Brooks's Law
Adding manpower to a late software project makes it later.
Tiny Summary
Brooks's Law states: "Adding manpower to a late software project makes it later." Communication overhead and onboarding costs more than offset the added capacity.
The Problem
Project behind schedule → Add 5 developers → Project gets MORE delayed
Why? New people need training, create communication overhead (n(n-1)/2 paths), and disrupt existing team.
The Mathematics
Communication Paths = n × (n - 1) / 2
5 people → 10 paths
10 people → 45 paths (+350%)
15 people → 105 paths
The Costs
Onboarding: 2-8 weeks to learn codebase, existing team loses 50% productivity mentoring
Communication: More meetings, coordination, context switching
Ramp-up: New hires make mistakes that slow others down
Example
5-person team, 10 weeks left:
Without adding: #################### = 10 weeks
Adding 3 people: #### (work) ## (onboarding) ############ = 14 weeks
Adding people late made it 4 weeks LONGER!
When It Works
Parallelizable work (independent features, no shared state)
Early in project (team structure not formed, simple codebase)
Specialized skills (adding DBA for database work, security expert for audit)
Long horizon (6+ months to amortize onboarding)
Handle Late Projects Better
Cut scope: Remove 20% of features → Ship on time → Iterate later
Extend deadline: Be honest about timeline
Improve efficiency: Remove blockers, automate testing, reduce meetings
Add people to future work: Current team finishes v1 → New hires work on v2
Key Insights
Small teams (3-5 people) move fastest due to minimal communication overhead. Amazon's "Two-Pizza Rule": if a team can't be fed by two pizzas, it's too large. If project >80% complete, adding people almost always delays it. Focus on removing blockers instead.
Use the simulation to see how communication overhead compounds!