Capacity forecasting: how do you forecast your team’s capacity?

Capacity forecasting is the process of predicting how much work your team can take on in a future period, then comparing that against expected project demand. You calculate it by tracking each person’s available hours, subtracting time off and non-billable work, and measuring the result against the hours your pipeline and open projects require.

What is capacity forecasting?

Capacity forecasting predicts the resources you will need over the next few weeks or months based on current workload and pipeline. It answers one question: will the team have enough available hours to deliver what’s coming, or is there a gap.

It’s often confused with capacity planning, which is the broader exercise of building the team, budget, and tools to support that forecast over the next year or more. Forecasting is short-term and recalculated often. Planning is longer-term and revisited quarterly or annually.

The two concepts share the same underlying data: utilization rate, billable ratio, and the capacity gap between what’s forecasted and what’s available.

How do you forecast your team’s capacity?

Forecasting capacity comes down to four steps, repeated on a schedule rather than done once.

1. Calculate available capacity. Start with headcount multiplied by hours per week, then subtract paid time off, holidays, and non-billable admin work. A full-time employee typically loses 160-200 hours a year to PTO and holidays alone, according to a 2025 breakdown by billing platform TimeRewards [5], which is why most teams forecast against 34-37 available hours a week per person instead of 40.

2. Forecast demand. Pull hours from confirmed projects and weight probable pipeline by close likelihood, a technique also called resource forecasting. A deal at 30% probability should only count as 30% of its projected hours in the forecast, not the full amount.

Here’s what weighted pipeline looks like at the deal level:

Weighted, those three opportunities contribute 384, 80, and 180 hours, not 1,540. The 900-hour rebuild at 20% is the dangerous one. Booked at full value, it makes your development bench look full for the entire quarter and stops you from selling smaller work that would have filled it.

3. Compare demand against capacity. Subtract forecasted demand from available capacity to find the gap, by person, by team, and by week. A negative number means a shortfall; a large positive number means people will sit on the bench.

4. Adjust and repeat. Reassign work, delay a start date, hire, or flag the project to sales before the gap becomes a missed deadline. Then run the forecast again the following week, since new time entries, PTO requests, and closed deals all shift the picture.

Here’s what that looks like with numbers. A team of 12 consultants at 37 available hours a week gives 444 hours of raw capacity. After a typical 15% allowance for admin work and internal meetings, that drops to roughly 377 billable hours a week. If the pipeline forecasts 460 hours of confirmed and probability-weighted work for the same period, the team is forecasting an 83-hour shortfall, worth raising with hiring or a subcontractor before it turns into slipped deadlines.

That team-wide number is a starting point, not a finished forecast. A comfortable team-level total can still hide one role running at triple capacity while another sits idle, since the shortfall and the surplus cancel out in the average. For that reason, run the same math per role, not just across the whole team, and see predictive resource forecasting for professional services for how to catch what a blended number hides.

The 5 inputs every capacity forecast needs

The four steps above describe the process. The quality of that process depends on what feeds it. Five inputs carry the weight.

Input 1, available hours per person. Start with headcount multiplied by hours per week, then subtract paid time off, holidays, and admin time. This is the single input most spreadsheets get wrong.

Input 2, confirmed project workload. Map committed work to roles, people, and delivery timing. A single revenue number for Q4 tells you nothing about whether your two data engineers are double-booked in week 3. Capacity forecasting is about future workloads, not future invoices.

Input 3, weighted pipeline hours. Weight pipeline hours by close probability rather than counting full deal value, because unweighted pipeline overstates demand and hides real shortfalls. A 20% deal booked at full hours makes a starving team look busy.

Input 4, employee availability and current workload. Someone who is already at 110% this month is not a candidate for the new project, however good the skill match looks on paper. Failing to consider current workload when assigning new work is one of the most common resource management mistakes in professional services firms [2].

Input 5, time-tracking history plus what the team tells you. Blend qualitative input from the team with quantitative time-tracking data. History tells you the QA pass on a migration always runs 30% longer than scoped. The project manager tells you the client’s legal review will eat two weeks. Neither shows up in a resource sheet on its own.

Which forecasting method should you use?

Most professional services teams end up blending two approaches rather than picking one.

Qualitative forecasting relies on manager and team input rather than historical data. It works well for smaller teams or new service lines where there isn’t enough clean time-tracking history to model from.

Quantitative forecasting uses time series or regression on past utilization and project data. It needs at least 12 months of consistent time entries to be reliable, which rules it out for teams still tracking time in spreadsheets or not tracking it consistently at all.

Rolling forecasts update the numbers every one to two weeks as actuals come in, instead of building one static forecast per quarter. This matters more than which model you use, because a monthly forecast is already stale by the time a scope change or a new deal closes.

Corporate Finance Institute documents the underlying method families, straight-line, moving average, regression, if you want to formalize the maths behind your reruns [3].

Preparing for the quarterly planning session

CEOs preparing for quarterly planning should choose the right facilitator, set the right environment, and review the previous quarter’s performance. The performance review matters most for resourcing. If you consistently forecast 85% utilization and land at 68%, the correction belongs in your model, not in a pep talk.

The cost of getting this wrong shows up in industry numbers. Billable utilization across professional services firms fell to 66.4% in 2025, with Level 3 maturity firms at 72.5% and Level 5 at 81.2%, according to a summary of the 2026 SPI Professional Services Maturity Benchmark Report published by Deltek [4]. A meaningful share of that gap comes down to firms not seeing a capacity shortfall or surplus early enough to act on it, not a lack of available work.

How does software change the forecasting process?

Dedicated forecasting tools pull actual time entries and open pipeline into one live view instead of a spreadsheet that’s outdated the moment someone logs new time. Birdview’s resource and capacity planning software, for example, recalculates a resource’s forecasted load automatically when a new task is assigned or a time entry is approved, and flags that resource once their allocation crosses 100% of capacity for a given week.

The platform covers the five functional pillars a capacity forecast depends on:

• Capacity planning, role-level available hours after leave and admin time, not flat 40s
• Scheduling, assignments that show conflicts before they reach the client
• Leave management, approved time off flowing straight into available capacity
• Time tracking, the actuals that feed next quarter’s estimates
• Reporting dashboards, utilization by role and week, visible before the gap becomes permanent

Beyond those five, Birdview PSA adds portfolio dashboards, scenario modeling for “what if this deal closes” questions, and project financials tied to the same hours you resource against. That combination is why teams comparing tools land here.

In an analysis of more than 40 discovery and demo conversations with professional services firms between 2025 and 2026, roughly 80% described managing resource planning in spreadsheets, and about 60% said they couldn’t reliably forecast resource demand more than a few weeks out. Neither number is a software limitation; both describe a process that has outgrown a manual tool.

For a longer planning horizon than a single weekly forecast, see how to build an adaptive 30/60/90-day capacity forecast, which breaks the same process into tactical, operational, and strategic windows.

What mistakes make a capacity forecast wrong?

Counting full pipeline value instead of weighting it by close probability. This overstates demand and makes the team look shorter on capacity than it actually is.

Forecasting once a quarter instead of weekly or biweekly. By the time a quarterly forecast is reviewed again, PTO requests, new hires, and closed deals have already moved the numbers.

Ignoring non-billable time. A forecast built on 40 available hours a week instead of the realistic 34-37 will consistently understate the shortfall until it shows up as a missed deadline.

Leaving out input from the people doing the work. A resource manager’s view of who’s “available” often misses in-progress client requests or informal commitments that never made it into the project plan.

Treating the forecast as a one-time exercise. A capacity gap caught at kickoff is a staffing conversation. The same gap caught in week three of a six-week project is a missed deadline.

A capacity forecasting checklist for operations leaders

• Calculate available capacity as headcount × weekly hours, minus paid time off, holidays, and admin time
• Convert pipeline into role-level hours and weight each opportunity by close probability
• Rerun the whole forecast weekly or biweekly, not once per quarter
• Prepare the quarterly session properly: pick the facilitator, set the environment, review last quarter’s actual performance
• Leave slack in the plan, over-tight schedules and overworked people break delivery
• Check each person’s current workload before assigning them anything new
• Track utilization by role weekly, compare against the SPI 2026 benchmark median of 66.4% and Level 5 performance
• Ask every delivery lead what the numbers are missing, and put those answers in the model

FAQ

Who is responsible for capacity forecasting? Usually a resource manager, PMO director, or operations lead with visibility across every project’s staffing plan, not each project manager working from their own view. Individual PMs are typically responsible for flagging accurate demand for their own projects, which the resource manager then aggregates.

How often should you update a capacity forecast? Weekly or biweekly for active delivery teams. Monthly updates are too slow for teams billing hourly, since PTO, scope changes, and new sales can shift the numbers meaningfully within a few weeks.

Can you forecast capacity without dedicated software? Yes, for teams under roughly 10-15 people with a small number of concurrent projects. Spreadsheets become unreliable past that point because merging individual availability, time entries, and pipeline data manually introduces lag and errors.

What’s a healthy utilization rate to forecast toward? Most professional services benchmarks put healthy billable utilization between 70% and 80% at the team level, depending on the service line. Treat that as a starting target, not a ceiling to plan every role toward, since a healthy team average can still hide one role running at double or triple capacity. See predictive resource forecasting for professional services for why role-level targets matter more than the team-wide number.

Is capacity forecasting the same as resource forecasting? No. Capacity forecasting looks at aggregate available hours across a team. Resource forecasting looks at which specific person, with which skill set, is needed for a task. See capacity planning and resource planning: what is the difference for the full breakdown.

What is the SPI 2026 benchmark for billable utilization? The SPI Research 2026 Professional Services Maturity Benchmark reports a median billable utilization of 66.4% across 509 organizations. Level 5 firms achieve 81.2%, which is 42% more billable utilization than Level 2 peers at 62.7%. PSA users average 66.4% versus 63.5% for non-users.

How does weighted pipeline improve capacity forecasting? Weighted pipeline multiplies each opportunity’s estimated hours by its close probability. A 200-hour opportunity at 70% probability contributes 140 hours to demand, not 200. This prevents overstaffing for deals that will not close and surfaces the gap between committed work and probable work early enough to act.

Sources

1. SPI Research, 2026 Professional Services Maturity Benchmark (509 organizations; 2025 billable utilization 66.4%; Level 5 firms show 42% more billable utilization than Level 2 peers) – https://spiresearch.com/reports/2026-ps-maturity-benchmark/
2. PMI, Project Management Body of Knowledge (PMBOK Guide), Resource Management – https://www.pmi.org/standards/pmbok
3. Corporate Finance Institute, Top Forecasting Methods for Accurate Budget Predictions – https://corporatefinanceinstitute.com/resources/financial-modeling/forecasting-methods/
4. Deltek, “2026 PSO Benchmarks: Insights from SPI Benchmark Maturity Report” (2026) – https://www.deltek.com/resources/articles/professional-services-benchmarks/
5. TimeRewards, “Billable vs Non-Billable Hours: Complete Guide 2025” (2025) – https://timerewards.com/billable-vs-non-billable-hours/