Deep-dive guide
Overtime vs New Hire Simulator: The True Cost of Covering Work With Extra Hours
Executive industry analysis
When work exceeds capacity, overtime is the easiest answer because it needs no recruiting, no contract and no commitment. The hours appear on the payroll next month and disappear when the peak passes. That flexibility is real and often worth paying for. The danger is that overtime is judged only by its pay line, which is the base wage plus a premium, while the larger costs arrive elsewhere: mistakes that must be redone, people who get ill or quit, and on-costs that follow the pay. A comparison that stops at the wage will always make overtime look cheaper than it is.
The structure of the problem is a trade between a variable cost and a fixed one. Overtime is variable: you pay only for the hours you use, at a premium. A hire is fixed: you pay for a full month whether or not the work is there, but the hour is cheaper and the person is rested. Which one wins depends on how many hours are needed, how long the need lasts and how quickly fatigue turns hours into errors. A small, short peak favours overtime; a large, lasting gap favours a hire, and the crossover is rarely where intuition puts it.
The domino cascade shows how it breaks. More hours spread over few people raises the load on each person, which lifts the error rate because tired people make more mistakes. Higher errors waste a growing share of the pay, and the same load raises the chance that someone burns out and leaves. Those exits push the real cost of a usable hour above its value, and the month result turns negative. Each stage feeds the next, so a rota that looks fine at fifteen hours each can fail at thirty, because fatigue grows faster than hours.
The iceberg separates the visible pay from the hidden cost. Above the waterline are the base pay and the premium, which every manager sees on the payroll. Below it are three costs that rarely get a name: the value lost to fatigue errors and rework, the cost of replacing people who burn out, and the payroll taxes and benefits that follow the extra pay. In a stretched team the hidden part can be as large as the premium itself. Use the simulator to find the load at which overtime stops paying, then decide whether to hire, share the load or cut the work.
Formula and methodology blueprint
The model works per month. Overtime hours are split across the people sharing them, which sets the load per person. The error rate starts at your base fatigue percentage and grows with that load, doubling at thirty overtime hours each and capped at 60%. Overtime pay is the base wage plus the premium. On-costs add 12% to overtime pay. Burnout adds 0.4 points of monthly quit chance for every ten overtime hours per person, and replacing someone costs three months of base pay.
- Load per person
- L = extra hours / people
- Overtime hours each person absorbs in the month, the driver of fatigue and burnout.
- Error share
- e = fatigue × (1 + L / 30), capped at 60%
- Share of overtime output that is wrong or redone. It doubles when each person works thirty extra hours.
- Overtime pay
- P = hours × wage × (1 + premium) × 1.12
- Base wage, premium and the 12% payroll on-costs that follow it.
- Burnout cost
- B = people × (L / 10 × 0.4%) × wage × 160 × 3
- Expected monthly cost of replacing people who quit under the load, at three months of base pay.
- Overtime result
- Net = hours × value × (1 − e) − P − B
- Monthly result of covering the gap with overtime. The break-even load is where this reaches zero.
Cascading elasticity
Value per hour and the error rate are the most elastic inputs, because they act on every hour at once and fatigue compounds with load. The premium and base wage are linear and matter in proportion to hours. Spreading the same hours over more people is powerful because it lowers fatigue and burnout together.
Three real-world case studies
Bootstrapped starter
A small shop covers a seasonal rush
A small shop with three people needs sixty extra hours for a busy month. Each person adds only twenty hours, errors stay low and nobody is close to burnout. Overtime is cheap and flexible here. Load it to see why a short peak is the right place for overtime.
Scaling mid-market
A large team shares a steady backlog
A twenty-person team shares three hundred extra hours at a good hourly value. Each person carries fifteen hours, so fatigue stays modest and the work is worth far more than its cost. Load it to see how spreading the load keeps overtime healthy at scale.
Black-swan crisis
Three people carry a permanent gap
Three staff cover two hundred and sixty extra hours month after month at double pay. Each works about eighty-seven extra hours, errors pass a fifth of output and people begin to quit. Load it to see why a lasting gap is a vacancy, not an overtime need.
Industry benchmark matrix
| Metric | Danger zone | Industry average | Top 10% elite |
|---|---|---|---|
| Overtime hours per person per month | Above 40 | 10–20 | Below 10 |
| Overtime share of regular hours | Above 15% | 3–8% | Below 3% |
| Rework share of overtime output | Above 20% | 8–12% | Below 6% |
| Cost of a good overtime hour against value | Above 90% | 55–75% | Below 50% |
| Duration after which a gap should become a hire | Beyond 6 months | 3–4 months | Within 2 months |
Variables and levers
Extra hours needed per month
Hours of work beyond normal capacity that must be covered across the whole team in a month.
Lever: Separate the permanent gap from the temporary peak; a gap that lasts more than a quarter is a vacancy, not an overtime need.
People sharing the overtime
Staff who split the extra hours. Fewer people means more hours each and faster fatigue.
Lever: Spread the extra hours over more people, because fatigue and burnout rise faster than hours when a few carry the load.
Base hourly wage
Normal hourly pay before any premium, the rate on which the overtime premium is calculated.
Lever: Cover peaks with the cheapest qualified hours first, and avoid paying overtime to your highest paid staff for routine tasks.
Overtime premium
Extra pay on top of the base wage for each overtime hour. Check current rules and your contracts for the legal minimum.
Lever: Use time off in lieu or flexible schedules where your rules allow it, so fewer hours attract the premium.
Value of one productive hour
What an hour of the extra work earns or saves, such as billed revenue or avoided penalties.
Lever: Spend overtime only on work that is worth more than its cost per good hour, and drop the rest.
Base error rate from fatigue
Share of overtime output that is wrong or must be redone at a moderate load. It rises as each person works more extra hours.
Lever: Cap hours per person and avoid long consecutive shifts, because error rates rise with every extra hour in a row.
How to use this simulator
- 1Enter the extra hours needed and how many people share them.
- 2Set the base wage and the overtime premium that applies to you.
- 3Estimate the value of a productive hour and the error rate at moderate load.
- 4Read the five dominoes and note where load, errors or exits turn amber or red.
- 5Open the Parallel universe tab to compare with hiring one person and share the link.
Mistakes to avoid
- Comparing wages only. The premium is the smallest part of the cost once errors, exits and on-costs are counted.
- Averaging hours across the team. Two people doing sixty extra hours each are in a different place from ten people doing twelve.
- Treating a permanent gap as a peak. If overtime has run for more than a quarter, it has become a vacancy paid at a premium.
- Ignoring idle time in the hire case. A hire costs a full month even if the work is thin, so size the gap honestly before choosing.
Frequently asked questions
When is a new hire cheaper than overtime?
When the gap is large and lasts. A hire has a lower effective hour and a rested worker, but also idle capacity if the work fades. The Parallel universe tab compares both over 24 months.
Is the overtime premium the same everywhere?
No. Minimum premiums and limits on weekly hours differ by country and contract. Check current rules and your agreements; the model only uses the percentage you enter.
Why do errors rise faster than hours?
Tiredness compounds. Attention and judgement fall with each additional hour in a row, so the same extra hour causes more mistakes when it is added to a heavy week.
How is burnout priced?
As a monthly chance of losing a person that grows with their overtime hours, times the cost of replacing them. It is an estimate, not a prediction for any individual.
What does the hire scenario assume?
Base pay plus 25% on-costs, about 82% of hours productive, no premium, and much lower fatigue. Idle time is paid for, which is why a small gap may still favour overtime.
How accurate is this?
It is an educational planning model with the assumptions shown, not legal, HR or financial advice. Replace the inputs with your real wages and hours.
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.