Connecting to Collaborate: An Interview With Rob Cross

We rely on organization charts because they portray the formal structure of power and division of labor in organizations. After a merger or acquisition, an internal reengineering, or simply personnel changes at the top levels of an organization, everyone wants to see the new org chart so they can place themselves in the new environment. We want to know who’s responsible for what and who reports to whom.

At the same time, we also know that org charts don’t give us the full picture. We know organizations work through lines of collaboration that are at least as much horizontal as vertical, but we have almost no idea what those lines look like. That’s what organizational network analysis can show us. By asking employees who talks to whom about what to get their work done, organizational network analysts can map those lines of collaboration, giving us diagrams that are the informal counterparts to org charts.

Rob Cross, an expert in organizational network analysis, has been conducting these assessments for Fortune 500 companies for years. An assistant professor of management at the University of Virginia’s McIntire School of Commerce, Cross is also director of the university’s Network Roundtable, a consortium of corporations and government agencies.

Professor Cross is co-author, with Andrew Parker, of The Hidden Power of Social Networks: How Work Really Gets Done in Organizations (Harvard Business School Press, 2004). We talked with Cross recently, and we asked him about the emphasis in that subtitle.

CSC World: Are people surprised to find that work gets done through informal networks?

Cross: No, but they are surprised by who is and isn’t influential in the network. Almost everybody knows that a network is an asset, that better-connected people do better. But when we do network diagrams for an organization and show them to the executives, they usually see only about a third of what they thought was happening. There are always substantial surprises, about people who are far more influential than they realized and people who are much more peripheral than expected.

CSC World: Is there any pattern to who is more and less influential than executives expect?

Cross: It’s common to find that people who one way or another get an executive’s ear — and there are great stories about how people do that — are not influential in the larger group. They’re more political, and executives expect them to be close to the center of a network. A lot of times, they show up on the edge.

Sometimes they’re not political. For instance, experienced hires at the executive level aren’t that successful in most organizations. These people are hired because they have depth, expertise, and a lot of contacts. But they have to prove themselves at the new organization before they’re listened to. That takes one to two years. Many aren’t willing to invest that much time, or simply quit when they can’t crack the network. A lot of these people joke that by the time they become influential, they’ve forgotten what they knew. They may joke, but it’s a very expensive error that can be avoided with a network perspective.

CSC World: How do you uncover these informal networks?

Cross: The two most prevalent ways are surveys and tracking e-mail. The networking software that tracks e-mail can help you see broad networks for an entire organization. But they don’t capture face-to-face interactions. And executives tell me e-mail isn’t how the bulk of their work gets done.

I prefer surveys. We’ve developed a Web-based survey tool here at UVA, which takes only 15 minutes to complete. The network diagram lets us see who people interact with to get their work done.

CSC World: What do you look for in those diagrams?

Cross: The first thing I look for is people who are overly connected. Whether because of job design or because they’re very good, they’ve gotten overloaded and are holding up people on the edge of the network — often without knowing it because they themselves are so busy. They become bottlenecks.

That’s a common leadership trap. Leaders see things happening around them all the time, but they lose sight of what’s happening, or not happening, out on the edge of the network.

Another problem is that people who are too central to a network can dramatically disrupt organizations when they retire, get promoted to a different area, or are hired away.

CSC World: What can you do about that?

Cross: We go to those people and find out what information they have, or what decisions they’re making, that others go to them for. We try to make information accessible elsewhere and reallocate decision-making rights so the group can operate around them. We do things to take those people out of the bottleneck position. In healthy organizations, as people rise in the hierarchy, they move out to the edge of a network, not the center. Formally or informally, they do things that decrease other people’s reliance on them.

CSC World: What else do you look for?

Cross: I look for brokers and people who aren’t connected enough.

Brokers may not have the most direct connections, but they bridge subgroups. By understanding what two different subgroups in a network know, they can have an impact on an organization’s ability to develop new products or services, or bring new ideas to clients. They can also bridge cultural differences between groups. These are people I’d want to pull into a leadership role if I were forming a community of practice or starting an organizational change program, because they have the ability to pull people together and get information out quickly.

I also look for very peripheral people. People who are stuck on the edge of a network are important because they’re underutilized resources and are three times more likely than better-connected people to leave an organization.

CSC World: Are there patterns to why people are poorly connected?

Cross: Some of them have performance issues and they’re working their way out of the organization anyway. Some are balancing work and quality-of-life issues. These are people you do not want to force back into the thick of things because you’re likely to lose them if you do. If you tried to force a single parent, for instance, or even a scientist or a subject matter expert, back into the fray, you probably wouldn’t get very good results.

Another group is peripheral because of poor staffing or because they weren’t onboarded very well. If they stay poorly connected, the organization will lose them over time. So these people are my high priority points in the network.

CSC World: Many organizations have tried collaboration tools. How well do they work?

Cross: The only effective medium for collaboration across physical distance that I’ve seen — and by this I mean a tool that seems to help networks stay connected across distance — is instant messaging. That’s a generalization, and there are exceptions. But I’ve yet to see any database come close to a human being as a source of the kind of information people need to get work done. Instant messaging recreates the kind of serendipitous conversation that happened around water coolers.

CSC World: How about team-building exercises?

Cross: I think they were useful. Where they fell down was not creating an awareness of who knows what. A lot of times I’ll go into an organization after a merger or a large-scale change effort, where there were plans to have certain groups connect in certain ways. But they don’t connect. It’s not because they don’t want to. It’s usually because people don’t know what to reach out to others for. They don’t know who knows what.

CSC World: So you also look for how people don’t collaborate.

Cross: I always look for disconnects. There are always holes at points where people need to be connecting. That happens in almost every place I’ve been. There are lots of reasons: distance, hierarchy, function. Our tool helps us find opportunity points, where people can get better connected.

CSC World: So you think networks can be managed.

Cross: Yes. There are limits to what you can do, but yes. When you see how work is and isn’t getting done, you can see how to make improvements. I think network analysis will provide crucial insights as we move further into knowledge-intensive work in delayered organizations.