on networks

13 October 2010

Networks, it feels sometimes, are every where. Everyone talks about them and every donor seems to think that they can solve any problem under the sun. At RAPID we have studied networks from a functional approach (following the basic rule that Form should follow Function; which we are also applying to the study of think tanks) and used he findings of our research to set up and facilitate some.

Simon Hearn and I were asked to present some lessons about networks at a workshop hosted by the Danish Development Research Network. Below is the outline of our presentation -and a link to it using Prezi. I hope to follow this post with a report on the discussion that we had at the event and some ideas that came out of it (so watch this space -particularly for a definition or description of networks).

1) When do we need a network?

  • Networks are a form of organisation. But are they the most appropriate form? Why not a project, hierarchy, partnership…

Theory of change

  • Many networks exist to bring about change of some kind – either policy change, social change… in a specific context with a specific history, culture, geography etc. Change happens in different ways in different contexts – we need to understand the dynamics of change in our contexts and the possible roles of a network therein.
  • But a recent article by Malcolm Gladwell suggests:

This structure makes networks enormously resilient and adaptable in low-risk situations. Wikipedia is a perfect example. It doesn’t have an editor, sitting in New York, who directs and corrects each entry. The effort of putting together each entry is self-organized. If every entry in Wikipedia were to be erased tomorrow, the content would swiftly be restored, because that’s what happens when a network of thousands spontaneously devote their time to a task.

There are many things, though, that networks don’t do well. Car companies sensibly use a network to organize their hundreds of suppliers, but not to design their cars. No one believes that the articulation of a coherent design philosophy is best handled by a sprawling, leaderless organizational system. Because networks don’t have a centralized leadership structure and clear lines of authority, they have real difficulty reaching consensus and setting goals. They can’t think strategically; they are chronically prone to conflict and error. How do you make difficult choices about tactics or strategy or philosophical direction when everyone has an equal say?

The Palestine Liberation Organization originated as a network, and the international-relations scholars Mette Eilstrup-Sangiovanni and Calvert Jones argue in a recent essay in International Security that this is why it ran into such trouble as it grew: “Structural features typical of networks—the absence of central authority, the unchecked autonomy of rival groups, and the inability to arbitrate quarrels through formal mechanisms—made the P.L.O. excessively vulnerable to outside manipulation and internal strife.

2) Defining the purpose of the network (The Network Functions Approach)

  • The purpose is the objective of the network and will help justify its existence; it is independent of the approach taken to achieve it. Identifying the purpose help answer the question ‘why are we building a network?’


  • The role describes how the network intends to serve its members. Two archetypal roles exist – support and agency. Support networks enable their members to act as agents of change while agency networks are charged by their members to be the agents of change.


  • The functions of networks describe the strategic orientation of the network – what it is that the network actually does?
  • Community building, filter, amplify, facilitate, invest/provide and convene –or anything else that appropriately describes a network

3) Setting up the network (the form)

  • Truly speaking networks are essentially a combination of relationships, and cannot be created or destroyed – only transformed.
  • When we talk about ‘setting up’ or ‘creating’ or ‘developing’ networks what we should be trying to do is identifying existing relationships and enhancing, adding-value to, expanding or formalising them.
  • However, we often attempt to create entirely new things around a new project, a new fund, a new idea.


  • Who are the existing members in the network (what are the existing networks?) what are the relationships?
  • What are the commonalities that bring members together or define them as members?
  • There are different kinds of people who would interact with the network: transients (those that come and go without committing), joiners (those that jump the first hurdle of making contact but don’t contribute or interact very frequently), boosters (those that contribute a lot and whose activity encourages others), leaders.
  • The relationships that make up networks can vary in strength – we can talk about strong ties and weak ties. The strength of the tie will determine things like trust, accountability, micro-structure, diffusion of innovation, consensus etc.
  • Large, diverse memberships (hence build on weak ties) are unwieldy and problematic.
  • Strong ties lead people to join the revolution; weak ties get you lots of friends on facebook


  • How are decisions made? How centralised are decision makers? How formal are the structures and rules?
  • Processes and structures are not the same as for hierarchical organisations but can borrow elements from them.


  • Networks are not static and can’t be depicted in the same way that hierarchies can. Structure within networks is often organic and fluid – it depends on the relationships, interests and resources of members. We shouldn’t attempt to design networks but we can identify and support emerging structures and we can define boundaries for the network.
  • Networks look different, some are labelled explicitly as networks, others not.
  • There are a number of archetypal network structures that can be used to analyse any network

4) Managing and facilitating networks

  • For networks to be of consistent value to members and remain relevant to their purpose, they need to be stewarded. Perhaps network management is the wrong term but there is certainly a necessity for some form of leadership in networks. This is usually a role taken up by a facilitator, secretariat, board, etc.

Building / maintaining community

  • A key role of the facilitator or secretariat is connecting members, ideas and activities in a way that will sustain the network’s functions and promote its purpose. Members’ interest and motivations are not always completely aligned and their contributions to the network are not always thought of with the network in mind; they are more likely to reflect their own priorities.
  • The role of the facilitator then is to attempt to make these private priorities add up to or contribute towards the priorities of the community.
  • Facilitation planning is a technique we have used to make explicit these links by scheduling events (broadest meaning) to maximise opportunities and added value of interactions.


  • Networks (unless they occur naturally) are resource intensive – there is a large overhead involved in working collaboratively and a large amount of administrative work is generated. Members must be willing to support this; otherwise there is no business case for a network. Funding can come from member fees or external donations/investments. But often what is more important than funding is in-kind contributions – members giving time, energy, ideas and experience for the benefit of the wider network.


  • Too much technology is not always good
  • Email is the most prolific communication tool in use today and may always be the most important tool for geographically dispersed networks.
  • Online social networking tools are useful for supporting structures that already exist but are not proven to be good at creating strong network structures
  • Because of the proliferation and cheap costs of online tools, there has been such an increase of so called ‘networks’ that becoming a member of a network now carries little meaning. This has the effect of decreasing the value of the network for the members and hence weakening the ties.
  • Free newspapers get dumped; while the ones you pay for (even a few pence) are carried around all day.

5) Monitoring and learning

  • Since things are changing in networks very frequently, often without coordination, monitoring needs to be a constant task to ensure the decision making structures have the relevant information.
    • And since the purpose, roles and functions of a network can be delivered through an infinite range of actions and structures, it is difficult to evaluate or assess what works and what doesn’t work (all the time).

Is the form appropriate for the functions / role / purpose?

  • Asking whether we have the right kind of network should be a regular reflection question. As functions change, the form or structure may have to adapt as well.

Are the functions appropriate for the role and purpose?

  • An objective of monitoring efforts is to assess whether the network’s functions (what it does) is in fact supporting the achievement of the role and purpose.
  • Network structures can drive the network to do things it might not have set out to do; its membership might steer the network to dedicate more effort on certain functions in detriment of others. At the same time, the context may change and members may demand new roles for the network.

Is there progress towards to purpose/goal?

  • The network exists to serve a wider purpose – if the network is not progressing towards this goal then it needs to be re-defined


  • How is accountability realised within the network? How are members accountable to each other? How is the network accountable to donors or beneficiaries?

6) Transforming and adapting

  • Networks are naturally resilient and will change and adapt to the changing environment. However, sometimes (perhaps as a result of the above monitoring) leaders in the network will need to influence changes in the formal processes and structures.

Responding to change in internal context

  • Changes can occur in the membership, relationships, resources of the network that may require careful changes in structure, function or role.

Responding to change in external context

  • Similarly, changes in external environment; social, political or economic context; discourse, knowledge or understanding can sometimes require changes – either to maintain relevance or to take advantage of opportunities.

Is a network still relevant or do we need something else?

  • Maybe we’ve come to the point where it is clear that a network is not the best way of addressing the challenges we have. Perhaps investing our resources in a project would be more effective.
  • Maybe the network can be a good platform from which to launch an initiative –driven by one or more members but in the shape of a hierarchical organisation.

6) Closing

  • Networks aren’t created nor destroyed. Formal structures can be closed down, funding can dry up but as long as there are relationships there are networks. But that said, sometimes it is important for us to realise that our investment in formal networks should come to an end – either to let a network flourish openly or to focus our resources on more controllable forms of organisation.