Analytics 101: The Map We’ve All Been Looking For

Share This
October 24, 2016 | Ankita Poddar | HCI

A good map is the only tool you need when lost and given HR’s perpetual ‘lost’ state with regard to its people, it should come as no surprise that a good map would come to the rescue in this case as well. There is of course much art involved in drawing the right map, which is precisely the raison d'etre of this blog.

Let’s begin with why we need a map and how maps help navigate the complex world of people. If you’ve worked in an organization for long enough, you know that the formal organization structure rarely paints an accurate picture of how work actually gets done. Often it is the informal networks that cut across boundaries and help work move forward. While, we’d like to believe our leaders understand these informal networks, actual mapping can either validate existing notions or challenge prevailing beliefs.

What organizational network analysis (ONA) does is scientifically interpret either survey or interaction data available in the ecosystem to help understand informal networks. It illustrates how people work together to solve problems and make decisions. This analysis helps reveal insights at both an individual and organizational level that most existing systems do not help identify. This analysis facilitates visualizing patterns of relationships and may discover unexpected networks, explain conflicts or behaviors, diagnose opportunities for change efforts, and outline specific developmental needs. The network map generated also uncovers leaders who effectively network to support collaboration for problem solving and innovation, uncover leaders that have tribal knowledge upon which a large number of employees rely, act as a mechanism for identifying talent that facilitate or potentially block innovation and can assist in identifying overleveraged leaders.

ONA can help organizations solve different problems based on identified parameters and network maps that they choose to create. E.g., the advice network shows prominent players in an organization on whom others depend to solve problems and provide technical information. Trust networks tells us which employees share information and back one another in crisis. Communication networks reveal the employees who talk about work related matters on a regular basis. 1

The below image depicts a sample network map on information sharing:

The above image helps recognize peripheral specialists (Kevin) and understand why the individual is disconnected from the larger group and if his role is structured in a way to encourage this. It also helps identify Paul as a boundary spanner who helps connect various networks within the identified boundaries. Paul’s exit from the network will lead to a breakdown of communication between the marketing, finance and manufacturing teams. Following a review of this map, leaders can choose to either increase collaboration or interaction between the identified teams, create a successor for Paul or choose to do both.

The data for plotting a network map can be either be collected via survey responses from identified target groups or from back-end data on the server (emails, communicator, meeting requests etc.). Given confidentiality, noise and control over data, I would recommend mapping networks via survey responses.

The picture, however, isn’t all rosy and like every other tool, there are a few risks and implementation challenges associated. Firstly, network surveys need a high response rate. Low responses lead to holes in network maps and may paint an inaccurate picture rendering the entire exercise futile. Mapping survey data is another challenge. While organizations have begun to realize the importance of informal networks, mapping them is yet to take mainstream spotlight. Thankfully, there are solutions like Kumu, UCINet, and other software that help make life easy. It is also possible to manipulate input data and it is important to help survey respondents understand the importance of providing accurate data. Confidentiality becomes key in drawing these responses. Leaders need to give confidence and comfort to the participants that the decisions will only be data informed and used to achieve larger organizational goals and not be used to base assessment/growth decision.

While the above may sound daunting, the successful implementation of network analysis by far outweighs any reservation that you may have. Few tools in the HR analytics world are as powerful as ONA. You would have to try it to believe it. I promise you will not be disappointed. 

[1] Krackhardt, D. and Hanson, J. R. (1993) “Informal Networks; The company behind the Chart”, Harvard Business Review