Capacity Planning Is a System, Not a Spreadsheet

Every program manager I know has a capacity planning spreadsheet. It is always in Google Sheets or Excel. It is always maintained by one person. It is always slightly out of date. And it is always the first thing that breaks when plans change.

I have been there. I ran capacity planning on spreadsheets for years. Then I realized that the spreadsheet was the problem.

Why Spreadsheets Fail

Spreadsheets are static snapshots of a dynamic system. The moment you update one, the underlying reality has already shifted. Someone got pulled into a production incident. A new hire started. A team member went on leave. By the time you share the sheet in your Monday planning meeting, the numbers are wrong.

The second problem is collaboration. When capacity data lives in one person's spreadsheet, nobody else can update it, validate it, or trust it. You become a bottleneck — the human API that everyone queries for availability information.

A Better Approach

I built a simple system using Python and the Jira API. It pulls current sprint assignments, calculates remaining capacity based on team members' availability, and cross-references it with the backlog. The output is a live dashboard, not a static sheet.

The key insight was making it self-service. My engineering leads can check capacity for their teams without asking me. Product owners can see whether their feature request has realistic staffing before they commit to a deadline. I went from being the bottleneck to being the architect of a system that runs without me.

Building the Framework

The framework has three inputs: people (who is available and for how many hours), commitments (what is already planned), and demand (what is coming). It outputs gaps — where demand exceeds capacity — and surfaces them early enough to act.

This framework is now standard practice across my portfolio. It started as a personal automation and became an organizational tool. That is the trajectory of most useful things: they start by solving one person's problem and end up solving everyone's.

Stop updating spreadsheets. Build systems.