In times of digital transformation and agile ways of working, we are constantly working in teams. But sooner or later, when putting together teams, the question arises as to how big the team should actually be? With how many people can a meeting still be effective and efficient? How can the cooperation in the team be designed in the best possible way?
The well-known Scrum process model for agile project development recognizes three roles:
- Product Owner: Responsible for maximizing the value and features of the product
- Scrum Master: Responsible for the smooth implementation of the Scrum method.
- Developer: Responsible for the implementation of the product The roles result in a minimum number of three people, where Scrum counts the role of Product Owner and Scrum Master exclusive of the defined team size. But then where exactly is the upper limit for a team? The official Scrum Guide recommends that the size of a Scrum team should not exceed ten people. Thus, it is still small enough to quickly and
agile, and large enough to also be able to deliver a significant result within a sprint. Should Scrum teams
become larger, the guide recommends splitting the team into more related teams, but still working on the same product.
Amazon founder Jeff Bezos also asked himself the same question, and came up with the following rule of thumb: “We try to create teams that are no larger than can be fed by two pizzas. We call that the two-pizza team rule.”. This lineup – an estimated four to eight members – keeps cation lines short, increases collaboration, and improves efficiency. There are many more opinions on how big the team size should be, but they all usually vary between three and ten members. These rules of thumb and opinions can also be formalized in some way by mathematical formulas such as those of Moore, Rogers, or Bell, which calculate the complexity of the group by the number of possible interaction systems between the members.As can be seen in the graph, increasing from five to six members already results in a huge jump in the possible interaction systems of the group. The group size should therefore be carefully considered.
