 Formalizing Employee Networks
A colleague recently sent me a link to an outstanding article by McKinsey on the importance of employee networks: "Harnessing the power of informal employee networks." Some of it's key findings are very similar to the findings in Six-Degrees: The Science of a Connected Age
which I also think is worth a read. The authors in the McKinsey piece
note that most the real information in large organizations does
not flow through traditional employee hierarchical structures, but
rather through ad hoc or standing employee neworks that exist
outside traditional structures.
One
of the interesting concepts in the study of social networks (and a big
focus in the Six Degrees book) is the idea of social clusters and
shared memberships or proximity. It's one of the core concepts
underlying the "six degrees of Kevin Bacon" idea. Typically, people
self-organize into groups around shared interests. Of course, since
people typically have mutliple interests, they actually participate in
mutliple groups. With a relatively small number of highly connected
people, it's possible for information to flow quite fluidly across
multiple disparate groups.
Here
is a real world example: my indoor soccer team is run by a guy named
Francis. I play with him on Sundays. He also plays on different teams
on Tues, Wed, and Sat. That's four different teams with at least 10
players each -- 40 people. He also skis, typically with a group of
10-30 different connections, some of whom are friends from soccer, some
of whom are new. So we have another 10-15 unique connections. He also
rides a motocycle and is part of an online BMW biking community where,
given his nature, he is has over 20 strong connections to other
members, none of whom are part of his other groups. And of course, he
works and has at least 15 strong work connections. In the course of a
single week, Francis is strongly connected to roughly 80 different
people across 7 different "groups." So if I know Francis, I'm one
connection from these people. In turn, he is one connection from my
group members, like my Facebook contacts, my buddies in California, or the guys I lift with at the gym.
These
connections, even in our personal world, enable us to be more
productive and efficient. When I was doing yard work and needed some
specialized work, I reached out to my buddy Rob, who in turn, reached
out to his connections, who put me in touch with the right contractor.
The McKinsey argues that this same exact thing happens in a work
environment. Take the following diagram they present:
In
this diagram, it seems clear from the formal strcuture that Jones, the
SVP, would probably know the most about what's going on in his
organization. Of course, the network version of these relationships
reveals this to be false. In fact, it's probably Cole who is the most
plugged in to what's going on, by virtue of his more extensive network
and interconnections. More importantly, Cole is the only one
connecting the Production group to Exploration and Drilling. He is a
lynchpin in the network and is in some ways, a lot more valuable to the
ongoing smooth operation of the business than any of the people above
him.
In
the McKinsey piece, the authors argue that this actually a bad thing.
While it's great that Cole is holding everything together, it's not so
great that the organization is this dependent on one individual. What
if Cole is a problem child, and uses his position to gain power or
horde information? What if Cole gets hit by a bus or decides to move
on? What if Cole gets way too busy for a period of time and doesn't
have the cycles to share information? On the one hand, Cole is a
gateway between the groups, but on the other, he's a potential
bottleneck.
One
of the main suggestions of the article is to overcome this fragility by
formalizing this ad hoc network into a standing employee community. In
this model, there is still a hierarchical structure, but there is a
parallel network structure that is independently managed by a formal
"network" manager. Both the network manager and the traditional "boss"
would be equally responsible for the success of team members. As with
more informal, ad hoc networks, members could join multiple groups and
thereby act as conduits between disparate groups within the
organization.
Interestingly,
one of the main benefits to this formalization is not explicitly
addressed by the McKinsey authors, namely the strengthening of
otherwise "weak" ties between groups. And the elimination of
"control." In the Exhibit above, one the biggest problems is that
there is a very low level of independence between members of the
Exploration and Production teams or between Drilling and Production.
In both cases, Cole alone has a very high level of control or
centrality. For all the reasons mentioned above, this is a bad thing.
One of the main goals of any formal employee network should be
to increase both strong and weak ties among members. Strong ties
increase the overall independence of the connections and decrease the
centrality of any individual member. This enables the network to
better sustain the loss of any individual member and avoid any
information bottlenecks or hoarding. Weak ties enable teams to better
connect outside the core group and to tap into extended expertise among
a more heterogenous network.
These
issues should be of particular importance with regard to retention and
the impending waves of Boomer retirement. How do companies prepare for
the loss of information and exertise that is rapidly approaching?
Establishing a formal network that is inherently structured to
withstand the loss of indvidual members is certainly a step in the
right direction, as is an ability for this network to tap into extended
expertise from a variety of unique sources.
Where
else in our HR world might these more formal networks provide value?
Maybe talent management? Look again at the Exhibit referenced above.
In the formal hierarchy, Cole is at the bottom of the hiearchy, just
another cog in the machine. But in the network structure, he is a
lynchpin, holding together disparate groups and connecting multiple
people within his own group. Why? Maybe he's a natural leader? Maybe
he has expertise others need? Maybe he is a great listener
or facilitator? Whatever the cause, it's clear that Cole is
worth considering somehow, whether that's for a leadership training
program, succession planning, or just promotion and compensation.
What about recruiting? Or learning
and development? How might formal networks help with these issues?
What are some of the negatives? Does anyone have any direct experience
with these sorts of formal employee networks?
Thu, Oct 18 2007 |
Blogs
