Worklytics Blog - Analytics on Work and Collaboration

Taking Network Analysis Beyond Email

Organizational Network Analysis (ONA) is fast proving itself as an invaluable tool with a broad set of applications in people analytics, ranging from organizational design, to operational efficiency and change management. Data on employee relationships within an organization can provide rich insight into how teams and departments operate and share information.

A large portion of the analyses in this space are now performed using data gathered from email networks, extracted from tools like Microsoft Exchange. These networks infer employee relationships by assuming that sending or receiving an email indicates some connection between two or more people. The more emails sent or received, the stronger the connections might be.

Email networks are of course a great source of network data but when used alone they can provide an incomplete picture. The number of digital tools used in most workplaces has ballooned and email has decreased sharply as a relative portion of digital collaboration. The average enterprise organization now uses 100s of different cloud tools, including new communication products like Slack and Microsoft Teams, which aim to replace email. Each of these new tools contains data on one, if not multiple employee networks. Even tools like Sales CRMs, ticket management systems, and intranets contain rich sets network data.

Below is a diagram of six networks collected from different tools within the same organization. Each diagram shows a completely different set of relationships. It's clear from this that there are a large number of strong employee relationships that exist in some networks but don't show up in email.

Analysis showing collaborative overload
Diagram showing some of the many networks that exist within tools in a single organization

If you were analyzing something like the growth of employee networks during an onboarding process, it would be misleading to look at only the email network above. You might incorrectly infer that certain groups of individuals don’t have any internal networks, when really they are just using a variety of different tools to collaborate.

In organizations we’ve analyzed, as many as 30% of employees may not send any email for months at a time and this percentage seems to be increasing. Our research indicates that email tends to be heavily used by more senior managers but that most team members make far greater use of other tools for collaboration.

It’s also worth thinking about the nature of the relationship indicated by each type of network. Email is a broadcast communication tool and it’s relatively common that large numbers of people are included in threads. Email networks tend to be very dense as a result. Everyone is connected to everyone else! But being included in an email thread may not indicate any real working relationship. On the other hand, a network from a document management tool such as OneDrive or Dropbox, shows people who have actively worked together on the same documents. They’ve actually edited or reviewed each other’s work. This is clearly a far stronger indication of a collaborative relationship!

The advantage of email network data is that it’s relatively easily accessible and simple to process. In many cases, however, it is not the best or only source you should consider. It’s worth taking time to think about which networks are best suited to each particular analysis. For some analyses, it makes sense to use multiple networks together to get a clearer global picture of an organization.