Team Size - Dr. Nick Keca

Implications of Team size

In this podcast I’d like to talk about the implications of team size on social interactions and coordination within, and between, organisational teams.

Team Size: Friend or Foe?

If you continually struggle to manage your email Inbox, or if you hold any kind of leadership role, this article may be of interest to you as it helps to explain how team size impacts you and your team.


In the past we’ve tended to visualise organisation structures hierarchically, as business units, functions, teams, and work groups, in some cases, we still do. Consequently, when thinking about composing teams, we tended to pay attention to spans of control, perhaps following a rule of thumb, or existing organisational norms. However, hierarchical structures, like this, are largely an artefact of the command and control era, and this no longer reflects the way that teams and groups in contemporary matrix organisations operate in practice.

Delayering and virtualising operations increases interaction volume due to Bossard’s Law, the impact of which increases the likelihood of dysfunctional performance and coordination failure, and means that:

  • Costs of interaction failure are largely ‘hidden’, and
  • Acceptable performance is increasingly difficult to sustain, and effort needed to do so diminishes returns

Bossard’s Law of Interaction

Bossard’s Law…

…explains the nature of interactions between two entities (actors). In this context, actorscould be people, teams or groups, functions, organisations, processes, IT servers, network nodes, and/or software applications.

In other words, Bossard’s Law influences all things that interact in dyadic relationships, i.e. between two parties. This is the basis for social network analysis and it explains, for example, why your LinkedIn relationships multiply as you add connections.

The problems for teams is that the relationship between team size and interaction volume is exponential.

Bossard’s Law is expressed in simple mathematical equations…where N is the number of people in the team and, in this example, there are 2 people in the team as follows…



Therefore, interaction volume is calculated as…Bossards Law

So two people have one interaction.

The more people you add, the greater the number of interactions generated, and it doesn’t take long before interaction volume is overwhelming and leads to coordination loss (failure cost). Obviously machines are created to handle high transactions volumes whereas people can generally only cope with an interaction volume of about 100, which approximately equates to a group size of about 15 individuals.

When there are 3 people in the team (N = 3)…

3 interactions


Therefore, interaction volume is 3.

Adding just one person results in two more interactions, and so on…





But, group structures in matrix organisations and virtual teams can be much bigger than this….which is where the problems begin. Although much depends on the nature of work complexity, the impact of this interaction volume on matrix organisations is profound because the nature of interactions in such organisations are more complex than the example given above, i.e. interactions tend to be between and within groups(which is a different equation of Bossard’s Law).

Things you can do to help

There’s no easy solution but there are things you can do to reduce the impact of interaction volume:-

  • Be aware that this isn’t about spans of control of people in groups, its about task complexity and how tasks are inter-related within your workflow processes since complexity and interdependence are proxies for one another
  • When you group people logically in your workflow (value stream), make sure you create structures that keep groups as small as possible
  • Keep people that do interdependent tasks (in the workflow sequence) as close together as you can
  • Select people that have positive pro-social behaviours and train them in boundary spanning processes
  • Understand that interaction volume creates a high stress situation for key people in your in-group as you don’t want to burn your people out, especially if they are business critical
  • Think about the personality types that are likely to be best suited for the pro-social behaviours required in your distributed organisation
  • Think about functional diversity and whether it helps or hinders coordination processes
  • Raise awareness and understanding of the group dynamics associated with interdependent working, i.e. high conflict – and the cost and performance impacts
  • Set priorities to reduce the number of interdependencies
  • Centralise the process of identifying interdependencies
  • Automate knowledge transfer as far as possible, particularly in the early project planning phases of a change
  • Develop the use of synchronous communication technology to all roles fulfilling interdependent (additive) tasks requiring high uncertainty and coordination


Leave a Reply

Your email address will not be published. Required fields are marked *