The eternal question is if size matters or not? One says it is important others think it is not, while there are people saying: it depends. What is the best size? Sure you may say the larger is better? You are maybe right. May be not... Come and read more to get the answer!
Certainly in this environment we are speaking about the size of organisations, applications or projects. Both the organisations and projects have a common measure, the team. I'd like to show you a mind about the team size and an opposing dimension to delivery capability. This aspect is the frustration may come from the task, the political forces of an organisation.
The one man show
This situation may mean a very productive situation. On the other hand it causes frustration because of the diversity of tasks, the absence situations and because of the feeling that there is no other view to check the results. Excluding the cases mentioned in the Do you need a team player? article, it is far to be the most effective size.
3 champions
Small teams can be simulated by the 3 champions case. Here the team consists some great talented people, who may help and substitute to each other. These teams working so good if the members are focusing on the team target and not playing rivalry games to acquire the leader or the best seen position. The negative side of this small size is that falling-out of a member usually kills the whole team at all.
Best fit
Here we speak about teams has just enough people to cover all functions with enough contingency. Let's define the team size between 5 and 15. Having this size you will have dedicated positions for the different tasks, while the positions are filled by good professional people. The members may substitute each other, and the size is enough to survive of losing a member. These teams are great to control each other and know enough about all of us. This size is when most people will say: the team cannot be successful without me and I cannot be successful without them.
Large teams
The typical issue in large organisations, that large, important projects appeals more and more people and the project team becomes large project organisation. Here the more people involved the more organisational politics come, the real source of frustration! In extreme cases the frustrating politics or self-serving tasks are eating up 2/3 of team resources. The result is no delivery, sometimes with "great" communication of mistakes, which will be shown as results.
How it can happen if the already referred Sun Tzu statement is right? - "The control of a large force is the same principle as the control of a few men: it is merely a question of dividing up their numbers." Do not forget that Sun Tzu assumed the full control of the forces, while it is typically missing in different organisational cultures. I would say luckily it is missing. There are other instruments in the management toolkit, like motivation through great targets, right people to get on board and so on, but it is beyond the topic of this article.
Anyway the best fitting team is when people are focusing on the real targets instead of internal fights, which is much more probable if the team size is around 10-15.
Turn a bit to the field of IT applications. There is an analogy between the size of applications and the teams, but in this case the "frustration" is the set of unnecessary functions. I saw often and often the case, when companies bought large "products" are full of functions, while significant percent of the functionalities were never been used. Those functions are slowing down all the changes, since no one can be sure about the potential side-effect of the never-used capabilities, therefore planning, design and the distress about the risks are growing and growing. Here the only exception is if you have a very professional team, who understands the main target of using the application and all of the functions inside and eliminates the negative forces.
Summary
You need the best fit size for all needs. Best fit is more simple if the team, or application is small; large teams and large applications are effective only if they are 100% keeping the real targets.