Sign In

Communications of the ACM

Communications of the ACM

Making Knowledge Work in Virtual Teams

When you rely on technology to get work done, as in large, global, distributed information systems projects, collaboration is going to break down at some point during the process. The basis for this breakdown can usually be traced to people and technology. When this happens, many project managers and team members in other leadership positions use technology facilitation to improve their team's collaboration; that is, they take actions to enable their team's more effective use of information and communication technologies (ICTs).

Technology facilitation has been an important, yet neglected topic for many years [7] about which we know little. Still, its importance seems to have increased as work has become increasingly computer-mediated. Whether involved in virtual teams focused on global, distributed projects or not, most knowledge workers currently spend most of their time working virtually through the ICTs, but do not take advantage of the benefits available—a key area for productivity improvement [2] that technology facilitation addresses.

These ICTs, from email and fax to complex collaborative integrated development environments and group decision support tools, can enhance teamwork performance, but convincing team members to use them effectively remains an ongoing challenge [3]. The fact that productivity using ICTs suffers even when collocated, suggests that the additional communication challenges imposed by virtuality, global distribution of teammates, and co-work on projects involving substantial task interdependence will heighten the difficulty of collaboration and need for specific, efficient intervention techniques for resurrecting interaction when it fails.

Here, we report findings of a study that addresses this need by isolating how virtual team (VT) leaders in the IS industry persuade their teams to effectively use ICTs through technology facilitation during team interaction. We know of no prior field study examining this topic.

Whether involved in virtual teams focused on global, distributed projects or not, most knowledge workers currently spend most of their time working virtually through the ICTs, but do not take advantage of the benefits available—a key area for productivity improvement that technology facilitation addresses.

We captured the moments of interaction breakdown and what was done to fix them in order to analyze their elements and isolate the specific technology/ICT interventions/changes that leaders were making. To do so, we applied critical incident technique (CIT) interview methodology—a practice applied in thousands of studies in the field of industrial and organizational psychology, but rarely used in IS research [4]. It enables a focused and in-depth capture of intact job behavior and its context [1, 5].

We conducted CIT interviews with 13 practicing VT leaders or project managers with experience in more than 20 organizations including more than half of the top outsourcing firms according to Information Week [6] to collect 510 pages of transcribed data in this study [8]. The interviewees were selected for their high performance and they had an average of five years experience leading virtual teams. They reported successful and unsuccessful technology facilitation critical incidents from 30 IS projects. These incidents were occasions when the leaders took action to improve the use of ICTs within their teams and the leader could point to a demonstrable impact on team use that led to definite positive or negative changes in team effectiveness.

The questions in these interviews were guided by an adaptive structuration theory (AST) framework that helped specify the general areas for questioning. AST explains how groups come to appropriate changes. Our focus was on changes related to ICTs and their features, but they may be related to team members' roles or behaviors. Changes might also relate to task methodologies. Changing one thing, such as ICT, in an intervention may require changing others simultaneously, and changing one may lead to a need to change another later. As a result, understanding the technology interventions leaders made required a carefully constructed and pilot-tested, two-hour interview protocol that tapped all of these areas and possibilities.

The leader in the facilitations sensed an opportunity to improve collaboration by changing the way the team members were using ICTs and then took one or more actions to make changes. We report best practices for technology facilitation found in this study. To be sure, we also found that high-performance VT leaders do not always follow these best practices, often resulting in failure. We would expect more problems with lower-performing or less experienced VT leaders.

Back to Top

Current IS Projects

We collected data on 52 incidents of technology facilitation in 30 projects. Many of the projects involved outsourcing. Longer projects lasted two to five years with a median monthly budget of approximately $625,000. The project work encountered fully represented the IT industry. Eleven projects involved the development of new software systems; six involved implementing large software packages, five of which were enterprise resource planning (ERP) systems. Another six projects involved broad change initiatives involving IT changes, sometimes in legacy systems as well as new system development, along with change management or restructuring initiatives that went beyond IT.

Tremendous volatility characterized these projects. Team membership sizes fluctuated over the course of projects. Size would often begin small (5–12 members), grow to perhaps 60 members and five organizations during development, cycle through several additional increases or decreases, and end with a new group added for implementation. More volatility came from competing commitments. Indeed, interviewees reported stoppages of even one day becoming major issues due to tight project timelines.

On top of task and team member volatility and time pressure, the quantity, variety, and interoperability problems of ICTs being used led to distinct interaction challenges for these virtual teams. Interviewees reported numerous work stoppages resulting from problems associated with technology use. Clearly, the research demonstrated that interoperability and coordination problems must be addressed by vendors. Some have begun to do so, as evidenced in products such as Microsoft Communicator and the Team Foundation Server included with the Visual Studio IDE.

Teams used a median of 12.5 ICTs in each project. Figure 1 shows the percentage of teams using a particular ICT. Several were "comfort technologies" available and used by all teams—email, phone, and audio conferencing—while others were less common. These included instant messaging (73%), integrated development environments (77%), more complex tools, and tools used by only one or a few members of a team. Some technology trends showed up as "other" ICTs, such as two teams using wikis (7%), several using customized open-sourced applications, some blog-like RSS integration, and experimentation with integrated systems development modeling ICTs with code round-tripping and project management milestone tracking features.

Given the volatility in team, task, and technology and time pressure, team interaction frequently proved difficult. Since technology is the major communication mechanism in virtual teams, these project teams had to adapt their technology use when problems arose or when opportunities to improve presented themselves. The team leaders used technology facilitation to make these adaptations.

Back to Top

The Technology Facilitation Process

Our technology facilitation interview findings are summarized in Figure 2. Interviewees answered questions about technology facilitation: its triggers, the actions leaders took, the changes that immediately resulted from those actions, the impact these changes had on project outcomes, and the beliefs knowledge team leaders held throughout this process. We then coded that data and analyzed it following the CIT methodology to identify the specific types of triggers, leader actions, and emergent changes common among the 52 incidents.

VT leaders engaged in monitoring actions constantly, leading to trigger identification. Next, VT leaders took initial actions to cause changes, and often they engaged in follow-up rounds of supporting actions. The oval-shaded area in the center of the model highlights the scope of a technology facilitation intervention. When successful, technology facilitations resulted in a variety of project outcomes including saving troubled projects, making budgets, innovating company processes, higher-quality products, and higher client satisfaction.

In one case, a leader came into an ailing project involving multiple organizations, including some offshore. The previous leader could not handle the senior staff in the other organizations. The new leader spent time assessing the situation, identifying the following change triggers: tool inadequacies (too much reliance on email), information visibility problems (shared task information could not be accessed easily), internal group structure problems (dispersion and team size made email unworkable as the main information sharing device), and cooperation problems (private communications between members that should have been shared and differing views on task information led to conflicts).

His foremost technology change was blocking the use of the project management tool and centralizing all the task information in an Excel spreadsheet and placing that spreadsheet in a shared team space where all members could view it anytime and update their portions. He also set some new rules and engaged in persuasion actions to encourage open communication. Participation and information processing capacity improved. He attributed this seemingly simple solution with resolving the conflicting data trouble, improving members' accountability to each other, improving morale, and saving the project from failure.

Figure 2 depicts a composite model drawn from the analysis of all the leaders' data. Hence, not all the leaders had this complete model. However, our analysis of the data indicates that the more complete their model, the more successful they were. Thus, this model provides a means to train VT leaders as well as to guide future research efforts.

Back to Top

Emergent Participation and Information Processing Capacity

Participation and information processing capacity formed an unexpected additional layer at the core of how the VT leaders in this study viewed team interaction. Through the data analysis and coding process we identified these two categories of higher-level change and three dimensions of each. Participation centers on the team members: How they cooperate or express their willingness to work together, how they coordinate information flows or are able to know when to send what, and how they communicate or are able to transmit messages between each other. Information processing capacity centers on the information at the core of work:

  • Visibility, or having accurate, accessible information in shared information repositories,
  • Manipulability, or having information that can be jointly visualized and manipulated simultaneously by dispersed team members, such as jointly rotatable 3D models or even screen-shared spreadsheets in which members share mouse movement and the ability to do "what-if" analysis, and
  • Exchangeability, or having information in forms that can easily transfer intact virtually from member to member.

The appropriation support actions focused on these higher-level changes, and we found that focusing on participation and information processing capacity as outcome goals was key to effective change.

Back to Top

Recognizing Triggers' Impact on Interaction

Focusing on participation and information processing capacity, the leaders could catch problems material to the team's interaction before they became seeds of major failure. We found five varieties of problem triggers (noted in the accompanying table) and one class of opportunity triggers. Multiple problem triggers existed in many incidents.

Only 12 of the 52 (23%) technology facilitation-critical incidents had an opportunity focus, meaning a focus initiated without something being wrong that was affecting interactions within the team. In these opportunity triggers, the leaders addressed some of the same topics as defined by problem triggers, such as how to improve tool capabilities (see the table); however, there was no problem present. They were just focused on making improvements. The general attitude we found was "if it's not broken, don't fix it." In a notable exception, a leader proactively convinced the team to acquire and install an e-meeting tool that enabled a critical client meeting to include the larger team without overwhelming the client. This proved to be a huge success, requiring training and pilot-testing the new ICT in mid-project. Seizing such opportunities leads to improvements in participation or information processing capacity while avoiding the setbacks in team interaction. VT leaders should be trained to spot such opportunities.

Back to Top

Taking Primary and Secondary Actions

Our data suggests two classes of leader actions. Leaders directly manipulated technology, task, and people structures. Such manipulations included installing and adding new ICTs, reconfiguring ICTs, blocking use of ICTs, changing rules and roles within the team, altering the task schedules and content, acquiring additional human resources, firing staff, and a variety of other actions. These direct actions led to low-level changes; that is, the establishment of changes prior to their actual usage and appropriation by team members. In addition to direct manipulation, we found successful leaders took supporting actions that promote appropriation of the new changes.

In one case, a leader found the collaboration portal for the team was causing several problems, particularly performance and access problems. Information could not be found. He developed a consensus that this tool was inadequate, assigned a team member to research new tools, acquired new portal software, developed norms for using this new portal, trained the team on it, assigned a member to administer it, and mandated its use. Team interaction improved with the new tool in place, leading to less stress, better morale, more efficiency, improved communication, and better information processing capacity.

This leader's actions lead to initial, emergent, low-level changes in technology, tasks, and people in the team. Over time and with the aid of additional appropriation support actions, these low-level changes led to higher-level changes in participation and information processing capacity. When introducing a new ICT to the team, direct manipulations hinged on two rounds of supporting actions for success, one in which the low-level changes took place and a second round to support higher-level changes in interaction enabled through the introduction of the new technology structure ().

The supporting actions used by leaders were initializing, persuading, training, and enforcing rules. The initializing actions involve establishing team interaction, designing the communication plan to clarify how and why to use the ICTs in the team's toolkit, setting up the ICTs, and being the first to use the ICTs (especially complex, new ones). Only 17 of the 30 (57%) projects included a formal orientation process that included initializing actions. This is a significant, missed opportunity, particularly in large projects with high pressure and members from multiple organizations with a variety of ICT use habits. Several reported problems traced to failure to take charge and perform initializing actions during the ramp-up phase of projects.

Training and persuasion served a critical role. We found effective leaders did provide these supporting actions to ensure their facilitations' success, and they were able to do it in the dispersed team setting. Training spanned a variety of possibilities. Formal trainings were often at the beginning or early stages of the project, but we also had three instances where team interaction was halted and everybody was flown in for a formal (mid-project) meeting. Formal training need not be colocated, and we determined that the occasions where project progress stopped in order to fly in members indicated a technology facilitation failure. Informal training included ongoing modeling of effective use of the ICTs and reminders of how to use the ICTs as well as using other team members as ICT trainers with other members having trouble.

Training efforts provided a platform to convey both how to use the ICTs and how these technologies would be useful to each team member, the team as a whole, and the organization. Persuasion required soft skills—discussing the use of the ICTs with the team to convince them the effort to use the tools outweighs the costs, establishing expectations of regular use of the ICTs during work, and reminding members of the motivation for using the technologies. Team members were not automatically convinced of the need to use new tools or change their use of existing tools, simply because leaders told them it was necessary. Only 25 incidents (48%) included training or persuasion actions. Training, even informal, offers an ideal setting for more easily setting up the base motivation for making ICT use changes. Not using training actions appears to be a serious mistake.

The more organizations involved in a project and the more physically distant, the more likely entering team members will have different preferences and habits regarding even the simplest and most comfortable ICTs: the phone and email. Leaders reported problems originating even from these two. Phones may be shared among cubicle-mates and therefore unavailable except with careful planning. Several leaders reported email breaking down in teams with more than five or seven people due to overload, problems tracking down correct versions of attachments, or inability to send large enough files. When even the most comfortable ICTs breakdown, the leader must create better participation and information processing capacity options. Overcoming members' high comfort level requires persuasion. Successful VT leaders were able to persuade dispersed team members from multiple organizations over whom they had little power or authority. This ability critically enabled technology facilitation success in many cases.

We must train VT leaders to be successful and more proactive in seizing opportunities before problems disturb or halt team interaction.

Back to Top

Final Thoughts

Technology facilitation is a heightened priority for IS project managers working in today's virtual contexts. They must learn to recognize the triggers, shift their focus to improving team interaction, and effectively take action, so that team productivity can be maximized. This study defined triggers, actions, and emergent changes found in technology facilitation and suggested possible relationships they have with success or failure. We must train VT leaders to be successful and more proactive in seizing opportunities before problems disturb or halt team interaction. Future efforts can take the output of this study and develop assessment instruments and training to help identify technology facilitation strengths and weaknesses within individual leaders to allow for more targeted training.

With better technology facilitation, team members can spend more time enjoying what they do, and less time under stress and working late-nights or weekends due to missed deadlines and failed VT interaction.

Back to Top


1. Borg, W.R. and Gall, M.D. Educational Research: An Introduction (4th ed.). Longman, New York, NY, 1989.

2. Davenport, T.H. Let them all be power users. The HBR List: Breakthrough Ideas for 2005. HBR, Cambridge, MA, Feb. 2005, 41–42.

3. Easley, R.F., Devraj, S., and Crant, J.M. Relating collaborative technology use to teamwork quality and performance: An empirical analysis. J. MIS 19, 4 (2003), 247–268.

4. Fivars, G. and Fitzpatrick, R. The Critical Incident Technique Bibliography; (accessed May 16, 2004).

5. Flanagan, J.C. The critical incident technique. Psychological Bulletin 51, 4 (1954), 327–358.

6. McDougall, P. Outsourcers fall short. Information Week. Nov. 22, 2005, 45–55.

7. Niederman, F., Beise, C.M., and Beranek, P.M. Issues and concerns about computer-supported meetings: The facilitator's perspective. MIS Quarterly 20, 1 (1996), 1–22.

8. Thomas, D.M. The Team Leader Technology Facilitation Role in Information Systems Project Virtual Teams. Unpublished dissertation, University of Georgia, Athens, GA, 2005.

Back to Top


Dominic Thomas ( is a faculty member in the Goizueta Business School at Emory University, Atlanta, GA.

Robert Bostrom ( is the Edmund L. Rast Professor of Business in the Terry College of Business at the University of Georgia, Athens, GA.

Marianne Gouge ( is a freelance IT researcher working in HCI, Oahu, HI.

Back to Top


F1Figure 1. Prevalence of ICTs used by teams studied.

F2Figure 2. The nature of successful technology facilitation interventions.

Back to Top


UT1Table. Problem triggers.

Back to top

©2007 ACM  0001-0782/07/1100  $5.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2007 ACM, Inc.


No entries found