20% of Our Group's Value Is Quitting

I've written about "social network analysis" previously, and described some of the initial work in this area by Rob Cross, a faculty member at the McIntire School of Commerce at the University of Virginia. In a nutshell, "social network analysis" (SNA) is about analyzing the real organizational networks inside a firm (as opposed to those that appear on org. charts or departmental diagrams), with a goal of making sure that people who should be in touch are in touch, that "best practices" are shared spontaneously and naturally throughout a firm, and that isolated pockets don't develop unintentionally.

While this may sound intuitive in theory, the obstacles have traditionally been (a) understanding what the firm's actual operating networks are, and diagnosing how they could be intentionally improved; and (b) demonstrating that the firm actually gets something for its efforts.

Now, using companies as diverse as Whirlpool, Sanofi-Aventis, and Halliburton, Booz-Allen's Strategy + Business has a piece on how to go about just that. At Whirlpool, for example, 400 employees from a range of departments were trained, starting in 2000, in "ideation" (biz-school speak for brainstorming) and divided into teams headed by "innovation mentors" designed to identify unmet consumer needs and target R&D accordingly. Whirlpool's new product introductions have gone from a handful per year to dozens, including the new "Gladiator" product line offering configurable combinations of appliances, storage, and workbench areas.

Despite success stories like this, opposition, some of it principled, remains. One school of thought treats social networks as "emergent communities" spontaneously formed by serendipitously shared interests, and therefore not susceptible to management. A related belief is that networks, as self-governing communities, cannot be interfered with without doing damage. To be sure, throwing money or collaborative software at them without concrete priorities specified and some way of measuring success can be wasteful, and installing Procrustean individual performance metrics can cause the communities to self-destruct.

Nevertheless, if one starts with a rigorously specified social network diagram, one can identify key lines of connection, see which individuals are "nodes," and who fills the role of "natural brokers" spanning two or more otherwise disparate communities. For example, Halliburton, less famous today for its core business of oil exploration and drilling than for its hapless performance in the Iraq theater of war, sought a way to fix highly disparate performance among its various "completion" groups—those responsible for taking oil wells from exploration and drillling to actual production. These groups, widely dispersed in places like the Gulf of Mexico, Canada, the North Sea, Nigeria, Angola, Brazil, and Saudi Arabia, were experiencing highly variable rates of success and difficulty in getting essentially the same job done.

A "network analysis" of the groups revealed, perhaps unsurprisingly, that they communicated relatively little among themselves. Nigeria talked to Nigeria, Brazil to Brazil, etc. The standout performer in the groups was the Gulf of Mexico team, which had created many of the communities' best practices and was improving performance steadily, while across the other six countries, "costs related to poor quality" (e.g., delays) were up 13%, a key poor-performance metric.

Based on its network map, Halliburton did two things:

• moved some individuals in the Gulf to other regions; and
• moved individuals with strong ties in other regions to the Gulf.

These moves were not random, but highly targeted, selecting people already identified as having high growth potential, and investing in their professional development, with highly visible promotions typically occuring on their return "home." A year after the transfers, an updated network analysis showed a far richer skein of interconnections, not just to and from the Gulf of Mexico, but between essentially all other nation "pairs." Concrete results?

• the number of personal referrals needed to connect someone with a question to someone with the answer dropped 25%;
• revenue increased 22%
• the "cost of poor quality" metric declined 66%
• overall productivity went up 10% (this is in just one year, recall); and finally
• customer dissatisfaction dropped 24%.

Impressive as that might be, we know it can be hard getting there from here. As Cross and his co-author describe it:

"Many people are reluctant to ask colleagues whom they don’t know personally for help, even within the same organization, and for a wide range of reasons: Will they think I’m stupid for asking the questions? Are they really experts? How can I trust them?"

One way to help overcome this is through giving people information about others in advance in hopes they will discover commonalities of interest, and letting human nature take its course. What type of information? Not just areas of expertise ("China," "project finance," "technology IP licensing") but personal information (alma mater, hobbies).

So what's the payoff, again? Consider the well-known reality that a consistent differentiator of high-performance individuals is their habit of cultivating ties outside their unit and outside the organization. For example, in one (unidentified) financial services organization, a key female leader of one community of practice was determined through SNA to be surprisingly central to the entire group. Indeed, by herself she accounted for nearly one-fifth of the entire unit's value creation.

"When we asked one of the company’s leaders what would happen if she left the organization, he blanched. It turned out that she had recently submitted her resignation."

Had the firm known the value of the woman's centrality earlier, they could presumably have taken steps to retain her.

The more far-flung organizations (and law firms) become, the more important it is to ensure the "knowledge workers" who inhabit them are well-connected. This nicely sums up its value:

“The system we have developed is intrinsically rewarding to the users,” says Guillermo Velasquez [instrumental in the Halliburton experience]. People participate because they see value. Experts get recognition. As time goes by and people in the community start to know each other, they develop reciprocity. An individual in need today may be tomorrow’s expert providing the knowledge to help solve a problem. Gradually, we see much higher trust, and the community changes from the mode of ‘getting the right information to the right person at the right time’ to truly start building on each other’s ideas to find a solution to a problem. In other words, that’s when we start creating knowledge.”

"Creating knowledge?!"