Before You Hire: The Capacity Math
For agencies of 5 to 15 people on retainers. Whether the team is full, or the hours are leaking into scope and low-margin clients, and what a hire costs before it helps.
- By Vinicius Diniz
- Published October 6, 2026
- Methodology
A team that feels full is a reason to consider a hire, and not always the right one. Before a new role, three months of hours by client answer a simpler question: is the team over its line, or is the work going to places that do not pay for it? The math below uses the same example as the client hours calculator.
What full looks like
A healthy agency is never booked near 100%. Parakeeto, an agency profitability consultancy, puts healthy utilization at 50% to 60% for the whole agency over a year, and 60% to 80% for people who mostly do client work.
50% to 60%healthy utilization for the whole agency over a year; 60% to 80% for people who mostly do client work.
Source: Parakeeto, utilization rate calculator. See all benchmarks
A team running above those ranges is not unusually hardworking. It is past the point where more work turns into later work.
Three reasons a team feels full
"The team is full" can mean three things, and only one of them is solved by hiring: client work above the team's line, work nobody pays for, or clients that take more hours than their retainer covers.
Out-of-scope work is a frequent leak. In Ignition's 2025 survey of 273 US agency managers and executives, 57% said they lose $1,000 to $5,000 a month to work outside the agreed scope, and 78% rarely or only sometimes bill for it. Those hours fill the calendar without filling the invoice.
Source: Ignition, 2025 Agency Pricing and Cashflow Report, 273 agency managers and executives, Apr 2025. See all benchmarks
Why 100% is not the goal
Booking a team close to 100% makes every delay longer, because there is no slack to absorb the next request. Queueing math shows waiting time rising slowly and then steeply past about 80% of capacity.
In the simplified version of the queueing result (Kingman's formula, 1961), waiting time at 80% occupancy is four times what it is at 50%, at 90% it is nine times, and at 95% it is nineteen. The model comes from factories and service desks, not agencies, so the exact numbers do not transfer; the shape does.
Longer hours do not rescue it. In records from British munitions factories during the First World War, Pencavel (Economic Journal, 2015) found output rising in step with hours up to about 49 hours a week, then less with each extra hour; output at 70 hours was close to output at 56. And the cost lands on the people who stay: in a 2018 Gallup survey of nearly 7,500 full-time employees, those who felt burned out very often or always were 2.6 times as likely to be looking for another job, and those who often had enough time to do their work were 70% less likely to experience high burnout.
Source: Gallup, Employee Burnout, Part 1: The 5 Main Causes, Nearly 7,500 full-time employees, article of July 12, 2018 (workers in general, not agencies). See all benchmarks
The math
In the example agency, one hire adds about 139 bookable hours a month. Whether that is needed depends on the team size, and on the 55 out-of-scope hours sitting in the same month.
| Team | Bookable hours a month | Client hours | Share of the line |
|---|---|---|---|
| 7 people | 971 | 590 | 61% |
| 5 people | 693 | 590 | 85% |
| 4 people | 555 | 590 | 106%, 35 hours over |
Assumptions
- Client hours last monthExample agency (client hours calculator)
- 590
- Of those, outside the agreed scopeExample agency
- 55
- Hours per person per weekPlanning assumption
- 40
- Bookable linePlanning assumption, from queueing math
- 80% of contracted hours
Result
- One more person adds
- 40 × 52 ÷ 12 × 80% = 139 bookable hours a month
- 4-person team, out-of-scope work billed or cut
- 535 of 555 hours, 96% of the line, without hiring
The 4-person team is the only one over its line, and it is over by 35 hours. The same month has 55 hours of out-of-scope work. Billing or cutting that work first puts the team back under the line; a hire would add 139 hours to fix a 35-hour problem.
A rule of thumb circulates in agency advice: hire when a role runs above 32 to 35 billable hours a week for three months. No published study backs those numbers. Three months of real hours against the line is a better test.
What a hire costs, and how long it takes
A hire takes about six weeks to reach an accepted offer, costs from about $1,200 to several thousand dollars before the first day, and takes longer still to carry a full load. That is why the decision needs three months of data, not one busy week.
In SHRM's 2025 benchmarking of US employers across industries, the median time to fill a nonexecutive role, from opening it to an accepted offer, was 44 days. The median cost per nonexecutive hire was $1,200 and the average $5,475, because a few expensive hires pull the average up. Neither number includes the weeks before the new person carries a full load, which no study measures for agencies.
Source: SHRM, 2025 Recruiting Executives Benchmarking data brief and press release, 2,371 SHRM members, January 9 to March 3, 2025 (US employers across industries, not agencies). See all benchmarks
A hire also has to be planned against revenue that may not be there. In Productive's 2025 survey, 45% of agencies said they could forecast revenue only one to three months ahead.
Source: Productive, State of the Agency Industry Report 2025, 93 agency leaders (agencies, consultancies and creative studios worldwide, $1M to $50M in revenue; Productive sells agency software). See all benchmarks
If the role being considered is someone to sell, not someone to deliver, the math is different, and it is covered in pipeline and referrals for agencies of 3 to 15 people.
Where the calculator helps
The client hours calculator turns last month's hours by client into margin, out-of-scope cost and room under the line, which are the three numbers this decision needs.
Run it with each of the last three months. If hours are not logged by client yet, start with how to ask the team to log hours; a month of data is enough to see the shape, and three are enough to decide.
When not to hire
A hire is the wrong move when the team is under its line, or when the hours over the line are smaller than the hours given away.
Skip if
If client hours stayed under 80% of the bookable line in each of the last three months, do not hire. The team is not full; the work is uneven.
If the hours over the line are fewer than the out-of-scope hours, bill or cut the scope first, then measure again.
If any client is below a 60% delivery margin on its hours, fix that client before adding anyone. A new person would mostly absorb that client's overruns. Parakeeto puts a healthy single project at 60% to 70%, higher than the line for the whole agency; 60% is the low end.
Next step
This afternoon, run the client hours calculator for each of the last three months and write down three numbers per month: hours used against the line, out-of-scope hours, and clients below 60% margin.
How capacity fits with margin and scope is covered in agency operations for 5 to 15 people.