Organization Network Analysis — Know your organization better !!!

5 min readMay 17, 2021

Preface

Organizational Network Analysis (ONA) is the set of scientific methods and theories to help understand interactions within an organization. It helps executives and managers to intervene at critical times, increase performance, and reduce costs.

Introduction to Organization Network Analysis (ONA)

There’s increasing pressure on executives to drive sustained, long-term growth. Yet, they lack the information they need to make informed business decisions and successfully initiate change. As organizations restructure departments to have fewer hierarchical levels, work increasingly occurs between social networks, rather than through prescribed reporting structures. Research shows that employees look to their networks to find information and to solve problems. Communication no longer flows solely from senior management to individual contributors — information moves through social networks, between colleagues and different teams. Organizations can analyze social networks to assess how information flows between teams and to intervene at critical times in order to improve how work gets done.

Consider this simplified comparison of the formal org structure and the informal network, drawn from a much larger ONA in the exploration and production division of a major petroleum organization. Names and identifying details have been changed for confidentiality.

This ONA identifies mid-level managers who are critical to information flow. Note the very central role of Mitchell; he is also the only point of contact between members of the production division and the rest of the network.

ONA Terminologies

Graph

A common way to visually represent social networks, consisting of two dimensions: actors and relations (also called nodes and edges).

Node

Nodes are the entities in graph (also called vectors). For example, if we consider Facebook friends as a graph, then every friend is a node.

Edge

These are the relationships between nodes. For example, if we consider Facebook friends as a graph then every friendship is an edge.

Centrality

There are multiple ways to determine a node’s importance, or centrality. The measure you use depends on how you define centrality. Several of these measures are:

Degree centrality

An important node is involved in large number of interactions. The number of edges connected with a particular node.

2. Eigenvector centrality

An important node is connected to important neighbors. This is a measure of influence of a given node in the whole network. The notion is how well-connected a given node is with other well connected nodes in the network. This is how, for example, Google determines page rank.

3. Betweenness centrality

An important node lies on a high proportion of paths between other nodes in the network. Model based on communication flow. A person who lies on communication paths can control communication flow, and is thus important.

4. Closeness centrality

An important node is typically “close” to, and can communicate quickly with, the other nodes in the network. Length of the average shortest path between a given node and all other nodes in a graph.

ONA Data collection — types

Benefits of Organizational Network Analysis

Revelation of the actual emerging relationships in an organization to identify, assess and streamline information flow in an organization.
Identification of key players and most influential employees to enhance communication flow and improve business results.
Identifying strategy and innovation groups to help you tap into them and find key resources to increase the competitive advantage for the business.
Establishing the new and more efficient way to coordinate and communicate with employees for better response. This is necessary particularly when initiating change like adopting new technology, culture or any restructuring.
Identification of sustainable linkages in a firm and how to improve such linkages.

ONA — A sample network analysis use case

Let us say we have a list of people in an organization and how they are connected to each other directly. We will now try to capture the shortest path between two people a and b and other possible ways to connect them.

The below graph indicates how two people ,Margaret Fell and George Whitehead, nodes of which are colored in light sea green, can be possibly connected(highlighted in orange) and the shortest path between them(highlighted in different colors)

There are 4 shortest paths between Margaret Fell and George Whitehead which are marked in 4 different colors and the remaining paths are marked in yellow, with a cut off of 5 which limits the connections between the two to only 5 nodes are fewer for each set. The width of the yellow line indicates the connection between them.

Python code to generate the above graph

aramachandran21/ona-sample1

Contribute to aramachandran21/ona-sample1 development by creating an account on GitHub.

github.com

Now, let us generate an interactive graph for another sample dataset that has an image for each node and also displays the connections emerging from the highlighted node. Let us also determine the importance of each node using eigenvector centrality principle which helps us understand how removing the node disrupts the entire network.

The images of the numbers are put across each node which can be customized to peoples’ pictures. The influencer rank indicates how influential the person is with 1 being the highest based on eigenvector centrality.