Capacity Planning with PTO & Holidays: How It Works
This page explains how Release Management, Roadmaps & Product Portfolio for Jira accounts for paid time off (PTO) and public holidays when calculating team capacity and forecasting release dates. It is a neutral, factual reference for how the capability works.
Why non-working days matter to a forecast
A release forecast is only as accurate as the capacity it assumes. If the model treats every calendar day as a working day, it overstates how much the team can complete and pulls the projected release date earlier than the calendar can support. PTO and public holidays are the most common reason a fully planned sprint or release slips, because they remove specific people on specific days rather than reducing the whole team evenly.
Tying PTO to real Jira users
PTO is recorded against the actual Jira user accounts on the team — the same accounts assigned to issues. When a person is marked as out, their availability is reduced only on the days they are away, so the capacity calculation reflects who is absent and when, not a blanket percentage applied to the whole team.
Importing public holidays in bulk
CSV import — a set of public holidays can be loaded from a CSV file rather than entered one date at a time.
Per-team application — an imported holiday set is applied to the relevant team, so those dates are treated as non-working days for everyone it covers.
Multiple regions — teams distributed across countries can carry different holiday calendars at the same time, and the plan respects each.
How capacity feeds the forecast
Working days are calculated as calendar days minus weekends, applicable public holidays, and individual PTO. That adjusted capacity is what the Monte Carlo forecast uses when it simulates how long the remaining scope will take. As a result, the p50/p85/p95 release dates reflect the team's real availability, and the burnup chart bends where the team is actually out instead of following a uniform slope.
Capacity adjustments are inputs to a probabilistic forecast, not a guarantee. They make time off visible so scope, capacity, and date can be traded off deliberately.
Where it runs
Capacity and forecasting run natively inside Jira Cloud on Atlassian Forge. No data is exported to an external service to perform the calculation.