Social networks and how things (really) get done

Peter Brown PhD FCIPD and Hannah Williams (FCIPD)

I knew the answer to my problem was out there, but there was just one thing that was making me anxious: I wasn’t sure who to ask or where to start.

That was me, on a spring morning sat in my office, staring out the window I knew I needed to get some heads around the table to discuss and make sense of a complex problem. I had that worrying feeling of who do I ask - my fingers hovered over my keyboard not sure who to email first. I always tend to speak with the same two or three people but this problem required more brain power and diversity of thought. So, I made a list on my sticky notes of people I should engage. I fired off the email and within minutes had replies of people willing to help; a few even said we should probably invite colleagues (who I didn’t know) from all corners of the organisation. In that moment I realised that our professional connections are crucial to how we get work done and that I must deliberately work on these to cultivate authentic, strong relationships. These connections and relationships are called our social networks.

That was my problem a few months back, and here’s what we did about it by exploiting our social networks.

?

What are social networks?

Networks are interconnected systems of people, knowledge, resources, teams and organisations, organised around a common purpose and outcomes. Social network analysis (SNA) sometimes called Organisational network analysis (ONA; which is our preference as it avoids confusion with social media, however may need framing for those in IT roles as it’s nothing to do with laying network cables), is a research method which enables the visualisation of the connections between people, ideas and things.

The reason this is such a useful tool is it surfaces otherwise hidden patterns and informal relationships which exists in an organisation. We often see the organisational structure through the lens of a hierarchy with managers managing other managers and so on. However, this industrial-revolution approach to describing how work gets done in organisations is outdated. Formal structure has a place, especially within the realms of accountability, but its sphere of influence is dominated by compliance. We know that how work actually gets done is in the gaps between formal structures, through our relationships, its often messy and difficult to represent on paper. In Japan, the process of gathering informal support for ideas before presenting these formally is called ‘Nemawashi’;? tapping into our relationships and maximising our network.

ONA offers us a glimpse at the underlying patterns of how work really gets done centred on social influence and value creation; broadly described by the emerging concept of organisational physics.

Now we're not saying to throw hierarchy out the window completely, but we should consider a balance between formal and informal structures when we considered the realms of our people practices and interventions.

?

Fundamental elements of ONA visualisation

Some of the fundamentals of ONA are worth describing. When we do it, we end up with a graph (not a map: see the moot point at the end of this article) with lots of dots which are called ‘nodes’ – in my examples below these nodes are people. People are connected to other people (nodes) by what’s called edges (these are the lines joining the nodes) and you can measure the distance, quality, weight and strength of these edges representing connections and relationships. There are no good or bad networks, they just are, but they do provide a useful insight for which we can amplify or dampen the interventions within a system (which may be a small team or an entire organisation).

When collecting data, individuals are asked to identify who they connect with on a variety of topic areas, for example making decisions, problem solving, managing their personal wellbeing, handling conflict or whatever you are interested in uncovering. They list them from the most connected to the least and can list as many as they like.

?

Analysing the data

Once you’ve collected your data using simple survey methods and imported it into a software package (for example we use Gephi, free open source software) ONA can offer a raft of quantitative data. However for the purposes of creating interventions within teams and divisions in our organisation, five statistics ?typically provide what’s needed. This includes:

• Visual inspection – what do you notice, the patterns, the obvious links, divisions, connectors and those on the periphery, what’s consistent with your hypothesis, what’s different or surprising
• Density – the relative number of connections to the total connections available
• Degree Centrality – the importance of a node within a network, the most connected and active people, who often have a strong affect upon the network due to their multiple unique connections and likely the go to people within a group.
• Betweenness Centrality – typically the brokers or gatekeepers in a network. They often connect different parts of the network, sometimes without knowing. They are often crucial in the flow of ideas and things across a team / organisation.
• Closeness centrality – this describes how quickly a node can reach another node, or any other node via the shortest path in their network; consider the theory of 6 degrees of separation. These people are efficient connectors and help the spread of information rapidly in their network and also key players during change.

?

Below we will present 2 examples from recent pieces of work.

Example One was a small group of experts who work across their region leading their own local teams, but the Head of the group had a sense that they were over reliant upon them for problem solving and making decisions. They wanted to enhance distributed leadership and empower the leaders to do for themselves, but they needed evidence.

Example two was a group of leaders in the same role, but within different functional departments in the organisation. There was a sense from their senior leader that the group were not well connected to each other, they weren’t coming together to share ideas and problem solve together. This was described as a limitation as they have similar challenges.

?

Example 1:        

Here is the ONA graph for scenario one. The leader was right, their thought that they were being overly relied upon was true. The leader was the key node within the group becoming a bottleneck through over reliance for decision making and problem solving. Interestingly a few other key nodes were highlighted too as key brokers. Strategies have now been put in place to amplify distributed leadership with only weak ties to the leader and to strengthen ties between the experts and the good practice of the secondary problem-solving network depicted.

?

Example 2:        

The senior leader’s hunch was again correct. Despite being in a similar role yet functionally disparate, the group were poorly connected when it came to supporting each other’s problem solving and decision making. Interventions have been designed and deployed to accelerate opportunities to connect, engage in facilitated creative problem solving and rapid sensemaking (using the Cynefin framework). Interesting there were members of the network that despite being in the same office block were not connected as they were in different departments; they had created an invisible barrier around themselves and used an intermediary rather than one another for support.?

?

So in summary...

The answer to my complex problem was out there and it arrived by bringing together a network of people some of who we knew and some of who we didn’t. There are many ways to interpret ONA and like most statistical analyses you can (and can be at risk of) cherry picking the metrics or visual interpretation which work for you to confirm your hypothesis.

So, it’s good to have a sense of what you are trying to evaluate before you start and we always do this in partnership with stakeholders and network members. For example, a very well-connected node may be considered an important broker in a network, however when you think about disease and infection prevention and control or negativity, they may be a significant source of viral transport. When we think about people who have few connections, we may think of them as isolated but evidence suggests that weak ties may be better for innovation.

So next time your fingers are hovering over the keyboard wondering who to speak with about a problem or decision, consider your network and whether it’s as efficient and effective as it can be to create change and value.

?

Footnote: Moot point worth noting for the purists!

ONA outputs are graphs, not maps. The process enables us to describe the current state, not map it. Maps as Simon Wardley describes need an anchor (in geographical terms think magnetic north), they need relative position (i.e. where you are in the world in relation to the compass) and consistency of movement (e.g. if you drive from the west cost to the east-coast you have been on a journey). In addition, and importantly, in maps space has meaning whereas in graphs they don’t.

To continue Wardley’s example, if you were to move Paris a few hundred miles north off the coast of England, that would have significant implications, on the contrary if you move a node on an ONA graph, its edges and connections follow, perhaps looking a little different but fundamentally it doesn’t change its message or interpretation. This is shown elegantly below in the images, from a tweet by Simon Wardley which you can view here (M1 and M2 in the example below represent motorway roads).

Also, some useful posts can be found here:        

The Power of Hidden Teams (hbr.org)

Org Physics: The 3 faces of every company | by Niels Pflaeging | Medium

https://www2.deloitte.com/us/en/insights/focus/technology-and-the-future-of-work/organizational-network-analysis-network-of-teams.html

https://scholar.harvard.edu/people_analytics/publications/how-can-organizational-network-analysis-ona-help-improve-company

https://www.mckinsey.com/business-functions/organization/our-insights/harnessing-the-power-of-informal-employee-networks

Download (gephi.org)

?

?