A founder decides they can afford a $90,000 hire because revenue supports $90,000. Six months later the runway chart is two months shorter than the model said and nobody can point to the line item that did it.
The line item is the hire. Salary is not what an employee costs. It is somewhere between 70 and 80 percent of what an employee costs, and the rest arrives quietly.
What sits on top of salary
For a US employee, the additions are fairly predictable:
- Employer payroll taxes. Social Security and Medicare are 7.65 percent of wages up to the wage base, plus federal and state unemployment. Call it about 8 to 10 percent all in for most employers.
- Health insurance. The largest variable. An employer contribution of $500 to $1,500 a month per employee is a common range, and family coverage sits at the top of it.
- Retirement match. If you offer one, typically 3 to 4 percent of salary for the people who take it up.
- Software and equipment. A laptop amortised over three years, plus the seats. Ten SaaS tools at $30 a seat is $300 a month before anybody has done any work.
- Payroll and benefits administration. Small per employee per month, but real.
Run that on a $90,000 salary and you land near $115,000 to $125,000 a year depending mostly on the health plan. That is the number that belongs in the model.
The rule of thumb worth carrying: multiply salary by 1.25 to 1.4 for a US employee. If your model says 1.0, your runway is wrong by the difference.
The part the multiplier misses
Two more costs do not show up as a percentage of anything.
Ramp
Almost nobody is productive in month one. For a specialist role, three to six months to full contribution is normal. You are paying the loaded cost throughout, so the first two quarters of a hire are closer to an investment than an expense, and the model should say so.
Management
Every hire consumes somebody's attention. In a company of eight, that somebody is usually a founder, and their hour is the most expensive hour in the business. This is the cost founders systematically forget, and it is the one that makes the difference between hiring one person and hiring three.
The test to run before you open the role
Three questions, in order.
1. How many months of runway does this hire cost?
Take the loaded annual figure, divide by twelve, and divide your current cash by the new burn. If a hire moves you from fourteen months of runway to ten, the question is no longer whether you need the role. It is whether you can raise or grow into it inside ten months, which is a different and much sharper question.
2. What breaks if you do not hire?
Write the answer down as a specific consequence with a date attached. If the honest answer is that things stay slightly harder than you would like, that is not a hiring case. If the answer is that a named customer commitment fails in November, it is.
3. Is this a job or a project?
A lot of first hires are really a project wearing a job description. Forty hours of bookkeeping a month is not a controller, it is a service. A one-off systems migration is a contractor. Full time employment is the right answer when the work is continuous, needs company context to do well, and you want the person to still be there in three years.
Contractor, fractional, or employee
These are not the same trade and the costs differ in shape, not just size.
- Contractor. Higher hourly rate, no benefits load, no ramp commitment, easy to stop. Best for defined scope with a finish line. Worth being careful about classification: the tests are about control and independence, and getting it wrong is expensive in back taxes and penalties.
- Fractional. Senior skill at a fraction of the time, a fixed monthly fee, no loading. Suits a function you need done well but not forty hours a week, which is most finance work under about $10M in revenue.
- Employee. Lowest hourly cost at full utilisation, highest fixed commitment, and the only option that really builds institutional knowledge. Right when the work is genuinely continuous.
The mistake is comparing a contractor's hourly rate to an employee's salary divided by 2,080. Compare it to the loaded figure divided by realistic productive hours, which after holiday, sick leave, meetings and ramp is closer to 1,600. That comparison usually looks different.
Build it into the model, not the spreadsheet margin
The practical fix is small. Put the loading factor in your hiring plan as an explicit assumption rather than a mental adjustment, model the ramp so the first two quarters cost more than they return, and re-run runway with the new burn before you post the role rather than after the offer is accepted.
A hiring plan that shows loaded cost and the runway impact of each role is a document you can make decisions from. One that lists salaries is a wish list.
We build hiring plans as part of the forecasting work for the companies we support, and there is a hiring plan template on the tools page if you want to start on your own. If the question is really whether the role should be a hire at all, that is worth a conversation.