Engaged and effective teams are at the core of almost any successful organization. After all, teams are where all real work and innovation happens. A healthy team will naturally tend to have higher morale, and be more innovative and productive. Many organizations we speak with are rightfully interested in how they can use data to identify the common patterns that lead to successful teams and how these patterns might be copied over to other parts of their organizations.
Organizational network analysis (ONA) is an invaluable tool for the analysis of data on team dynamics and culture. It can provide great insight into the day-to-day workflows and collaboration that drive a team, and can help to identify many potential issues. We’ve applied ONA to analyze data on thousands of teams and are continually surprised by the fascinating insights we are able to glean. In case you’re considering exploring the culture on your teams, below is a summary of some of the common techniques we use to gauge how healthy teams are.
Before I get into the detail, it’s important to note that every team is different. The same set of behaviors may be great on one team and problematic on another. For example, some of the work patterns that lead to successful sales teams would wreak havoc on a team of researchers and vice versa. Context is critical when analyzing whether a team culture is healthy.
It’s critical to consider what measurable outcome you expect from a healthy team culture: Happy Employees, Low Attrition, Productivity, Sales Target Completion, Customer Satisfaction etc.
We also spend a considerable amount of time ensuring that we have good outcome variables to compare against. It’s critical to consider what measurable outcome you expect from a healthy team culture, e.g., Happy Employees, Low Attrition, Productivity, Sales Target Completion, Customer Satisfaction etc.
Below are some of the techniques we use to measure the health of teams.
One of the first things we look at when analyzing teams is the sheer volume of collaboration and communication and who is driving this activity. Some teams spend a lot of time collaborating internally, others do a lot of work with other groups in the organization or externally with customers or partners. The right mix as always depends on context but there are certainly behaviors that could be cause for further analysis.
For instance, when looking at an agile team of software developers it might be concerning if external teams are driving a disproportionate amount of collaboration as this may be a sign of a distracting environment or too many external dependencies. Conversely, one would expect the average sales team to spend the bulk of its time with customers. Spending a disproportionate amount of time with internal teams could be a sign of process bottlenecks or sales reps struggling to get access to critical information.
Below are a couple of examples of how we profile team collaboration by source and compare teams against baselines taken from successful internal or external examples.
This analysis is somewhat related to the I/O analysis above but the goal is to determine whether a team may be isolating itself from the rest of the organization. Isolated teams may have limited access to key information from within the org or may silo their own information from the rest of the organization. Both of these cases are likely suboptimal and may lead to slower decision making, poor collaboration and less innovation.
In this analysis we typically look at the strength and number of connections between team members and other groups in the organization. We compare teams against either an internal or external baseline and look for examples of extreme isolation/silos. Other variations of this analysis include looking for siloed managers and or leaders, who aren’t doing enough to connect the team to the rest of the org. We also look for siloed individuals who may be at higher risk of low morale.
Another variation of the I/O analysis above is to analyze network data looking for teams that are overly dependent on external resources or decision makers. This is a very common issue in matrixed organizations or departments using waterfall processes.
Network data can highlight some teams that are overly dependant on an external resources or teams who they require to get things done. Having key resources outside of a team increases the probability they are shared/limited and can make is harder to communicate, prioritize and make decisions.
Network data can also identify teams who spend a disproportionate amount of time reaching out to external leaders. These teams may not have the autonomy required to make quick and effective decisions. A key external stakeholder can create a significant bottleneck and make it hard to get things done. They are also likely to lack context on issues within the team and the quality of their decisions may suffer as a result.
Managers are often pivotal to the success of a team so analyzing their relationships with the team can yield interesting insights about team health. We’ve written at length about how to evaluate managers using network analysis. Below are a few of the questions that network analysis can help answer:
Another area we look at is the trade-off between time teams spend collaborating versus doing individual focused work. Network analysis can help us identify teams which are too far on either end of the spectrum. Teams spending a disproportionate amount of time on communication may be at risk of collaborative overload and individuals may have trouble finding time to focus. Teams that spend next to no time collaborating may be underutilizing collaborative tools or isolating themselves.
Below is an example of how we analyze the amount of individual focus vs collaborative time teams have:
When analyzing team health we typically use network and activity data to better understand a teams meeting habits. Too many meetings can indicate that team members lack autonomy and need to schedule meetings to make decisions by consensus. It can also indicate a team environment where meetings are used to share information instead of more efficient tools like email, chat, shared documents or task management platforms. Below are some indicators of unhealthy meeting habits, which can lead to lower productivity and engagement:
Finally, we often use network analysis to identify indicators of whether a team environment is inclusive or not. In this analysis we look at whether groups of individuals have disproportionately low access to key meetings, projects, leadership or other connected individuals.
We typically generate a list of the most important work events in a team, including key meetings, shared documents, projects and email threads. This is done by scoring work events with a weighted function of the seniority of employees involved in the work event, the role they played in the work (e.g., creator vs editor) and the frequency/volume of their contribution (e.g., 50 document edits). We then score each work event for level of inclusion by looking at the relative number of different demographic groups involved in the event. In the ideal scenario there is a reasonable level of representation by various groups in the most important work events. This indicates that people of various groups are given equal opportunity to participate, learn new skills and prove their ability.
ONA provides a powerful set of tools for the analysis of teams. It can be used to deliver in-depth insight into a vast number of common issues known to impact team performance and engagement. I hope this post provides a helpful summary of a few of the techniques you could use in your own analyses. We along with many others continue to invest in this technology and are excited by the huge number of new techniques and use cases being identified on a regular basis. I look forward to sharing these additional updates as they come!
