Rob Cross and Inga Carboni recently published an article by the above title in the MIT Sloan Management Review. The authors identify six common patterns of team dysfunction. The patterns are derived from organisational network studies (ONA) looking at the collaborative practices in teams across industries and geographies. The six patterns are illustrated below: 

The above organisational network patterns are archetypes. The idea is that if your team network structure matches one of the above archetypes, then the wise words of advice provided by the authors can be aptly applied to your team. The theory is sound, but the practice can often be a challenge.  

Prior to forming SWOOP Analytics six years ago, two of the co-founders ran an ONA consulting practice for nearly a decade, conducting more than 100 client projects in that time. There are some practical issues though in applying ONA to teams:  

• How often and how easy is it for teams to apply an ONA? Many ONA studies are survey based. How easy is it to get your whole team to respond on a regular basis? To be credible, teams need 100% response rate or close to it. 

• Even if you do conduct an ONA study, how soon before the insights date? How regularly would you need to repeat the study? 

• Your organisation may have hundreds, if not thousands of teams. Is it practical to do ONA studies on all of them? How do you manage coordination when many teams are involved? 

• What if I do an ONA and find my team’s relationship patterns don’t comfortably fit any of the above archetypes, or appear as a combination of a few? It is rare to find a neat match to a network archetype. And context always matters.  

There will always be some interpretation challenges when using archetypes. For example, the hub and spoke pattern is offered as a dysfunction. In the context of, say, an emergency services crew, this is precisely the structure you would want when there is no time for discussion or debate. Context is important.  

Using SWOOP to identify Team Dysfunction (or not) 

In this article we address how SWOOP can be used to flag possible dysfunctions along the lines identified above, but also potential high performance contexts for those very same archetype patterns. One of the reasons we formed SWOOP was to be able to do continuous ONA on as many teams or communities as possible, in as close to real time as possible. Rather than using surveys to collect network data, we infer relationships from digital tracking from collaboration tools like Microsoft Teams, Yammer and Workplace by Facebook. This is sometimes called “passive” ONA, in contrast to survey-based methods being labelled “active” ONA.  

We appreciate the nuances that can be derived through ONA surveys, but we think it is more than compensated for by the scale and frequency of online ONA. The potential to monitor team dysfunctions across all teams, groups or communities in close to real time is key. The COVID-19 inspired acceleration to remote working has resulted in a compulsory movement of teams to digital, making passive ONA even more attractive.  

The SWOOP Team Networking Map provides the network representations that can be compared with the archetype dysfunctional patterns shown above. Team network maps tend to not exactly match archetypical patterns. We find complementary SWOOP measures can provide more extensive evidence of potential dysfunctions, or in fact sometimes the reverse.  

Time frames are also important and will depend on the tempo of team activities. A three-month period is long enough to characterise most team patterns, though some highly active teams can show enduring patterns after only a few weeks. It’s worth trying several different activity window sizes to settle on what the enduring pattern of your team collaboration is.   

Hub and spoke and their variants  

The “Hub and Spoke”, “Disenfranchised Nodes” and “Misaligned Nodes” archetypes are variants of what we call “Single Leader” Teams. The example below is taken from a real time interaction pattern that has been classified by SWOOP algorithms as a “Single Leader” Team: 

It is rare to find a nice clean match to a single archetype pattern but in this case we can see Zachery is someone central, and in fact a good proportion of the network rely on him for connections, but not exclusively so. Hence, the Key Player Dependency of only 60% compared with a classic hub and spoke network where it would be 100%.  

We can see elements of the “Misaligned Nodes” archetypes to the left and right of the network map, yet the map shows a connection path to most nodes, hence to 94% connected core. The 23% gallery score shows the proportion of team members that have not been active at all and are therefore not even shown on the map. This speaks to the “Disenfranchised Nodes” archetype. The ‘quality’ of the connections are not shown for the archetypes, but from the network map we can see many of the nodes are only connected by one-way interactions (dotted links) resulting in only a 26% reciprocity (%Two Way). Additionally, the interaction level rating is only a single star (bottom 20% of all teams); so we can see elements of all the first three archetypes in this example.  

Overwhelmed nodes and their variants  

The “Overwhelmed Nodes” Archetype in SWOOP terms is our Self-Directed Team Persona. Some self-directed teams could very well be overloaded, but a good proportion could be an aspirational self-sufficient agile team. We need to look at more data to find out which is the case. 

The above network map has a pattern consistent with the “Overwhelmed Nodes” Archetype, where everyone is connected to everyone else (%Core is 100% and Key Player Dependency is zero). With 85% of the connections being reciprocated and the level of activity in to top 20% of all teams, (chat, calls and meeting interactions are not included here) is a real possibility for communication overload. When we look at the quality of the links, they are uneven (thickness of the links), so if indeed overload is occurring, it is uneven and not all team members may be overloaded. This, of course, can provide opportunities for ‘sharing the load’ more evenly.  

What the archetypes do not consider in terms of overload is that staff are often active members of more than one digital team. If some of these members are sustaining large digital networks beyond this team, the potential for overload is even greater. We address this issue in our article on ‘burnout’ at work.   

SWOOP provides an individual comparison chart for each team member. The Multi-group Participation index shows the degree to which a staff member is active across multiple teams. We can see that Madison Waxem and Austin Thdoz are highly active in multiple teams and are also central to this team; so good candidates for being overloaded. In contrast, Brian Carearte is equally central to this team but far less active than Madison and Austin in other teams. 

The “Isolated Networks” archetype reflects a connected core of members and what we call a “gallery” of inactive (isolated) observers. This isn’t always a dysfunction. Many digital teams are used as discussion forums, where the active core of organisational leaders may be wanting to communicate en masse with many staff members. In this case, the gallery is an information consumer. We need to look at more data to decide which context it is. 

The “Priority Overload” archetype suggests interference from a set of external stakeholders with competing priorities can result in confusion and cross purposes. Alternatively, this same pattern could describe a “Community of Practice”, with the discipline experts at the core and the external less connected nodes being the novice practitioners looking to learn from the experts.   

Let’s look at another real-life example: 

In this case we have a reasonably well-connected core (64%) with a less connected periphery, relatively low-key player dependency, a good %Two-way (reciprocity), but also a large gallery where the nodes are not shown on the map, making the periphery even larger. While we have classified this team as potentially a Community of Practice, there is still a possibility that it could be the “Isolated Networks” or the “Priority Overload” archetype. We would need to survey the members of the gallery to decide if they were feeling isolated or simplly happily passive observers and learners.  

Another hint about internal versus external nodes can be discovered by looking at which formal lines of business the team members are being drawn from, and therefore how much cross-organisational interactions are being helped by this team. For the above team, we find what is referred to as the “Kite”: 

In the Kite itself we see Professional Services, IT, Europe and Head Office are interacting freely. In the tail we have Health and Safety and Operations weakly connected to the head of the kite. The team members from these divisions are likely more isolated than influencing priorities. 

Summing Up  

In this article we have tried to expose some of the nuances and practicalities in applying ONA to diagnose dysfunctions within teams. While the theory and advice behind the archetypes provided by Cross and Carboni is sound, context is everything. An “Overwhelmed” team archetype map could be a high performing agile team. An “Isolated Network” pattern could, in fact, be a thriving community of practice or information-sharing forum. By looking at other auxiliary network data we can better discriminate between what is dysfunctional or in fact quite the opposite.