Lindahl, a computer scientist who co-manages the digital initiative unit for the University of Rochester's River Campus Library, hired Nancy Fried Foster two and a half years ago as lead anthropologist and co-manager of the seven-member group.
He says Foster helps her co-workers see problems and solutions that they might otherwise miss. "The values that her profession brings raise the quality of the work," Lindahl says.
The same is true at Stamford, Conn.-based Pitney Bowes Inc., which provides software, hardware and services to help companies manage their flow of mail, documents and packages. Jim Euchner, vice president of advanced technology and chief e-business officer, first worked with an anthropologist about 15 years ago when he was an IT executive at the former Nynex Corp. Based on his successful experience there, he brought the practice to Pitney Bowes in 1999. He says his decision is paying off, since his two anthropologists continually bring unique perspectives to projects.
A different approach
The IT world has a bias that automation is always good. Technologists bring that bias to the drawing table when they design products, and it can sometimes blind them to the true needs of users. Enter anthropologists, who are trained to ask questions about how people work, how they relate to others, which tools they use and which ones they don't. That kind of research allows anthropologists to see the world from users' perspectives.
Although IT anthropologists are far from common, some companies and IT shops are hiring them to provide that insight, which in turn helps technologists develop applications and systems that best meet users' needs. IBM computer scientist Eser Kandogan sums up the relationship like this: A technologist can make a tool usable; an anthropologist can make sure it's used.
Kandogan works with anthropologist Jeanette Blomberg, manager of the people and practices group at IBM's Almaden Research Center in California. Blomberg says she and other anthropologists use surveys, focus groups, interviews and observations to learn what people need to do their jobs.
"People who aren't trained as anthropologists often come to solutions very quickly," Blomberg says. "They don't often take the time to ask, 'Why are people doing it that way?'"
At IBM, Blomberg studied how systems administrators did their jobs and found that they developed their own local tools -- spreadsheets, programs and bits of code -- to help them manage their systems.
Based on Blomberg's observations, Kandogan and others designed a program that facilitated the systems administrators' efforts to develop local tools. Moreover, Kandogan's program contained a collaborative element that enables systems administrators to share their individual tools -- a feature that came specifically from Blomberg's observations.
At the University of Rochester, Foster studied how faculty worked. Her research led to a customized application that allows faculty members to create a page showcasing their work. "At all the other institutions where this is installed, no one has come up with what we have," Foster says. She adds that understanding the faculty's needs was instrumental in the decision to customize.
No boundaries
That dynamic doesn't surprise anthropologist Patricia Sachs, founder and president of Social Solutions Inc., a consulting firm in Half Moon Bay, Calif. "Technologists tend to look at the user and the user's relationships to the technology. It tends to be very task-focused. And the funding for a lot of technology tends to be done by the business unit, so it has boundaries around what it's going to try to fix," she explains. "Anthropologists look at the missing social layer."
That's clear when anthropologist Eleanor Wynn describes her work. As a social technology architect at Intel Corp., Wynn was part of a cross-functional group that studied how employees work together across time and distance. Wynn asked questions such as "How much do you socialize with teammates outside of meetings?" and "How often do you socialize with people sitting around you who aren't your teammates?"
"That's something that might not be asked by a regular IT person but still influences what kind of social aspects are built into a tool," Wynn explains.
As a result of her work, she says, Intel technologists are designing a program prototype for collaborating virtually that enables multiple ways of communicating, such as a shared whiteboard and instant messaging.
"Some of the data she uncovered was key to our design of our environment," says Cindy Pickering, a principal engineer in Intel's IT Collaboration Research Lab.
Still the exception
Despite such successes, though, the use of anthropologists within IT organizations is limited, says Ed Liebow, a senior research scientist at Battelle Memorial Institute, a company in Columbus, Ohio, that develops and commercializes technology and manages laboratories for customers. "It's still exceptional rather than the main pattern," he says. But that could change. Liebow, a past president of the American Anthropological Association, says there's a growing demand in corporate America for practicing anthropologists.
But there are still considerable obstacles to widespread use of anthropologists in IT, Euchner says. He cites two main factors. "It's hard to justify the cost in ROI terms," he says. "The dominant trend -- routinizing the work, offshoring where you can -- drowns out other voices."
In addition, Euchner says many IT departments don't have the kind of environment that would welcome anthropologists. "An IT initiative that starts with the presumption that work is essentially something that needs to be processed, mapped, rationalized and systematized and that the people need to conform with that -- that's really not a place where an anthropologist will have a voice," he says.
Ironically, though, IT departments that work like that are the ones that could most benefit from the insight that anthropologists offer, Euchner says. "What can get lost under those pressures is a focus on how people do work, and that's what anthropologists bring to the table," he explains.
One of those anthropologists is Alexandra Mack, who works in research and development as part of the Advanced Concepts and Technology group at Pitney Bowes. She helps study work processes alongside other team members, including technologists. The team then uses the information to develop new technologies for various markets.
One of her teams recently studied how small businesses handle bulk mailing. As a result of that research, Pitney Bowes is developing a software application that will help them be more efficient in that process.
Bringing such practices to IT fits in with the overall push to align the tech world with the business realm, Mack and others acknowledge. But even given this alignment trend, Sachs says technologists still have limited ability to garner such insight on their own. "It's a very important thing that technologists are being asked to have a broader view, but they [still] see a problem from their frame of reference and they see a solution from their frame of reference," she explains.
But Wynn says even companies that don't hire anthropologists are benefiting from their work, as anthropological tools and approaches used in places such as Intel, IBM and Pitney Bowes bleed out to other companies. "The methods and approaches of anthropology have spread a lot," she says, "and with that, there is a potential for a very large impact."
Sidebar
Anthropologists in IT: The ROI
Studies showing the return on investment that anthropologists bring to IT are hard to come by -- if they exist at all, but Jim Euchner has some figures that proved the value to him.
As vice president of process improvement at Nynex in the early 1990s, Euchner oversaw the deployment of a system designed to diagnose problems with phone lines. To his surprise, workers in the 42 maintenance centers equipped with the new Maintenance Administration Expert (MAX) used it differently from site to site. Some managers loved it, while others hated it, saying it made their work harder.
Euchner was confused by the differences and, at the urging of another manager, hired anthropologist Patricia Sachs to figure out what had gone wrong and help fix it.
With her assistance, he discovered that there were problems with the way people perceived the system, the way management perceived the work and the way people were measured. "But with Pat's help, we were able to tweak the system to make it work differently for different people, so in the end everyone used it," he says.
Euchner estimated that it cost Nynex about US$1 million to develop and deploy MAX, which helped the company save $4 million to $6 million annually. Euchner points out that a healthy chunk of those savings would have been lost if MAX hadn't been used in all 42 maintenance centers.
Convinced that Sachs' insight as an anthropologist could have a significant and tangible impact, Euchner used her services once again in the mid-1990s. At that time, Nynex was losing market share to competitors that were able to provide high-speed data lines much more quickly than it could. Nynex officials wanted to know the reason for the lag.
Sachs led a team of Nynex workers charged with finding some answers. They learned that orders for high-speed lines took weeks to fill because they passed through multiple workers and systems. Sachs worked with Nynex employees to redesign the flow using existing technology.
As a result, Nynex cut the cycle time for orders from 30 days to just three, Euchner says. The company gained market share and cut the costs of filling an order in half.
Pratt is a Computerworld contributing writer in Waltham, Mass. Contact her at marykpratt@verizon.net.