Academy ol Management Executive, 2002, Vol. 16, No. 3

Five challenges to virtual teamsuccess: Lessons from

Sabre, Inc.

Bradley L Kiikman, Benson Rosen, Ciistina B. Gibson, Paul E. Tesluk, and Simon O. McPheison

Executive SummaryAdvances in communications and information technology create new opportunities for

organizations to build and manage virtual teams. Such teams are composed of employeeswith unique skills, located at a distance from each other, who must collaborate toaccomplish important organizational tasks. Based on a comprehensive set of interviewswith a subset of team members, team leaders, general managers, and executives on 65virtual teams at Sabre, Inc.—an innovative organization in the travel industry—weidentify five challenges that organizations can expect to encounter in establishing,maintaining, and supporting virtual teams, e.g., building trust, cohesion, and teamidentity, and overcoming isolation among virtual team members. Both leaders andmembers of virtual teams face particular difficulties in selecting team members whohave the balance of technical and interpersonal skills and abilities required to workvirtually and in evaluating the performance of individuals and teams working in virtualspace. Examination of Sabre's strategies for coping with each challenge should beinstructive to other organizations using or considering virtual (earns.

Off the coast of Mexico, a team of five peoplestruggles to stay afloat on a raft they assembled onshore. Waves crash around them, their raft beginsto tip over, and two members fall into the sea. Athird member helps the submerged members backonto the raft. Finally, the raft is righted and theteam paddles furiously onward.

While this might sound like a scene from a real-ity TV show, these events are actually part of team-building at Sabre, Inc. for its virtual teams. Thecrashing waves symbolize unanticipated andrapid change, the construction of the raft from ev-eryday materials demonstrates creativity and re-sourcefulness, and the entire exercise shows thatyou either sink or swim as a team. Just one yearearlier, virtual team members at Sabre had spentthree days in face-to-face team-building activitiesdesigned to launch a division-wide virtual teamsinitiative. This raft exercise reinforced the knowl-edge and skills learned in earlier teambuilding.We examine the challenges of building and man-aging virtual teams and present five importantlessons learned from Sabre's experience.

Virtual Teams: Origins and Trends

While work teams were used in the U.S. as early asthe 1960s, the widespread use of teams and qualitycircles began in the Total Quality Managementmovement of the 1980s. In the late 1980s and early1990s, many companies implemented self-manag-ing or empowered work teams. To cut bureaucracy,reduce cycle time, and improve service, line-levelemployees took on decision-making and problem-solving responsibilities traditionally reserved formanagement. By the mid-1990s, increasing num-bers of companies such as Goodyear, Motorola,Texas Instruments, and General Electric had be-gun exporting the team concept to their foreignaffiliates in Asia, Europe, and Latin America tointegrate global human resource practices.' Now,due to communication technology improvementsand continued globalization, virtual teams haveincreased rapidly worldwide.

Virtual teams are groups of people who work in-terdependently with shared purpose across space,time, and organization boundaries using technologyto communicate and collaborate.^ Virtual team mem-

67

Academy of Management Executive August

bers may be located across a country or across theworld, rarely meet face-to-face, and include mem-bers from different cultures.^ Many virtual teams arecross-functional and emphasize solving customerproblems or generating new work processes.'' Virtualwork allows organizations to combine the best ex-pertise regardless of geographic location.^

Due to employee travel restrictions resulting fromthe 2001-2002 recession and the events of September11, 2001, virtual teaming will likely increase expo-nentially. This increase will parallel that of telecom-muters, or employees who work from remote loca-tions and communicate electronically.^ Two out otthree Fortune 500 companies currently employ tele-commuters.' The United States Labor Department re-ported that 19 million people worked from home on-line or from another location in 2001,̂ and theGartnerGroup estimated that by 2002 over 100 mil-lion people worldwide will be working outside tradi-tional offices.̂ While many organizations have em-braced virtual teaming, little is known, beyondanecdotal evidence and conventional wisdom, aboutwhat makes virtual teams work or even how theydiffer from face-to-face teams.

Virtual teaming -will likely increaseexponentially.

Anthony Townsend and colleagues (August 1998)first discussed in detail the concept of virtualteams for The Executive.^^ The authors defined vir-tual teams, specified why they have become pop-ular, discussed communication technologies, andprovided preliminary guidelines for building vir-tual teams. More recently in The Executive (August2000), Wayne Cascio examined virtual workplacesmore generally, discussing the disadvantages ofvirtual teaming and methods for training teammembers and leaders." Building on and extendingthis work, we examined specific challenges en-countered by virtual team leaders and members.Much of the previous virtual-teams research ema-nated from anecdotal evidence or case studies.'^We comprehensively studied a large number ofcross-functional virtual teams in a high-technol-ogy company to challenge the prevailing conven-tional wisdom regarding virtual teams. To extractimportant lessons, we drew on our experienceswith 65 cross-functional virtual teams at Sabre.

Sabre, Inc: Business at Internet Speed

In 1960, Sabre began as the computerized reservationsystem of American Airlines and is the inventor of

electronic commerce for the travel industry. In March2000, Sabre spun off from AMR (the parent companyof American Airlines) and became 100 percent pub-licly traded. Headquartered in Dallas/Fort Worth, thecompany currently employs over 6,000 employees in45 countries. Sabre processes over 400 million travelbookings annually (40 percent of the world's travelreservations) and is used by over 60,000 travel agentsin 114 countries. Sabre also owns Travelocity.com,the world's leading online business-to-customertravel site; and it owns GetThere, the world's leadingsupplier of Web-based business-to-business travelreservation systems. Sabre's competitors include:Galileo (owned by Cendant Corporation), Worldspan(owned by Delta, Northwest, and TWA), and Ama-deus (majority owned by Air France, Iberia, andLufthansa). Major competitors oi Sabre's Travelocity-.com include Expedia.com and Orbitz.com.

Our research focused on executives, vice presi-dents, and virtual team leaders and members inSabre's North American Sales and Service, Opera-tions, and Financial Services Division. Sabre's 65virtual teams are cross-functional, based in theU.S. and Canada, and often span several states orprovinces. With over 500 members, they averageabout eight members per team. Virtual team mem-bers are located in both field and employee homeoffices and in the company's Texas headquarters.On each team, account executives sell reservationsystems, field service technicians install systems,training representatives teach travel agents howto use the systems, installation operations coordi-nators schedule installation and training appoint-ments, account management specialists handlecustomer billing and collection, and customer ser-vice representatives field inquiries throughout theprocess. Clearly, Sabre's virtual teams are highlyinterdependent. To coordinate activities, memberscommunicate using e-mail, telephone, video con-ferencing, and Web-based conferencing.

Sabre switched from functionally based workteams to market-based, cross-functional virtualteams in 1999 to integrate different functions toimprove customer responsiveness. Functionalsilos were limiting Sabre's ability to satisfycustomers. For instance, from 1996 to 1998, Sa-bre's North American customer satisfaction rat-ings fell from a 79 percent satisfaction rate to 68percent, while competitor ratings remained un-changed. The purpose of virtual teams was tostrengthen customer focus to increase productiv-ity, market share, and profitability.

Cross-functional virtual teams represent a spe-cific, albeit common, type of virtual team. Indeed,there are many types of virtual teams, each pre-senting unique management challenges. For ex-

2002 Kirkman, Rosen. Gibson. Tesluk, and JVlcPherson 69

ample, global virtual teams must overcome cul-tural and communication barriers.'^ Virtual teamsassigned to accomplish specific projects oftenhave high start-up costs. Some research on cross-functional teams has shown that as teams becomemore cross-functional, both positive team pro-cesses (e.g., information sharing, team task andstrategy agreement, and flexibility) and outcomessuch as unit performance decrease.''' The chal-lenge for Sabre was to recognize obstacles con-fronting teams that are both cross-functional andvirtual. Lessons learned from Sabre should applyto cross-functional virtual teams and other virtualteams that create synergies based on the speciaiexpertise of members in distant locations.

Functional silos were limiting Sabre'sability to satisfy customers.

The Dimensions of Virtuality

Sabre's virtual teams are only moderately, and notcompletely, virtual. A virtual relationship is one thatis primarily conducted using technology, but virtualteams vary in the degree to which this is the case.^^In fact, degree of virtuality is a complex multidimen-sional construct.'^ One dimension of virtuality is theproportion of time that team members work face-to-face compared to virtually. A second dimension isthe proportion of team members at any one location.Along this dimension, the highest degree of virtual-ity would be when all members work at distant lo-cations. A third dimension is the proportion of timemembers devote to a virtual team compared to timespent on other duties. In some instances, individualsmay work only a few hours a month on a virtual teamproject, while devoting most of their time to activitiesunrelated to virtual work.

Sabre's virtual teams meet face-to-iace only once ayear. While about 15 percent of the members work atthe same location, most teammates work at distantlocations. Virtual team activity is the primary focusof each member. Accordingly, our findings could ap-ply to other cross-functional virtual teams.

Five Challenges of Virtual Teams

From our interviews with over 75 executives, teamleaders, and team members, we uncovered some sur-prising insights about meeting the challenges ofmanaging and working in virtual teams (we summa-rize our interview methodology in the Appendix). Foreach of the five challenges, we first present the con-ventional wisdom (i.e., what consultants and re-searchers are saying), we then highlight Sabre's in-

novative responses, and finally we extract thelessons learned from Sabre that should help otherorganizations using virtual teams. Table 1 presents asummary of conventional wisdom, virtual team chal-lenges, and lessons learned.

Challenge 1: Building Trust Within Virtual Teams

Conventional wisdom: Most consultants and re-searchers agree that building trust is the greatestchallenge in creating successful virtual teams andorganizations.'^ Trust has been called the glue ofthe global workplace.'^ As Charles Handy, authorof "Trust and the Virtual Organization," stated.

Most of our organizations tend to be arrangedon the assumption that people cannot betrusted or relied upon, even in tiny matters.... Itis unwise to trust people whom you do not knowwell, whom you have not observed in actionover time, and who are not committed to thesame goals Trust needs touch . . . high techhas to be balanced by high touch to build high-trust organizations. Paradoxically, the more vir-tual an organization becomes, the more its peo-ple need to meet in

In Wayne Cascio's Executive article on working vir-tually, he stated, "Lack of trust can undermine everyother precaution taken to ensure successful virtualwork arrangements."20 Furthermore, Sirkka larven-paa and her colleagues have stated, "In virtual or-ganizations, trust requires constant face-to-faceinteraction—the very activity the virtual form elimi-nates."^' The conclusion from conventional wisdomis that trust is very difficult to build and requiresfrequent face-to-face interaction. Thus, a specificchallenge for virtual teams, compared to face-to-faceteams, is the difficulty of building trust betweenteam members who rarely, or never, see each other.

Our findings at Sabre, however, question thisconventional thinking. Consider the following in-terview quotes about trust;

When you are working with people you neversee, you can develop trust, but you must re-spond to that person. Follow through. If youtell them you are going to get back to a cus-tomer, get back to them.

(Dallas/Fort Worth team member)

I think trusting someone in a virtual team islinked directly to their work ethic. It is taskfirst. The trust has been built through thetask-based relationship that has evolved.

(Account executive)

70 Academy of Management Executive August

Table 1Conventional Wisdom, Virtual Team Challenges, and Lessons Learned from Sabre

Conventional Wisdom Virtual Teana Challenge Lessons Learned from Sabre

Building trust in virtual teams isextremely difficult, given the limitedface-to-face interaction.

Establishing trust based onperiormance consistency ratherthan social bonds.

Rapid responses lo virtual teammates ioster trust.Establishing norms around communication patternsis key.Team leaders play important roles in reinforcingtimeliness and consistency of team interaction.Levels of trust based on performance compensateior lack oi social interaction.

Virtual teams will struggle withcreating synergy.

Overcoming group-process lossesassociated with virtual teams.

Extensive training in virtual teamwork helpsovercome process loss. Training in virtual teamleadership, conflict management, and meetingsmanagement is particularly valuable forovercoming process loss.Adaptation of decision-making software facilitatesproblem solving and decision-making.

Virtual team members experienceisolation and detachment.

Creating a virtual environment ofinclusiveness and involvement.

• Consider individual differences in preferences forworking virtually when selecting virtual teammembers.

• Give virtual team members a realistic preview oithe potential for feeling detached.

• Team leaders play a critical role in maintainingcontinuous contact with remotely situated virtualteam members.

• Redesign job assignments to provide virtual teammembers with occasional face-to-face customercontact to reduce isolation.

• Convene iace-to-iace meetings for virtual teammembers at company-sponsored conferences.

Because of the need to communicate Identifying virtual team membersvia information technology,selection of virtual team membersoveremphasizes technical skills andunderemphasizes interpersonal andteamwork skills.

who have a healthy balance oftechnical and interpersonalskills.

Use behavioral interviewing techniques andsimulations as part of the selection process.Use panels of current virtual team members io helprecruit and select new team members and ensurethe appropriate balance of technical andinterpersonal skills. The panel approach has theadditional benefit oi building support andfacilitating socialization oi the newly selectedvirtual team member.

Assessment and development ofvirtual team members is verylimited in the virtual teamenvironment.

Establishing the appropriatequantitative and qualitativedata ior accurate assessment oivirtual team members.

Developing creative approachesfor providing feedback,coaching, and support iorvirtual team members.

• Use oi a comprehensive "balanced scorecard"approach provides valuable quantitative data onteam performance.

• Monitor group communication archives to assesssubjective iactors, including idea generation,leadership, and problem-solving skills.

• Use team-member peer reviews to assesscontributions to team eiiectiveness.

• Use "richer" communication media, including videoconferencing, for performance evaluation feedback.

• Identify on-line training and development resourcesto address virtual team members' knowledge, skills,and abilities in need oi further improvement.

You gain the trust in people when they deliverwhat they promise, when all are contributingto the same idea and goal. I think that on avirtual team you start trusting each otherwhen you start meeting those results and ev-erybody has their role within the team and

knows what their responsibility is and takesownership to achieve results.

(Canadian team member)

We concluded that trust can be built virtually anddoes not require face-to-face interaction. The key

2002 Kirkman. Rosen, Gibson. Tesluk. and McPherson 71

issue is the type of trust developed. For example, inface-to-face teams, members trust their teammatesafter spending time with them, sharing meals, dis-cussing personal matters, or socializing outsidework. People trust others when important informa-tion they share stays confidential. Researchers callthis type of trust benevolent or interpersonaltrust.^^ In contrast, we found that trust in virtualteams grows through team member reliability,consistency, and responsiveness when dealingwith teammates and customers, or what is knownas ability-based or task-based trust.^^

Lessons learned: One lesson learned is that build-ing trust requires rapid responses to electronic com-munications from team members, reliable perfor-mance, and consistent follow-through. Unlike face-to-face teams, where trust develops based on socialbonds formed by informal chats around the watercooler, impromptu meetings, or after-work gather-ings, virtual team members establish trust based onpredictable performance.^-'' Accordingly, team lead-ers should coach virtual team members to avoid longlags in responding, unilateral priority shifts, and fail-ure to follow up on commitments.^^ A team charterthat explicitly identifies important types of teammember behaviors (e.g., responding to all e-mailsfrom team members within 24 hours) is a comple-mentary strategy that leaders may use to developtrust among virtual team members. These actionswill build strong team norms about types of behaviorthat foster

Unlike face-to-iace teams, where trustdevelops based on social bonds formedby informal chats around the watercooler, impromptu meetings, or after-work gatherings, virtual team membersestablish trust based on predictableperformance.

Upon first glance, the challenge of building trustmay seem typical for any team or organization.However, building trust is a unique challenge forvirtual teams because managers cannot rely onpast methods of trust-building based on social in-teraction, face-to-face meetings, and direct obser-vations of fellow team member commitment. Work-ing virtually magnifies and exacerbates trustissues confronting almost all teams. The need for aproactive approach to identifying and facilitatingtrust based on speed, consistency, and responsive-ness of virtual team members is the first importantlesson learned from Sabre.

Challenge 2: Maximizing Process Gains andMinimizing Process Losses on Virtual Teams

Conventional wisdom: Many researchers and con-sultants believe that the group-process gains (posi-tive synergy) produced in face-to-face teams aremore difficult to obtain in virtual teams and thatprocess losses (negative synergy) are more likely.̂ ^Cascio stated, "The major disadvantages of virtualteams are . . . [that] the synergies that often accom-pany face-to-face communication [are \os\]."^^ In dis-cussing telecommuting, Nancy Kurland and DianneBailey said, "Managers may find it difficult to createteam synergy and to overcome the absence of infor-mal, interactive learning."^^ Beverly Geber statedthat in order to convince executives to incur the ex-pense of getting virtual team members face-to-face,"It's best to use the synergy ploy. Remind executivesthat often a company's best ideas are bom out ofchance encounters in a hallway or around a watercooler. Letting virtual team members get togethersometimes for extended water cooler discussions im-proves the chances for serendipity. "̂ ^ Steve Alex-ander quotes a manager as saying, "I think virtualteams are less productive in the sense that you'remissing out on those corridor talks between the salesand the technical people that sometimes bring aboutvery good results . . . it's not as perfect as havingeveryone sit in the same building."^'

This conventional wisdom suggests that generat-ing synergy (and avoiding process losses) is difficultin virtual teams because members rarely interactface-to-face. Thus, another challenge specific to vir-tual, but not face-to-face, teams is creating synergywithout daily physical encounters. In response tothis challenge. Sabre invested in teambuilding aspart of its virtual team launch. Pre-launch classroomactivities included developing team mission state-ments and core values to help members set objec-tives, clarify roles, build personal relationships, de-velop team norms, and establish group identity.Sabre also encourages virtual team members to as-semble once or twice yearly. While pre-launch team-building and scheduled face-to-face meetings can-not fully compensate for the lack of daily informalinteraction, these interventions do help team mem-bers establish a common set of goals, expectations,and operating principles.

To further instill shared purposes and goals, everyteam must complete a business plan outlining itsannual goals and objectives. One team membercommented, "Virtual teams need to understandmuch more so than co-located teams what goal theyare working towards because you are working insuch different areas and, in our case, in differentcountries. It plays a much stronger role if you know

72 Academy of Management Executive August

what your ultimate target is going to be. Everyone isworking toward the same thing." To help teams runeffective virtual meetings, brainstorm solutions, re-solve conflicts, and take action. Sabre relies on con-tinuous virtual team training. Each team completes aseries of CD-ROM training modules developed bythe Hillsboro, Oregon consulting firm The BelgardGroup. Nicknamed Tour de Teams by Sabre, the 15training modules contain exercises and scenariossuch as developing a team charter, managing ateam meeting, resolving conflicts, and selecting newteam members. One team member commented,"When we complete the team training modules, wehave a conference call and go over all of the points.We take them very seriously. So when we have vir-tual meetings, we now have tools to help us stay ontrack and communicate effectively."

Sabre's experience also shows that working virtu-ally can reduce team process losses associated withstereotyping, personality conflicts, power, politics,and cliques commonly experienced by face-to-faceteams. Diversity research shows that visual cuessuch as race/ethnicity and gender can decrease teamintegration and performance in highly diverseteams.^2 Virtual team members may be unaffected bypotentially divisive demographic differences whenthere is minimal face-to-face contact. And, while theresearch is still relatively new, some researchershave found that electronic collaboration generatesmore minority participation, which might increaseoverall integration and level of attachment of minor-ity members.33 Other studies show that electronicgroup decision support systems help virtual teamsmake higher quality decisions than face-to-face

Sabre's experience also shows thatworking virtually can reduce teamprocess losses associated withstereotyping, personality conflicts,power, politics, and cliques commonlyexperienced by face-to-face teams.

Lessons learned: A critical priority for virtualteam leaders is helping their virtual teams maxi-mize process gains and minimize process losses.Sabre's well-trained virtual teams overcome pro-cess losses. Sabre's training begins with team-building and continues with efforts to help virtualteams create charters and mission statements,clarify goals, and develop operating norms. Sabresustains virtual team effectiveness with an ongo-ing virtual training program to build new skills inmeetings management, problem solving, decision-

making, and other team processes. Sabre's com-prehensive training efforts allow virtual teams tocreate synergy by bridging barriers of time andspace and collaborating effectively. Moreover, ste-reotyping, gossip, politics, and conflict are oftenminimized in virtual teams like Sabre's.

Challenge 3: Overcoming Feelings of Isolationand Detachment Associated WithVirtual Teamwork

Conventional wisdom: Regarding isolation invirtual teams, Cascio stated, "The major disad-vantages of virtual teams are the lack of physi-cal interaction—with its associated verbal andnonverbal cues," and "Some level of social inter-action with supervisors and coworkers is essen-tial in almost all jobs. Without it, workers feelisolated and out of the loop."^^ Charles Handyhas stated, "The loneliness of the long-distanceexecutive is well documented. Few are going tobe eager advocates of virtuality when it reallymeans that work is what you do, not where yougo."̂ ^ Telecommuting researchers Kurland andBailey stated, "Probably the most commonly ex-pressed challenge of telecommuting is overcom-ing the isolation caused by the separation of thetelecommuter from the social network of the tra-ditional work space."^' In a separate article, Kur-land and Terry Egan said, "Employees' primaryreluctance about telecommuting rests on con-cerns about isolation," and "Employees commentthat they miss the informal interaction they gar-ner by being in the presence of colleagues andfriends."3s Finally, Paula Caproni warned, "Manyresearchers and practitioners are concerned thathigh-quality relationships may be particularlydifficult to achieve in teams in which team mem-bers are geographically dispersed."^^

This conventional wisdom suggests that virtualteam members will be less productive and satis-fied than people working face-to-face due to feel-ings of isolation and detachment. Thus, a specificchallenge for virtual team leaders that does notconfront face-to-face team leaders is overcomingmember feelings of isolation. Researchers havelong held that people are motivated and satisfiedat work, in part, as a result of interactions withcoworkers."*" Colleagues share stories and pic-tures, have lunch or take breaks together, and cel-ebrate promotions or the birth of children. In face-to-face teams, these activities occur naturally andfrequently and build esprit de corps. Team leadersuse team-building activities like ropes courses,bowling nights, or barbeques to solidify team co-

2002 Kirkman. Rosen. Gibson. Tesluk. and McPherson 73

hesion and spirit. In virtual teams, most of thesepossibilities are lost.

Indeed, some virtual team members have re-ported feelings of alienation.^' Consider thesecomments from Sabre team members:

We get left off a lot of things because thereare meetings we can't go to for cost reasons.We miss out on those opportunities to gettogether and bond as a group, and that istough sometimes. And you do feel like step-children sometimes.

(Dallas/Fort Worth team member)

I find that by working at home, my work is myhome and I miss that interaction. I don't haveas many people to network with on issues orsuccesses. Sometimes I can't reach anyone byphone and it's frustrating. When you work inan office, you just look over their cube andthere they are.

(Account executive)

As much as we want to go and run the worldfrom our bedrooms, in our slippers, we are hu-mans, we have to be touched, seen, and heard.

(Account executive)

Sabre counteracts feelings of virtual team mem-ber isolation first by recognizing individual needsfor social interaction or lack of it. While individu-als with strong social needs may find virtual team-work difficult, others desire independent, virtualwork. Some Sabre employees welcome minimalsocial interaction and reductions in gossip, poli-tics, and minor disruptions that often accompanyface-to-face work. Others tout the advantages ofworking from home including reduced travel time,proximity to young children, and flexibility. In se-lection interviews. Sabre questions candidates ontheir suitability for virtual teamwork. Sabre usesrealistic job previews^^ ^Q counter team-memberisolation, which allows candidates to select out ofisolating positions. Some virtual team membersinteract with customers, partially satisfying socialneeds. Sabre also gives employees options forworking from home or an office where they haveopportunities to interact with other Sabre employ-ees who may, or may not, be virtual teammates.

While individuals with strong socialneeds may find virtual teamworkdifficult, others desire independent,virtual work.

Sabre's team-building and training sessions helpovercome team-member isolation. Most consultantsand researchers believe initial face-to-face meetingsare critical for all team members to build personalrelationships with teammates.'*^ Annual companymeetings and occasional special events provide ad-ditional opportunities for face-to-face meetings. Gen-eral managers frequently communicate with individ-ual team members to reduce feelings of isolation.Since regular face-to-face interaction is not feasible,managers communicate with routine phone calls ore-mails to keep isolated team members in the loop.One Canadian manager encourages the most iso-lated team members to build networks of contactswithin the company and to stay in close communi-cation with headquarters. Other general managershave established mentor-protege relationships giv-ing isolated team members a sense of inclusion. Asone manager noted, "I work constantly to counteractthe out-of-sight, out-of-mind problems with virtualteam members. My goal is to keep everyone fullyinvolved."

Lessons ieained; General managers admittedthat it took some time to recognize how to deal withvirtual team member isolation. Initially, thesemanagers interpreted minimal communication asa signal that all was well. Over time, however,managers recognized that some virtual team mem-bers needed more frequent and some almost dailycommunication. Sabre's experience with isolatedvirtual team members is that feelings of detach-ment and alienation, while possible, can be over-come with careful attention to social needs. Psy-chological testing identifies members with strongsocial needs, realistic job previews shape expec-tations of prospective employees, increased clientcontact and teambuilding meet social needs, andvirtual team leaders proactively reach out to far-flung team members. While isolation can hamperteam functioning. Sabre uses a variety of tech-niques to minimize potential problems.

Challenge 4: Balancing Technical andInterpersonal Skills Among VirtualTeam Members

Conventional wisdom: Since face-to-face interac-tion is minimal, some managers assume that in-terpersonal skills for virtual team members areless important than for face-to-face teams. For ex-ample, Beverly Geber suggests that managersshould "select people who are comfortable sharinginformation and working with computers."'''* An-thony Townsend and colleagues note, "What isdifferent about the virtual team is the amount oftechnical training that is required to empower the

74 Academy of Management Executive August

team member to function in the virtual environ-ment. Training to maintain technical proficiencywill be an important component of any virtualteam member's continuing education program."'*^A manager in Martha Haywood's book ManagingVirtual Teams: Piactical Techniques for High-Tech-nology Managers stated, "I don't care about thisguy's feelings. I want to know when he's going tocall me back."''^ Such sentiments illustrate the lackof emphasis on interpersonal relations.

"/ don'f care about this guy's feelings, Iwant to know when he's going to call meback."

Conventional wisdom assumes that virtual teammembers should be selected almost exclusively fortheir technical skills. After climbing a steep learningcurve. Sabre's experience has been quite different. Adivisional vice president captured the importance ofstriking a balance between technical and interper-sonal skills by stating, "In our hiring in the past, wewere guided by the level of technical skill, but nowwe are more sensitive to the level of interpersonalskills an individual brings to the equation, becausethis is a very key element in how these teams inter-act. We are more sensitive to a well-rounded person.If the work ethic is there and their ability to work withothers is there, we can train them to be very effectiveat their jobs."

We asked team leaders and members to discussideal virtual team member skills. An overwhelmingmajority mentioned ability to communicate as mostimportant. A close second was desire to support ateam and teamwork in general. Team members alsolisted flexibility and adaptability in playing manydifferent team roles. Other members discussed theimportance of giving and receiving feedback. Othersmentioned a sense of humor. One Canadian teammember commented, "Technical job skills are impor-tant, but I tend to look at their ability to be part of ateam, how they adjust to working with others, andtheir people skills." Managers often mentioned work-ing independently, being a self-starter, thinking out-side of the box, and taking initiative. Task-relevantskills were low on their lists. One manager com-mented, "It is not what the job is about. We can teachthem the job. It is the right personality and the abilityto get along with other team members. I don't care ifthey know twenty different kinds of software or not. Iam more interested in how that person is going to fitinto that team."

A significant challenge for virtual team leaders isrecruiting, selecting, and retaining team members

who have a good balance of technical and interper-sonal skills. Clearly, virtual team members musthave financial, marketing, or technology skills tocarry out specific tasks. Moreover, all employeesmust be well versed in using the communicationstechnology necessary to coordinate the efforts of across-functional virtual team. However, Sabre's ex-perience suggests that virtual team members mustalso possess excellent interpersonal skills. In re-sponse to challenges of recruiting and selecting vir-tual team members with the right balance of techni-cal and interpersonal skills, managers at Sabre haveadjusted their selection procedures. Many use be-havioral interviewing and scenario-based questionsto assess communication and teamwork skills. Forexample, a Canadian manager presents this situa-tion: "I will say, 'You haven't seen me for a month.You have been flying around the Northwest Territo-ries. You are out of touch. How are you going to stayconnected to us?'"

A second approach to selecting virtual team mem-bers is panel interviews. Using teleconferences,prospective future virtual teammates interview jobcandidates, and virtual team members assess inter-personal skills and team fit. A secondary benefit isthe extra effort teammates make to welcome andsocialize candidates they have selected. In futureyears, managers may develop a variety of simula-tions to aid virtual team candidate assessment.

Lessons learned: At Sabre, clearly the selectionof virtual team members involves assessments ofboth task and interpersonal skills. Contrary toconventional wisdom, just because team membersseldom interact face-to-face does not mean interper-sonal skills will be less important than task-relevantskills. Indeed, interpersonal skills may be more im-portant as team members attempt to communicateeffectively without relying on traditional non-verbalcues.

Challenge 5: Assessment and Recognition ofVirtual Team Performance

Conventional wisdom: Again, Charles Handy put itbest when he asked, "How do you manage peoplewhom you do not see?" and "We will . . . have to getaccustomed to working with and managing thosewhom we do not see That is harder than itsounds."^' Kurland and Bailey stated, "A major chal-lenge for managers is their inability to physicallyobserve their employees' performance. They ques-tion, 'How do you measure productivity, build trust,and manage people who are physically out of sight?'If a manager can't see her subordinates in action,then she can't note where the employee is strugglingand where he is strong... monitoring and measuring

2002 Kirkman. Rosen, Gibson. Tesluk. and McPherson 75

[employeel performance remain problematic and asource of concern."''s Cascio said, "By far the biggestchallenge is performance management."^^

How do you measure productivity, buildtrust, and manage people who arephysically out of sight?

Conventional wisdom suggests that it is extraor-dinarily difficult for virtual team leaders to assessmember performance and ensure fairness formembers they rarely see. At Sabre, however, vir-tual team evaluations and rewards are a priority.Accordingly, Sabre developed a comprehensive,multi-tiered assessment process. First, Sabre de-veloped a balanced scorecard^o for each team. Atthe organizational level, balanced scorecards typ-ically elevate non-financial measures such as em-ployee or customer satisfaction to the level of im-portance typically held by financial measuressuch as stock price or return on equity. At the teamlevel, Sabre's balanced scorecard consists of:

• Growth (share of the market);• Profitability (costs versus revenue generated for

each travel booking);• Process improvement (cycle time, or the time

required to order and install customer hard-ware); and

• Customer satisfaction (assessed with surveydata collected from actual customers).

Sabre makes extraordinary efforts to monitor eachteam's customer satisfaction including setting an-nual customer service goals. Managers collect sur-vey data quarterly from each team's external cus-tomers. Sabre posts team customer service scoreson its intranet. Members know exactly where theirteam stands relative to other teams. Closely mon-itoring client satisfaction helps Sabre create in-tense customer service focus. Moreover, virtualteam leaders have an external, albeit subjective,basis for evaluating and rewarding virtual teams.One developmental need for virtual teaming atSabre is allowing more team-member input in set-ting customer service goals. Presently, manage-ment still controls this decision-making aspect.

Another important element of the balancedscorecard is process improvement (i.e., cycle time)or the extent to which team members reduce thetime from the placement of an order to the instal-lation of, and training on, a reservation system.Process improvement is an objective measure ofteam learning at Sabre. Thus, Sabre's teams are

responsible for both day-to-day work and continu-ous improvement of their processes and cycle time.

Managers also assess individual team memberperformance. Sabre tracks objective individual per-formance measures such as number of installations,development of new business, number of individualstrained to use a system, accuracy of financial con-tracts, and customer retention. Because these mea-sures are objective and quantifiable, evaluations aremuch less susceptible to stereotyping, favoritism, orother contaminating perceptual biases. Ironically,virtual team member evaluations may be more ac-curate than evaluations of face-to-face team mem-bers.^' Biases induced by demographic differencessuch as race, gender, and age can lower both indi-vidual performance ratings''^ and team ratings.^^Contamination of evaluations by perceptual biasesis less likely when team leaders have extensiveobjective data.

In addition to objective measures, general manag-ers track subtle virtual team member behavior suchas taking leadership roles during virtual team meet-ings, suggesting internal quality improvement strat-egies, coaching new team members off-line, andother intangible actions that enhance team effective-ness. General managers monitor electronic discus-sions, team e-mails, and other team activities. Man-agers often have more accurate records of individualcontributions to virtual teams than they do from in-formal observations of face-to-face teams. They alsosupplement their own evaluations with peer evalu-ations. Using modified 360-degree performance eval-uations, general managers collect peer and evencustomer input electronically.

General managers also provide timely feedbackand resolve performance problems quickly. Somemanagers emphasize choosing appropriate commu-nications media to deliver constructive feedback.They recommend using two-way communicationchannels such as teleconferences so feedback de-livery can be followed immediately by interactiveproblem-solving or counseling sessions. Communi-cating constructive feedback from a distance forcesmanagers to do more research, collect and analyzeall of the relevant facts, and carefully craft mes-sages. Knowing that their ability to resolve mis-understandings is constrained, managers empha-size careful preparation. Moreover, managersidentify ahead of time resources that teams cantap in responding to problems. Other managershold regularly scheduled monthly virtual meetingswith each team member. One manager said, "Ithas to be a two-way street. They have to feel com-fortable being honest and straightforward with meeven with the bad stuff. I find that communicatingelectronically overcomes some of the interpersonal

76 Academy of Management Executive August

issues that might have made me hold back in thepast."

Communicating constructive feedbackfrom a distance forces managers to domore research, collect and analyze all ofthe relevant facts, and carefully craftmessages* Knowing that their abilityto resolve misunderstandings isconstrained, managers emphasizecareful preparation.

Lessons learned: Sabre has built a comprehensiveperformance review system. The balanced scorecardprovides an excellent approach for measuring teameffectiveness. Sabre assesses individual contribu-tions to team effectiveness by monitoring electroniccommunications and systematically collecting datafrom peers and direct reports using 360-degree as-sessments. Performance data provide a solid foun-dation for recognizing and rewarding team and in-dividual performance, developing new trainingprograms to assist virtual teams, and identifying in-dividual team members who can benefit from off-linementoring and coaching. Managers have developedeffective techniques to deliver feedback. As onemanager stated, "Most everyone's work is measuredin the results they produce and through statistics,and it can all be pulled out systematically for eachindividual." In the virtual workplace, team memberscan be judged more on what they actually do ratherthan on what they appear to be doing.

Continuous Improvement at Sabre

While we have focused on positive aspects of Sa-bre's virtual teams, some lessons were learnedonly after trial and error. For example, managerssaid that recognizing the sense of isolation amongvirtual team members took time. Similarly, somegeneral managers reported initial reluctance toprovide strong negative feedback virtually, prefer-ring face-to-face meetings.

Our research identified a variety of other problemsassociated with managing and supporting virtualteams at Sabre. General managers still struggle withfinding effective strategies for empowering virtualteams. Human Resources continues to fine-tune thecontent and delivery of virtual team training. Lastly,Sabre still struggles with the appropriate mix of re-wards for individual contributions and team perfor-mance. While we have emphasized Sabre's positivelessons learned, these lessons were learned after

false starts, stumbles, and reassessments. Moreover,new problems require continuous fine-tuning. Work-ing with similar organizations, we found that thisongoing process of adaptation and adjustment iscrucial to maximize virtual team effectiveness. Per-manent, inflexible programs or policies such as arigid structure or one-shot training do not provideadequate support for collaboration such as virtualteaming, in which members themselves are ex-pected to grapple with uncertainty, innovate, andremain flexible.

One key to promoting ongoing adaptation at Sabreis communities of practice^" (or what Sabre callsCenters for Excellence) where virtual team membersin the North American division and beyond (includ-ing Sabre employees in Latin America, Europe, andAsia) share best practices. The Centers for Excel-lence, started in 2001 in the areas of sales, technol-ogy, training, and operations, are designed to pro-vide innovative process solutions from Sabre'sglobal operations. An excellent example is Sabre's24/7 Learning Cafe. Developed by the technologyCenter for Excellence, the Cafe is an on-line trainingscheduler that allows Sabre employees worldwide toschedule and access virtual training modules suchas "Coaching and Developing Others" and "NewProduct Training." More recently, the operationsCenter for Excellence developed a standardizedproduct-demonstration tool that allows account ex-ecutives, regardless of their experience level or loca-tion, to provide consistent demonstrations to custom-ers of state-of-the-art reservation systems. TheCenters for Excellence allow learning to occur withinand across Sabre's virtual teams despite the conven-tional wisdom that process improvements occur inserendipitous face-to-face encounters in traditionalwork settings. The ability to establish and re-estab-lish equilibrium between changing needs and re-quirements is a critical competency for organizationsutilizing virtual teams.

Have Virtual Teams Improved Sabre's Bottom-Line Results?

Most of the interviewees at Sabre agreed that thetransition from traditional, functional, face-to-faceteams to cross-functional, virtual teams improvedcustomer service. Regarding objective measures, af-ter Sabre introduced cross-functional virtual teams,customer satisfaction ratings improved each yearfrom a low of 68 percent in 1997 to 85 percent in 2000.In addition. North American market share increasedfrom 43 percent in 1997 to 50 percent in 2000. Alsoduring this period. Sabre's number of travel bookingsincreased significantly each year. While we cannotattribute all of these improvements solely to Sabre's

2002 Kirkman. Rosen. Gibson. Tesluk. and McPherson 77

implementation of virtual teams, clearly customershave responded positively to the changes at Sabre,which include using virtual teams.

However, benefits such as improved customer ser-vice are only half of the equation needed to assessvirtual teaming effectiveness. In determining any re-turn on investment, managers must also assess costsof implementing organizational designs. A centralissue for virtual teams is the difficulty of assigningmonetary values to costs that are not easily quanti-fied.̂ ^ These may include opportunity costs associ-ated with internal resources devoted to the team.Several researchers have suggested calculatingcosts of team-member time and support-person timebased on average salary and time spent with virtualteams.^s Data regarding costs and benefits of virtualteaming can then be used to compare different vir-tual efforts using the same metric. The goal of suchan analysis is determining if a virtual team's charteris consistent with a company's bottom-line objec-tives. Given the substantial resources necessary tosupport virtual teams, these are important questionsto address in designing virtual teams and settingthem up for success.

A central issue for virtual teams is thedifficulty of assigning monetary valuesto costs that are not easily quantified.

Virtual Teams: The Way Business Is Evolving

Using Sabre's experiences, we have highlighted fivecritical challenges that organizations face when im-plementing and using virtual teams. As organiza-tions expand globally, the need to tap the talents,experience, and special skills of employees workingin distant locations will increase. Most corporate ex-ecutives predict that technology-mediated communi-cation and virtual teaming will increasingly replacephysical travel. However, creating and supportingvirtual teams is a very difficult assignment. Identify-ing challenges ahead and learning from the Sabreexperience represent a good place to start. In thewords of one Sabre executive, "I think that virtualteams is inevitably the way business is evolving. Weare working hard to get ahead of the curve."

of virtual teams. Our work with other organizations using virtualteams suggests that Sabre's use is representative and comparableto that ol the typical organization. From Sabre's 65 cross-functionalvirtual teams, members of a representative subset of 18 teams (i.e.,varying by division, region, size of customer, and country) wereselected to participate in face-to-face interviews. From those 18teams, a representative subset of 58 team members (34 percent ofthe total of 169 members) was interviewed based on variance injob function, demographics, and organizational tenure. We alsointerviewed the 11 team leaders of these 18 teams (some teamleaders led more than one team), six divisional vice presidentswho supervised the team leaders, and the executive vice presi-dent oi the North American Division (for a total of 76 interviews).No team members or leaders declined to be interviewed.

Separate interview protocols were developed for each of thethree organizational levels, and all interviewees within eachlevel were asked identical questions. Each of the researchers(i.e., the first four coauthors of this paper) interviewed a roughlyequal number of team members and leaders. The researchershad no prior relationship with Sabre or any virtual team mem-bers before the two entities entered into a research partnership.The researchers traveled from New York to California and fromQuebec to British Columbia to meet virtual team members andleaders. Each interview lasted for one hour. The interviewswere tape recorded and in some cases videotape recorded. Fulltranscriptions of each interview were prepared. All of the re-searchers participated in the divisional and executive vicepresident interviews via conierence call. The interviewees weretold that Sabre and the researchers had formed a partnership toexamine the key drivers of, and significant obstacles to. virtualteam effectiveness. All respondents were assured that theirinterview and survey responses were the property af the re-searchers and that only summary data would be returned toSabre. Thus, their responses were confidential. Total time toconduct and analyze the interviews was six months. Samplevirtual team member interview questions included:

• Describe the main differences between the teams you haveworked on before in this company and your current virtualteam.

• Describe the special challenges you have encountered work-ing virtually.

• If you were involved in the hiring of a new member of yourvirtual team, what characteristics would you look for?

• What specific behaviors has your team leader (general man-ager) demonstrated that particularly help the functioning ofyour virtual team?

Regarding data analysis, all researchers read each interviewtranscript and created their own categories and themes. Eachresearcher then collected representative comments under eachcategory. The researchers then met face-to-face to compare thecategories. Discrepancies were resolved, and the researchersagreed upon a consensus set of categories. Representative com-ments were then collated by category. This process allowed usto retain only those themes that were represented by a largenumber of respondent comments.

Appendix

Summary of Methodology Used to Study VirtualTeams at Sabre

We selected Sabre's North American Sales and Service, Opera-tions, and Financial Services Division based on its extensive use

Endnotes

' Far more information about global work team implementa-tion, see Kirkman, B. L., Gibson, C. B., & Shapiro, D. L. 2001."Exporting" teams: Enhancing the implementation and effec-tiveness of work teams in global affiliates. Organizafionai Dy-namics. 30(1): 12-29.

^ Lipnack, J., & Stamps, J. 2000. Virtual teams: People working

78 Academy of Management Executive August

across boundaries with technology. 2"'^ ed. New York: Wiley. Seealso, Duarte, D. L., & Snyder, N. T. 2001. Mastering virtual teams.2"'' ed. San Francisco: Jossey-Bass.

^ Maznevski, M. L, & Chudoba, K. M. 2000. Bridging space overtime: Global virtual-team dynamics and effectiveness. Organiza-tion Science, 11(5): 473-492; and Montoya-Weiss, M. M., Massey,A. P., 8f Song, M. 2001. Getting it together: Temporal coordinationand conflict management in global virtual teams. Academy ofManagement /ournai, 44(6): 1251-1262, for issues involving virtualteams composed of members from different countries.

'' Chase, N. 1999. Learning to lead a virtual team. Qualify, 38(9):76.; and Geber, B. 1995. Virtual teams. Training, 32(4): 36-40; andBell, B. S.. & Kozlowski. S. W. J. 2002. A typology oi virtual teams:Implications for effective leadership. Group & Organization Man-agement, 27(1): 14-49.

^ Gibson, C. B., & Cohen. S. G., forthcoming. VirtuaJ teamsthat work: Creating the conditions for virtual team effectiveness.San Francisco: Jossey-Bass.

^ See Cooper, R. C. 1997. Telecommuting: The good, the bad, andthe particulars. Supemsfon, 57(2): 10-12; McCune, J. C. 1998. Telecom-muting revisited. Management Review, 87(2): 10-16; and Pearlson,K. E., 8f Saunders, C. S. 2001. There's no place like home: Managingtelecommuting paradoxes. The Academy of Management Executive.15(2): 117-128, for more infonnation on telecommuting.

' Cascio, W. F. 2000. Managing a virtual workplace. TheAcademy of Management Executive. 14(3): 81-90; Kurland, N. B.,& Bailey, D. E. 1999. Telework: The advantages and challengesaf working here, there, anywhere, and anytime. Organization a]Dynamics. 28(2): 53-67; and Kurland, N. B., & Egan, T. D. 1999.Telecommuting: Justice and control in the virtual organization.Organization Science, 10(4): 500-513.

^ Pearlson & Saunders, op. cit.^ Elkins, T. 2000. Virtual teams: Connect and collaborate. HE

Solutions. 32(4): 26-32.'" Townsend, A. M., DeMarie, S. M., & Hendrickson, A. R. 1998.

Virtual teams: Technology and the workplace of the future. TheAcademy of Management Executive. 12(3): 17-29.

" Cascio, op. cit.'̂ Maznevski & Chudoba, op. cit.'̂ Boudreau, M. C. Loch, K. D., Robey, D., & Straud, D. 1998.

Going global: Using information technology to advance thecompetitiveness of the virtual transnational organization. TheAcademy of Management Executive. 12(4): 120-128; andMaznevski & Chudoba, op. cit.

*̂ Bunderson, ]. S., & Sutcliffe, K. M., in press. Comparing alter-native conceptualizations of functional diversity in managementteams: Process and performance effects. Academy oi Managementjournal: Dougherty, D. 1992. Interpretive barriers to successfulproduct innovation in large firms. Organization Science. 3(2): 179-202; Lovelace, K., Shapiro, D. L., & Weingart, L. R. 2001. Maximizingcross-functional new product teams' innovativeness and con-straint adherence: A conilict communication perspective. Acad-emy of Management Journal, 44(4): 779-793; Ancona, D. G., & Cald-well, D. F. 1992. Demography and design: Predictors of newproduct team performance. Organization Science, 3(3): 321-341;and Parker, G. M. 1994. Cross-Zunctionai teams: Working with al-lies, enemies, and other strangers. San Francisco: Jossey-Bass.

'̂ Maznevski & Chudoba, op. cit.'̂ Gibson & Cohen, op. cit." See, far example, Coutu, D, 1998. Trust in virtual teams.

Harvard Business fleview, 76(3): 20-21; Jarvenpaa, S., & Leidner,D. 1999. Communication and trust in global virtual teams. Or-ganization Science. 10(6): 791-815; Jarvenpaa, S. L., Knoll, K., &Leidner. D. E. 1998. Is anybody out there? Antecedents of trust inglobal virtual teams. Journal of Management InformationSystems, 14(4): 29-64; Platt, L. 1999. Virtual teaming: Where is

everyone? Journal of Quality & Participafian, September/October: 41-43.; Cascio, op. cit.; and Townsend, op. cit.

'̂ O'Hara-Devereaux, M., & Johansen, B. 1994. Global work:Bridging distance, cuJture, and time. San Francisco: Jossey-Bass; Hart, P., & Saunders, C. 1997. Power and trust: Criticalfactors in the adoption and use of electronic data interface.Organization Science, 8(1): 23-42; and Sheppard. B. H., & Sher-man, D. M. 1998. The grammars of trust: A model and generalimplications. Academy of Management fieview, 23(3): 422-437.

'̂ Handy, C. 1995. Trust and the virtual organization. HarvardBusiness Review. 73(9): 40-48.

°̂ Cascio, op. cit., 83.'̂ Jarvenpaa, et al.. op. cit., 30.

^̂ For a more complete discussion of trust, see Mayer, R. C.Davis, J. H., & Schaorman. F. D. 1995. An integrative model oforganizational trust. Academy of Management Review. 20(3):709-734. For a discussion about the impact of trust on coopera-tion and teamwork, see Jones. G. R., & George, J. M. 1998. Theexperience and evolution of trust: Implications ior cooperationand teamwork. Academy of Management Review. 23(3): 531-546.

^̂ Mayer, et al., op. cit.^̂ Geber, op. cit.. 39.^̂ Gibson, C. B., & Manuel, J.. forthcoming. Building trust:

Effective multicultural communication processes in virtualteams. In C. B. Gibson & S. G. Cohen (Eds.), VirtuaJ teams thatwork: Creating the conditions for virtual team effectiveness. SanFrancisco: Jossey-Bass.

^̂ For more on creating team charters and other team-devel-opment tools and interventions, see Fisher. K., Rayner. S., &Belgard. W. 1995. Tips for teams. New York: McGraw-Hill.

^' The nation of process gains and losses is explained inmore detail in J. R. Hackman's work. See Hackman, J. H. 1987.The design of effective work teams. In J. W. Lorsch (Ed.), Hand-book of organizational behavior. Englewood Cliffs. NJ: Prentice-Hall: 315-345: and Hackman, ]. R. (Ed.). 1990. Groups that work(and those that don't). San Francisco: Jossey-Bass.

^̂ Cascio, op. cit., 84.^̂ Kurland & Bailey, op. cit., 59.^° Geber. op. cit., 39.'̂ Alexander, S. 2000. Virtual teams going global. Infowoild,

22(46): 55-56.^̂ Ancona & Caldweil, op. cit. See also, Lichtenstein, R.. Al-

exander, J. A.. Jinnett, K.. & Ullman, E. 1997. Embedded inter-group relations in interdisciplinary teams: Effects on percep-tions of level of team integration. Journal of Applied BehavioralScience. 33(4): 413-434; and Timmerman, T. A. 2000. Racial di-versity, age diversity, interdependence, and team performance.Smaii Group fiesearch, 31(5): 592-606.

3̂ McLeod. P.. Baron, R., Marti. M., & Yoon, K. 1997. The eyes haveit: Minority influence in face-to-face and computer-mediatedgroup discussion. Journal of Applied Psychology. 82(5): 706-718.

=^Gallupe. R. B.. Bastianutti, L. & Cooper. W. H. 1991. Un-blocking brainstorms. 7ournaJ of Applied Psychology. 76(1): 137-142. See also, Lam, S. S. K.. & Shaubroeck, J. 2000. Improvinggroup decisions by better pooling information: A comparativeadvantage of group decision support systems. Journal of Ap-plied Psychology. 85(4): 565-573.

^̂ Cascio. op. cit., 82, 84.^̂ Handy, op. cit.. 4.•̂̂ Kurland 8E Bailey, op. cit., 61.

3̂ Kurland & Egan. op. cit., 502.^^See Caproni. P. J. 2001. The practical coach: Management

skills lor everyday life. Upper Saddle River. NJ: Prentice-Hall(see specifically Chapter 8. entitled "Diverse teams and virtualteams: Managing differences and distances": 247-287).

*" Emery, F. E. 1959. Characteristics of sociotechnicai systems.London: Tavistock.

2002 Kirkman. Rosen. Gibson. Tesluk. and McPherson 79

•" Geber. op. cit., 36; and Cascio. op. cit., 82.^̂ Philips. J. M. 1998. Effects of realistic job previews on mul-

tiple organizational outcomes: A meta-analysis. Academy ofManagement Journal. 41(6): 673-690.

•*̂ Joinson, C. 2002. Managing virtual teams. HffMagazine.47(6): 69-73.

'̂' Geber. op. cit.. 40.^̂ Townsend. et al., op. cit., 26.••̂ HaywQod. M. 1998. Managing virtuai teams." Practical tech-

niques for high-technology managers. Boston: Artech House.^'Handy, op. cit., 3, 4.^̂ Kuiland & Bailey, op. cit., 59.••̂ Cascio, ap. cit.. 87.^"For more information on the balanced scorecard. see:

Kaplan, R. S.. & Norton, D. P. 1996. Using the balanced scorecardas a strategic management system. Harvard Business Review.74(1): 75-85.

'̂ Alexander, op. cit.^̂ See, for example, Kraiger. K., & Ford, J. K. 1985. A meta-

analysis of rater race effects in performance ratings. Journal ofApplied Psychology. 70(1): 56-65. See also, Pulakos, E. D.. Op-pler, S. H., White, L. A., & Borman, W. C. 1989. Examination ofrace and sex effects on performance ratings. Journal of AppliedPsychology. 74(5): 770-780.

^̂ Baugh. S. G., Graen. G. B. 1997. Effects of team gender andracial composition on perceptions of team performance incross-functional teams. Group & Organization Management.

22(3): 368-383. See also, Kirkman, B. L., Tesluk, P. E.. & Rosen, B.,in press. The impact of demographic heterogeneity and teamleader-team member demographic fit on team empowermentand effectiveness. Group & Organization Management.

*̂' Brown, J. S., & Duguid, P. 1991. Organizational learning andcommunities of practice: Toward a unified view of working,learning, and innovation. Organization Science. 2(1): 40-57.

" See Levenson, A,, & Cohen, S. G., forthcoming. Meeting theperiormance challenge: Calculating ROI for virtual teams. InC. B. Gibson & S. G. Cohen (Eds.), VirtuaJ teams that work:Creating the conditions for virtuai team e//ectiveness. San Fran-cisco: Jossey-Bass; and Levenson, A., iorthcoming. ROI andstrategy for teams and collaborative work systems. In M. Bey-erlein, C. McGee, G. Klein. L. Broedling, & J. Nemiro (Eds.). Thecollaborative work systems Held book. San Francisco: Jossey-Bass/Pfeiffer, for more information about assessing the costsand benefits of virtual teaming.

^̂ Levenson and Cohen, op. cit.

Simon McPherson is a project manager for Sabre Inc. He imple-mented and developed high-performance work teams forSabre's Sales & Service. Installations, and Financial Serviceswork groups. In addition to implementing virtual teams,he oversaw the breakdown of three distinct silo departmentsinto a single customer-centric department and the collocation ofover 200 customer support personnel. Contact: simon.mcpherson@sabre.com.

Bradley Kirkman is associateprofessor oi organizational be-havior in the DuPree Collegeof Management at the GeorgiaInstitute of Technology. He re-ceived his Ph.D. in organiza-tional behavior from the Uni-versity of North Carolina atChapel Hill. His research inter-ests include work-team effective-ness, cross-cultural manage-ment, organizational justice, andorganizational change. Contact:brad.J:irJrman@ mgt.galech.edu.

Benson Rosen is Hanes Profes-sor oi Management in theKenan-Flagler Business Schoolat the University of North Caro-lina at Chapel Hill. He receivedhis Ph.D. in social and indus-trial psychology from WayneState University. His researchinterests include team develop-ment, career management, andinternational ethics. He is anAcademy of Management Exec-utive editorial board member.Contact: ben r03en@unc.edu.

Cristina Gibson holds a joint po-sition as associate professor atthe Center for Effective Or-ganizations and the MarshallSchool of Business at the Univer-sity of Southern Caliiornia. Shereceived her Ph.D. in organiza-tional behavior iram the Univer-sity of California-Irvine. Her re-search interests include teameffectiveness and internationalmanagement. She is an Acad-emy of Management Journal ed-itorial board member. Contact:cgibson@inarshaii. usc.edu.

Paul Tesluk is assistant proies-sor of management and organi-zation at the Robert H. SmithSchaol of Business. University ofMaryland. He received his Ph.D.in industrial/organizational psy-chology from Pennsylvania StateUniversity. His research interestsinclude knowledge managementin teams, organizational innova-tion, and management/leader-ship development from workexperience. Contact: ptesiuJc@rhsmith.umd.edu.