Team Capacity Planning: A Practical Guide

Most teams don’t plan capacity — they discover they’re out of it. A quarter gets committed on optimism, two people quietly absorb the overflow, and the first hard signal is a missed deadline or a resignation. Capacity planning is the practice of seeing that coming.

This guide covers the model: how to turn headcount into real deliverable hours, how to weigh demand against it, and how to keep the plan honest once work is underway.

Start with the question, not the spreadsheet

“Do we have enough people?” is too blunt to act on. The useful questions are more specific:

• Can we deliver everything committed this quarter with the team we have?
• Is load spread evenly, or are two people carrying five people’s worth?
• If we win the deal in the pipeline, when exactly do we run out of room?

Each of those needs the same underlying number — realistic deliverable capacity — so that’s what to build first.

Calculating capacity

Raw headcount overstates what a team can do. Work down to deliverable hours in three steps.

Available hours  = scheduled hours − PTO − holidays − known non-project time
Deliverable hours = available hours × utilization target
Team capacity     = sum of deliverable hours across the team

Worked example — a team of six, one quarter

• Six people, ~480 scheduled hours each per quarter → 2,880 scheduled hours
• PTO and holidays across the team: 210 hours
• Standing internal commitments (on-call, recruiting, all-hands): 300 hours
• Available hours: 2,370
• Utilization target: 78%

Team capacity = 2,370 × 0.78 ≈ 1,849 deliverable hours

That ~1,850 is the number to plan against — not the 2,880 that a headcount-times-hours calculation would suggest. Committing to the bigger number is how teams end up structurally over capacity from day one.

Model demand the same way

Break the incoming work into estimated hours by week or sprint, and lay it against capacity. Three outcomes:

• Demand well under capacity — room to take more on, or to invest in internal work.
• Demand roughly matching capacity — fine, if estimates hold. Watch closely.
• Demand over capacity — decide now: cut scope, move a date, or add people. The one option that doesn’t work is hoping.

The value isn’t the snapshot, it’s the lead time. A crunch visible eight weeks out is a hiring decision. The same crunch discovered the week it lands is a crisis.

The visibility gap

Capacity plans drift because the inputs are guesses:

• Utilization target is assumed, not measured. If you plan at 80% but the team actually sustains 65% once meetings and interrupts are counted, every plan is 20% optimistic.
• “Non-project time” is underestimated. Support, context-switching and unplanned interrupts are real capacity consumption and rarely show up in the spreadsheet.
• The distribution is invisible. A team at 80% average can hide two people at 95%.

Calibrate the plan with activity data

The fix is to feed the model from what actually happened last quarter instead of what you assumed:

• Real utilization — measured active time on delivery work ÷ available hours, per person and team, so your target is grounded in this team’s actual sustainable rate.
• True non-project load — how much time genuinely goes to interrupts, support and coordination, so the deduction is real rather than a round number.
• Per-person distribution — who is consistently near the ceiling, surfaced while there’s still time to rebalance.

That’s what Optimetrix is built to provide: activity metadata rolled up into utilization and workload by person, team and week — the calibration layer that keeps a capacity plan from being fiction.

Keep it honest while work is underway

A capacity plan is a living document, not a quarterly ritual.

1. Re-check weekly. Compare planned load to actual activity. Where they diverge, find out why before the gap compounds.
2. Watch the outliers, not the average. The person 20 points above the team mean is the early warning.
3. Adjust commitments, not just effort. If capacity is genuinely short, the answer is scope or timeline or headcount — not asking the team to run hotter.

The bottom line

Capacity planning is comparing two honest numbers: what’s coming in, and what the team can actually deliver. Most of the failure modes come from inflating the second one — planning against headcount instead of deliverable hours, assuming a utilization rate instead of measuring it, ignoring the distribution. Get those inputs from real activity data and the plan becomes something you can commit against, and adjust from, instead of something you find out was wrong.