Capacity Planning Metrics (2026) – What to Track & When to Act

Capacity planning metrics are workforce measures that compare a team’s available hours with booked and expected work. They are tracked per team and per period, usually a week or a month.

Each metric measures supply, demand, or the fit between them. Supply is the hours your team can work in the period. Demand is the hours already booked plus the pipeline work you expect to win. Fit metrics compare the two to show whether supply covers demand.

Resource planning differs because it assigns hours to named people and projects. These numbers cover the team’s total hours, before anyone is named. When you run the same arithmetic person by person, the practice is called resource capacity planning.

For a firm that sells time, the unit is people’s hours, not machine output. Server response time and production yield do not apply, so every formula counts hours.

The metrics that show whether your team can take on more work are available hours, weighted demand, utilization, capacity gap and bench time.

These numbers come before resource allocation and resource planning. You confirm the hours exist first, then decide who gets them.

Available hours are contracted hours minus leave and public holidays for the period. Team capacity for a week is the sum of every person’s available hours.

Internal meetings and other non-billable time stay inside available hours. They are booked work, and total utilization counts them. Subtract them from supply instead, and billable utilization looks higher than it is.

The denominator shrinks while billable time stays the same.

Some demand is not booked yet, such as a proposal due to start in three weeks. Weighted demand is confirmed bookings plus each pipeline deal’s estimated hours multiplied by its win probability. Win probability is the chance, in percent, that the deal closes.

The weighting keeps a 50% deal from counting at full size, which would overstate the workload. Leaving the deal out entirely understates it. Lead time (weeks from a deal closing to its work starting) shows which weeks the pipeline affects.

At two weeks, pipeline belongs in next month’s demand; at eight weeks, next month is confirmed work only.

All three utilization formulas share one denominator, available hours, so their results compare directly.

Total shows how full the team is. Billable shows how much of that time earns revenue. Forecast shows the same view for weeks that have not happened yet. Any of the three gets called a utilization rate, so name which one a report shows. Some teams call total utilization resource utilization.

Before you accept the next deal, the capacity gap shows whether the team has the hours for it. It is available hours minus booked hours minus weighted pipeline hours for the same period.

A positive gap means spare hours, and a negative gap means the team is short.

A negative gap more than a month out is a scheduling problem you can still fix by moving dates. A gap that stays negative across a quarter is a hiring question.

People with no bookings for a week are on the bench. Bench time is available hours with nothing booked, divided by available hours. Calculate it week by week and person by person. A team total can hide one person with no bookings for a month while colleagues are overbooked.

Some bench is healthy, because it covers unplanned absence, estimate overruns and deals that start early.

You calculate all three from available hours, booked hours and weighted pipeline hours. Each input must cover the same period, such as a quarter or a month ahead.

Example: an illustrative team of 10 people, each on 40 contracted hours a week, over a 12-week quarter.

For bench time, add up each week’s unbooked hours: 591 hours, or 12.8% of available hours. Netting the quarter instead gives 4,624 minus 4,060, or 564 hours and 12.2%. The 27-hour difference is two overbooked weeks offsetting idle hours in other weeks.

If both deals close at full size, the team needs 1,580 hours against 1,536 available. That is a gap of minus 44 hours and 102.9% forecast utilization.

Monthly averages hide overbooked weeks because a busy week and a quiet week cancel out in the month’s total. Dividing a month’s booked hours by its available hours blends them into one number.

Grouped into four-week months, the averages are 89.6%, 89.2% and 84.6%. The quarter’s total workload of 4,060 hours fits inside 4,624 available hours. Yet weeks 4 and 7 are over capacity by 14 and 13 hours.

Track overbooked weeks and weekly spread alongside the monthly average.

In the example, assume the 27 overbooked hours were worked as overtime hours to deliver the bookings. Check timesheets for the peak workload days in each overbooked week, and note which projects filled them.

Plan total utilization to 100% minus one weekly spread, or minus two spreads when overbooked weeks are costly. For the example team, a 10.4-point spread gives 89.6%, and two spreads give 79.2%.

Each level allows a predictable share of overbooked weeks, assuming weekly utilization is roughly bell-shaped. One spread below 100% leaves about 15.9% of weeks overbooked, or one in six. Two spreads below 100% leave about 2.3%, roughly one week a year.

The example team averaged 87.8%, just inside one spread, and had 2 overbooked weeks in 12. That is close to the one in six the rule predicts.

SPI Research’s 2025 benchmark of 403 firms found billable utilization fell to 68.9% in 2024. SPI treats 75% as optimal, and the top 20% of firms averaged 76.2%.

Those are billable figures, so set them against the example’s 77.4% billable figure, which clears SPI’s 75%. The 89.6% plan level is total utilization, which also counts internal time.

When finance leaders set a utilization target from margin, show them the weekly spread next to it. Clinic scheduling research shows that throughput (work delivered per week), extra hours and waiting time trade off.

Pushing utilization toward 100% raises both extra hours and delays. Plan to the spread, not the average: a target set on averages ignores the overbooked weeks.

When a number crosses its threshold, take the first action tied to it and set a date to recheck. The five triggers are overbooked weeks, a negative capacity gap, high bench time, hours past contract and estimate overruns.

Treat each threshold as a starting rule for your own data. Log each trigger and the action taken in your monthly management reporting.

The person who is over their hours is the bottleneck for every project they work on. Productive’s capacity indicators in the By Person view show each person’s booked hours against available hours.

In the Resource Planner, half-year and year zooms group time by week, giving visibility into resources across a quarter.

Hold the start date until resource planning has named people for it.

In Productive, you can book pipeline work as tentative bookings, which stay out of scheduled hours until confirmed. Switching on Include tentative factors them into the planner’s booked totals, showing the resource allocation if the deal closes.

Recurring bench for the same person points to a skills or sales mismatch. For skills, train or reassign the person; for sales, review whether the pipeline needs what they do. Give spare capacity internal work with an end date, so the person is free again when a deal closes.

Five percent over contract is about one extra day in a four-week month. Moving bookings to colleagues under 80% aims for balanced workloads across the skill group, checked on next month’s timesheets.

SPI’s research flags project overruns above 10% as damaging to delivery and client relationships. Project estimate accuracy (actual hours divided by estimated hours) shows how far off your estimates run.

Estimates set booked hours, so an estimate that runs 20% short books 20% too little.

You report capacity numbers by giving each stakeholder only the numbers behind their decision, at the cadence they decide.

Put billable and forecast utilization in the same monthly pack as the financial reporting they explain. Reading utilization beside financial reporting shows whether a margin change came from hours or from rates.

The monthly view then goes into management reporting, and the quarterly gap trend goes to whoever approves hiring.

Every number in these reports depends on timely actuals: timesheets submitted by the end of each week. A late timesheet leaves last week’s booked hours standing in for hours worked. Before each Monday review, count the timesheets still open.

In Productive, Pulse sends a saved report on a daily, weekly, bi-weekly or monthly schedule. It can go by email, Slack or Teams, to named people or whole teams. That lets each stakeholder get their own report at their own cadence.

An average shows whether the quarter fits; weekly numbers show whether each week does. Weekly numbers stay consistent only when available hours, bookings, leave and pipeline come from one system. Split across tools, each input updates on its own schedule, and capacity planning becomes reconciliation.

If you want those four inputs in one system for your Monday review, book a demo with Productive.

A good target for billable time is 75%, which SPI treats as optimal, against a 2024 average of 68.9%.

You subtract leave and public holidays from contracted hours, so the example team’s 4,800 becomes 4,624.

The difference is internal time, like meetings and admin, which total utilization counts and the billable figure excludes.

Review both weekly per person and monthly per team, matching the cadence each stakeholder decides on.

A weekly review needs five capacity planning KPIs: available hours, total utilization, overbooked weeks, capacity gap and bench time.