Track 4 — Capstone & Applied Practice

Module 4.3: Engineering Org Scaling

Scaling laws, hiring economics, organizational design, and span of control. How to grow an engineering org without destroying productivity.

3 Lessons~50 minAdvanced
1

Lesson 1: Scaling Laws for Engineering Orgs

Adding engineers doesn't linearly increase output. Brooks's Law, communication overhead, and coordination costs mean that doubling the team may only increase output 50-70%. Understanding these scaling laws prevents expensive mistakes.

Communication Overhead

Communication paths = n(n-1)/2. A 5-person team has 10 paths. A 10-person team has 45. A 50-person org has 1,225. Each path costs coordination time.

This is why team size sweet spot is 5-9. Beyond that: split into separate teams.
Ramp-Up Cost

New engineers are net-negative for 1-3 months: they consume senior engineer time for onboarding, code reviews, and mentoring while producing less output.

Budget: $30K-50K in lost productivity per new hire during ramp-up.
The "Mythical" Engineer

Management often expects adding 1 engineer = 1 unit more output. Reality: 0.5-0.7 units after communication overhead. This gap causes chronic underestimation.

Plan for 60-70% marginal productivity per additional engineer in a growing team.
📝 Exercise

Calculate your current team's communication paths. If you added 5 more engineers: how many new paths? What's the estimated coordination cost increase?

2

Lesson 2: Hiring Economics

Hiring is one of the most expensive activities in engineering. The total cost of a hiring mistake (wrong hire + replacement) can exceed $200K. Getting hiring right is literally an economic imperative.

Cost per Hire

Recruiting fees (15-25% of salary), hiring team time (interviews, reviews), onboarding costs. Total: $25K-50K per hire before the person writes a line of code.

At $180K salary: recruiter fee ($36K) + interview time ($5K) + onboarding ($10K) = $51K
Bad Hire Cost

Salary during underperformance (3-6 months), impact on team morale and velocity, severance, and re-hiring costs. Total: $150K-250K per bad hire.

The strongest signal for hiring quality: structured interviews with rubrics reduce bad hires 40%.
Time-to-Productivity

Months until a new hire reaches 80% productivity. Junior: 3-6 months. Senior: 1-3 months. Staff+: depends on organizational complexity.

Faster onboarding = lower cost: invest in documentation, tooling, and mentoring.
📝 Exercise

Calculate your total cost-per-hire (recruiting + interviews + onboarding). Then estimate the cost of your last bad hire if applicable. Build the business case for structured interviews.

3

Lesson 3: Organizational Design for Scale

How you structure teams determines what you can build. Conway's Law guarantees it. Intentional organizational design is the most powerful lever an engineering leader has.

Span of Control

Engineering managers should have 5-8 direct reports. Below 5: manager overhead per engineer too high. Above 8: insufficient coaching and development.

Optimal: 6-7 direct reports per EM. Add a new EM before exceeding 8.
Team Autonomy Index

How often does a team need another team's help to ship? High autonomy (80%+ independent): fast shipping. Low autonomy (< 50%): coordination bottleneck.

Target: each team can ship 80% of their features with zero cross-team dependencies.
Architecture vs. Org Alignment

Architecture and org structure must match. Misalignment creates friction, delays, and bugs. When scaling: redesign architecture and org together.

Audit: for each service, exactly one team owns it. Shared ownership = no ownership.
📝 Exercise

Draw your current org structure and your architecture diagram side by side. Do teams align with services? Identify misalignments and propose corrections.