed better had the company done more research into how a product is actually used in its natural setting, says Stephen Wilcox, principal for Design Science, Philadelphia.
Better interfaces can literally translate to the bottom line. A well-designed interface can be the difference between a successful product and an underachieving or even failing product. It can also affect, positively or negatively, future sales and public opinion of the company.
A October 2005 study by J.D. Power and Associates, an industry forecasting company, hints that consumers are becoming frustrated with the user interfaces on their cell phones. "As more services are added to mobile phones, the ability to navigate the handset in an easy and straightforward manner becomes paramount," says Kirk Parsons, senior director of wireless services at J.D. Power. He says that manufacturers need to make it easier for consumers to learn how to operate the wireless phones. "In fact," he adds, "the handset can enhance the consumer's view of the wireless carrier by making the service experience much more enjoyable, which will promote future purchases of additional products."
Increasingly, companies looking to build customer loyalty are turning to experts in the field of human factors. These experts may conduct usability tests or ethnographic research.
A usability test studies a representative sample of potential users by having them interact with the product. As they do so, observers note any difficulties the users meet and ask the subjects to think out loud while they work so that observers can understand how and why they do the things that they do. Multiple usability tests maybe required as an interface is developed, a feedback loop may alter the design as the process progresses, or if the user's skill levels may vary between everyday users and those whom Kath Straub, chief scientist at Human Factors International, Fairfield, Iowa, calls "superusers."
An example of how this testing effected design can be found in India where a large American manufacturer of automatic teller machines wanted to introduce a new ATM machine in that market and compete against another manufacturer's machine. Sarit Airora, an interface designer with Human Factors International based in India, says one flaw was that the money and receipt came from the same place and users didn't expect that. The biggest flaw, however, was a simple fix.
"They were using green symbols for money all across the software user interface," he says. "Money is never green in India and the subjects were very wary of accepting that process."
Ethnographic study requires that the researcher go into a real-world situation to videotape and study how and why people do the things that they do. This may include asking questions as the tasks are performed, after watching the video, and by careful examination of the data revealed.
Wilcox says that conducting human factor studies could generate at least double the amount of information about users and help to redesign better interfaces.
Straub adds that many organizations are starting to get it. "Usability is a pillar for creating brand loyalty and trust," she says. "If it is aesthetically pleasing and is easy to use, the user is more likely to stick with it. But if they struggle with it, then they will have to think twice before they make that choice again."
Market research flawed?Knowing if an interface is appropriately designed can be tricky. Too often, experts say, these decisions are made in closed rooms with engineers and designers making assumptions about what the consumer wants and what the user can actually accomplish based on what they themselves know and what they can do.
"That is the real problem with many user interface designs," says Barry Beith, CEO of HumanCentric Technologies, Cary, N.C. "They are the object of their design so their use of terminology is almost always off in some way."
For a design engineer, the biggest danger lies in designing an interface for another engineer.
"Engineers process information differently than the average individual," notes Ken Klask, CEO of Amulet Technologies, a supplier of user-interface ICs based in Santa Clara, Calif. "Engineers can be blind to a problem because they already know where the menus and sub-menus are headed; they already know how the product works. The users don't."
Another method that many companies employ to learn about their customer's needs is traditional market research such as a survey. These answers can be helpful, especially in the early stages of interface design, but, at the end of the day, is it pointing the manufacturer in the right direction?
Any data gathered must answer what Beith, an interface expert, says are the four basic questions of interface design: what are you doing? Why are you doing it? Should you do it. And if so, are you doing it the correct way?
Traditional market research may not answer all of these questions, and most likely wouldn't answer them with as much data as can be collected through a rigorous user centered design process. Traditional market research can be flawed for a number of reasons. The subject's ability to articulate why they like something is one reason, but certainly not the only one.
"Market research does not allow us to get enough depth about the subject," says Wilcox, an expert in the field of ethnography. "The things that people do and what they want are largely unconscious decisions. That is why market research is an imperfect predictor."
Or, they simply may not think of it, he adds. Some people just assume the obvious, what he calls the cupholder phenomenon. "If we ask people what they would want to make their car better they might say to move the cupholder, they won't say ‘I want a better internal combustion engine,' because that is a given," says Wilcox.
Wilcox cites a vacuum cleaner example. Prior to the ethnographic study conducted by Wilcox and his team, the manufacturer had conducted focus groups on vacuum cleaners, specifically canister-type cleaners, and a typical answer addressed was to improve the ease of using the electrical cord.
"For this product, we went into houses and studied the use of the vacuum cleaner," says Wilcox. "We discovered that nobody mentioned bending over, and this is something that we witnessed hundreds of times. To them, bending over was just a given. Armed with this information, we helped the manufacturer develop the very first electronic vacuum with the controls in the handle."
A simple fact: nobody likes to bend over every time they need to turn on or off the vacuum or change a setting. Yet, none of the traditional market research had pointed this out and this data help create a market leading product. To use a delicious analogy, if market research were a scoop of vanilla ice cream, then human factors' might be more on the order of mocha ice cream with nuts and candies mixed in.
For more information email:Amulet Technologies: firstname.lastname@example.org
Design Science: email@example.com
HumanCentric Technologies: firstname.lastname@example.org
Human Factors International: email@example.com
Wiklund Research & Design: firstname.lastname@example.org
SIDEBAR: Decisions, DecisionsDesigning a user interface requires that the designer make a number of decisions that may affect the usability of the product and the satisfaction level of the customer. Due to the myriad of variables involved, a list of design tips could fill a book, but there are a few general design tips to help guide the interface designer.
When facing a difficult decision, Barry Beith, CEO of HumanCentric Technologies, Cary, N.C., says to be obvious. "Don't rely on their intuition," he says. "Map what you need someone to do in the interface in such a way that they can read the map and understand. Don't rely on their intuition unless you absolutely have to."
He adds that designers should, instead, rely on very direct information in the interface to tell the user what to do. Do not use ambiguous information. Do use consistent information. "The information that maps the way people do things in the real world."
Michael Wiklund, president of Wiklund Research and Design, Concord, Mass., says that most problems can be fixed without much trouble. Which is not to say that these issues won't cause trouble - think of poorly designed interfaces for medical equipment-but rather that they can be easy to fix.
Wiklund points to a number of design tips including reducing screen density, providing navigation cues and options, limiting the number of colors, simplifying type, and being consistent with icons.
For instance, Wiklund says that a user may have trouble with menus and submenus; for users, getting lost in a menu system is a common problem, he says.
To help alleviate this problem, he suggests breaking the information down into five or so major components. To think of it in another way, break them down into five doorways through which the user can find additional information.
Beith also cautions the designer about objective and subjective design. In other words, does it look complicated, but is easy to use; or, does it look simple, but is difficult to use?
For instance, he points to old PBX telephone systems that looked complicated because they featured a lot of buttons and switch points. However, once the user got past this fact, the interface was easy to use because each button represented an individual person. Compare this, he says, to other telephone systems that look easy but each person on the system has a two-digit or three-digit code that the user needs to know or figure out in some fashion.
"That is really some of the confusion in product design that we run into today," adds Beith. "Designers want to make their appliances look simple to use, but the very act of them making them look simple complicates their use."
SIDEBAR: Interfacing with DemocracyCan interface design affect the future of democracy? Kath Straub, chief scientist for Human Factors International, Fairfield, Iowa, believes it can.
When it comes to voting machines, Straub says that usability issues result in problems such as registration mix-ups, faulty polling equipment and confusing ballot design.
"Much of the focus on electronic voting is on security: is my vote secure and confidential," she says. "But another problem is the usability of the interface. The question is this: is the vote I intended to cast the one I actually cast?"
Straub points to a July 2001 study, the CalTech/MIT Voting Technology Report, that estimates that 4 million to 6 million votes were "lost" in the 2000 Presidential election. Straub, writing in Human Factors International's May 2003 e-newsletter says that 1.5 to 3 million votes were lost to registration mix-ups.
Another 1.5 million to 2 million votes were lost to faulty polling equipment and confusing ballot design and up to one million votes were lost because of polling station policy problems in which ballots were lost or mishandled because of poorly written standards for identification.
She says that having professional interface designers to create the ballots and paperwork and routine usability testing would have helped to ensure that votes were correctly counted.
"In the election cycle," says Straub, "there is clear evidence that if you look demographically at who uses a computer and who doesn't, those people who don't may be afraid of technology, they may be afraid of looking foolish, and they may choose not to vote."