The “Bermuda Plan” or how to speed up project by removing people

Summary

The "Bermuda Plan" is a controversial project management strategy that involves reducing team size by removing underperformers to accelerate project completion, despite the human resource implications and the potential for short-term gains at the cost of long-term team dynamics.

Abstract

The "Bermuda Plan" is a project acceleration tactic that involves sending most team members on vacation, effectively reducing the team to a core group of top performers. This strategy is based on the premise that smaller teams with fewer coordination demands can work more efficiently, especially when underperformers are removed. It is a response to Brooks's Law, which states that adding more people to a late project only delays it further due to increased onboarding time, indivisible tasks, and a combinatorial explosion of communication channels. While the plan can lead to a short-term increase in productivity, it is not without limitations. It is primarily applicable to software engineering projects but may extend to other fields. The strategy is sensitive from a human resources perspective and can negatively impact team morale and dynamics. It also does not account for the potential positive impact of adding experienced individuals or those with critical skills. Alternative solutions include improving project organization, communication, and newcomer integration, or dividing the project into smaller, independent sub-projects.

Opinions

The "Bermuda Plan" is seen as a way to cut down on coordination costs and focus on delivery by retaining only the most productive team members.
It is considered a corollary to Brooks's Law, which suggests that increasing manpower on a delayed project will only exacerbate the delay.
The strategy is criticized for its potential negative impact on team morale and the sensitive nature of personnel decisions involved.
It is acknowledged that adding experienced individuals or those with essential skills can be beneficial and may contradict the implications of Brooks's Law.
The article suggests that the "Bermuda Plan" should be a last resort and that other methods of reducing coordination costs, such as improving project organization and communication, are preferable.
The author believes that dividing a project into smaller, independent sub-projects can be a more effective way to manage team size and efficiency than removing team members.

The “Bermuda Plan” or how to speed up project by removing people

The “Bermuda Plan” is a strategy of removing most people from a project (understand sending them on vacation to Bermuda) to allow the remaining core team to complete the project more quickly. The goal is to reduce team size and the coordination work at the same time, keeping the top performers focus on delivery and removing the under performers who are slowing down the projet.

Bermuda Beach is a great place to spend vacation

The “Bermuda Plan” is a corollary of Brooks’s Law. Fred Brook is a Software Engineer and the author of The Mythical Man-Month published in 1975, that says:

“Adding manpower to a late software project makes it later”

Innocently, you’d think that adding more people to a project would make it move even faster, but that’s not entirely true. First of all, doubling a team does not make the project progress twice as fast. Then, to some degree, adding more people to a project will slow down team velocity because of the coordination costs.

Here are Brooks explanations:

People already in the project need to onboard newcomers: this reduces team velocity until newcomers can start delivering independently and this period can last a few months depending on the complexity of the project ;

Some tasks cannot be easily divided, so adding more people won’t help: nine women can’t make a baby in one month ;

More people means more communication channels due to the combinatorial explosion: communication works in a network and not in a tree, adding a new nth manpower will n new connections (eg: going from 3 to 4 people means going from 3 to 6 connections, going from 4 to 5 people means going from 6 to 10 connections).

More people leads to more complex communication network

That’s why we invented the “Bermuda Plan”. When you reach the point where the coordination cost is greater than the value of the last resources, you can consider removing some people from the project and returning to your highest velocity.

Limitations

Brook was mainly talking about software engineering project, but I truly think this applies to every project.

Please note that the “Bermuda Plan” is a very short-term strategy. There are other ways to reduce coordination costs by improving organisation, communication channels and the integration of newcomers. Moreover, this is a very sensitive human resources decision, which can have a negative impact on the remaining team.

Brook’s law also considers all people from a project equal, but this is not true. Adding people with a lot of experiences or with a valuable knowledge about the project or with critical missing skills will most likely help to reduce the delay.

Of course, more people on the same project can slow down the project. Another solution is to divide the project into small independent projects to have a smaller and more efficient team on each sub-project.