Deep-dive guide
Agency Capacity & Margin Stress Test
Executive industry analysis
A services business sells time, and time cannot be stored. Every hour a person is paid but not billed is gone for good, which makes utilisation the first lever of the whole model. A team of eight at sixty-eight percent utilisation bills about seven hundred and sixty hours a month, while the same team at fifty-two percent bills nearly two hundred hours fewer, a gap that no amount of cost cutting recovers. The simulator starts with capacity because everything downstream, from revenue to margin, is a multiple of the hours that actually get sold.
Scope creep is the quietest leak. When a client asks for one more revision or a small extra deliverable and the team says yes, the hours are worked and paid for in salary but never invoiced. Fourteen percent of delivered hours going unbilled sounds modest, yet it removes fourteen percent of revenue while costs stay exactly the same, so it usually removes more than a third of profit. The way to read the realisation node is as a pricing test: if a tenth of your delivered work disappears between the timesheet and the invoice, your real rate is lower than the one on your proposals.
Cash is the third constraint and the one that ends agencies fastest. An agency pays salaries monthly but is paid in thirty, sixty or ninety days, so its growth is funded by its own clients. Every extra month of payment delay ties up a month of revenue in receivables, and if that money is borrowed the interest is a direct cost. Late invoices are also more likely to become bad debt. The model makes this visible by charging a financing cost and a bad-debt reserve that grow with the delay, which is why shortening payment terms can raise profit without selling a single extra hour.
The scenario comparison tests a structural fix rather than a push for more volume. Fixed-scope packages with a deposit cut creep and payment delay sharply and justify a higher rate, at the price of slightly lower utilisation while the sales process adapts. In the twenty-four-month view this plan pays back its switching cost within a few months and then widens the gap, because margin improvements compound with growth. The breaking point tells you which of capacity, pricing or collections is closest to turning a profitable agency into a loss-making one.
Formula and methodology blueprint
The model computes one month of delivery from eight inputs. Hours, billing and cash are all derived from the same team capacity so a change in one input flows consistently through the rest. Constants for productive hours and reserves are typical agency assumptions.
- Billable and billed hours
- billable = team × 140 × utilisation; billed = billable × (1 − scope creep)
- 140 productive hours per person per month; creep removes hours that are worked but never invoiced.
- Revenue
- revenue = billed hours × billing rate
- Only billed hours earn money.
- Payroll and overhead
- payroll = team × cost per person; overhead as entered
- These do not change when hours go unbilled.
- Cost of waiting
- financing = revenue × (delay ÷ 30) × annual rate ÷ 12
- Receivables equal revenue times the delay in months; carrying them costs interest.
- Bad debt reserve
- revenue × 0.6% × (delay ÷ 30)
- The longer invoices stay open, the likelier some are never paid.
- Administration
- 3% of revenue
- Proposals, invoicing and account management time.
- Net outcome
- net = revenue − payroll − overhead − financing − bad debt − administration
- Break-even is where this reaches zero.
Cascading elasticity
Cascading elasticity: payroll and overhead are fixed, so every lost billed hour falls straight through to profit. A three-point drop in utilisation or a three-point rise in scope creep each removes thousands of revenue while costs stay put, which is why capacity and pricing usually top the sensitivity ranking.
Three real-world case studies
Bootstrapped starter
Bootstrapped starter
Two freelancers share clients and a laptop. With almost no overhead and invoices paid within a month, a modest rate and ordinary utilisation still leave a small, healthy profit.
Scaling mid-market
Scaling mid-market agency
Thirty-five people, higher rates and large overhead. The agency is profitable at high volume, but a few points of utilisation or scope creep now move tens of thousands each month.
Black-swan crisis
Black-swan crisis
A big client leaves, utilisation collapses, remaining clients stretch payments to nearly ninety days and borrowing costs rise. Load it to see what it takes to climb back to break-even.
Industry benchmark matrix
| Metric | Danger zone | Industry average | Top 10% elite |
|---|---|---|---|
| Billable utilisation | Below 55% | 65–72% | Above 78% |
| Realisation (billed ÷ delivered) | Below 80% | 85–92% | Above 95% |
| Payment delay | Above 55 days | 30–45 days | Below 20 days |
| Revenue per payroll unit | Below 1.3× | 1.5–1.8× | Above 2.2× |
| Net margin | Below 0% | 10–18% | Above 22% |
Variables and levers
Team size
People delivering client work, in full-time equivalents.
Lever: Do not hire ahead of demand: add people only when current utilisation holds above seventy percent for a full quarter.
Billable utilisation
Share of productive hours spent on client work.
Lever: Fill idle time with productised services and retainers, and move non-billable internal work into scheduled blocks.
Average billing rate
What you charge per billed hour, blended across the team.
Lever: Raise rates on new work every quarter; a small increase on the same hours flows almost entirely to margin.
Cost per person per month
Salary plus taxes and benefits.
Lever: Review the seniority mix; use senior time for scoping and review and let well-supervised juniors do production work.
Scope creep
Share of delivered hours that are never billed.
Lever: Write scope into every statement of work, price change requests separately and report hours against budget weekly.
Average payment delay
Days between sending an invoice and getting paid.
Lever: Invoice on milestones, take a deposit upfront, set late fees and chase at day seven, not day thirty.
Overhead per month
Rent, software, insurance and non-delivery staff.
Lever: Cut subscriptions and space that do not serve delivery, and review overhead against revenue each quarter.
Cost of financing
Annual interest rate on the cash tied up in unpaid invoices.
Lever: Shorten payment terms and consider invoice finance only as a bridge, never as a way of life.
How to use this simulator
- 1Start with the Realistic preset or enter your team and billing figures.
- 2Read the five nodes to see whether capacity, pricing or cash turns red first.
- 3Press Find my breaking point to learn how far each lever can move before margin reaches zero.
- 4Compare the fixed-scope plan with today's plan across 24 months.
- 5Copy the share link to discuss the scenario with your partners.
Mistakes to avoid
- Quoting from the sticker rate instead of the realised rate. If a tenth of delivered hours never reach an invoice, your real rate is a tenth lower than the one on your proposals.
- Hiring for the busiest month. Salaries are paid every month while work arrives in waves, so a team sized for the peak runs at low utilisation for most of the year.
- Treating slow-paying clients as a cash problem rather than a pricing problem. Every extra month of payment delay ties up revenue and adds financing cost, so charge for it or shorten the terms.
- Letting overhead grow in good months. Rent, tools and non-delivery staff are rarely cut when work dries up, which is why they decide how fast an agency drops below break-even.
Frequently asked questions
What is utilisation and why does it matter?
Utilisation is the share of productive hours spent on billable client work. Because salaries are paid whether or not hours are sold, every point of utilisation flows almost entirely to profit.
How do I measure scope creep?
Compare hours tracked against hours invoiced for the same projects. The difference as a share of tracked hours is your creep. Anything above ten percent is a pricing or contract problem, not a delivery problem.
Why does payment delay appear as a cost?
Because unpaid invoices are an interest-free loan to the client that you must fund. If you borrow to cover it, interest is a real cost; if you do not, it is the return you give up on that cash.
How accurate are the numbers?
They are a planning model with explicit assumptions rather than accounting. Use your own figures and read the output as a guide to which lever matters most.
What do the Monte Carlo results mean?
The tool reruns the scenario 1,000 times with every input varied randomly by around fifteen percent and reports how often monthly margin ends at zero or below, with the range of likely outcomes.
This is an educational planning model with stated assumptions. It is not financial, tax or legal advice. Replace the inputs with your own figures and verify important decisions with a qualified professional.