A firm doing two and a half million dollars at a twelve percent net margin hires its first real operator. Chief of staff, general manager, COO, the title varies. Call the package two hundred thousand dollars all in.
Twelve months later the margin is down to four percent and the owner is wondering what happened.
Nothing happened. The arithmetic did what it was always going to do, and almost nobody runs it before signing. Two hundred thousand against two and a half million is eight points of margin. Twelve minus eight is four. That isn’t a bad hire. It’s a hire that hasn’t paid for itself yet, which is a different thing and is the normal condition of the first year.
The mistake isn’t hiring. It’s not knowing which of those two you’re looking at when the margin drops, because they look identical on a P&L and they require opposite responses.
The number to work out first
Take the fully loaded cost of the seat. Not salary — salary plus payroll taxes, benefits, equipment, and any variable component you’d owe in a decent year. Divide it by revenue.
That’s the margin you’re spending, and it’s the number the hire has to give back before anything else counts as progress.
At two and a half million, a two hundred thousand dollar seat costs you eight points. At six million, the same seat costs three. At one and a half million, it costs thirteen, and thirteen points is more margin than most professional services firms have to give. That’s the arithmetic reason a business can be too small for the person it needs, and it has nothing to do with whether the person is good.
Owners think about this as a salary they can or can’t afford. The useful version is a margin they’re choosing to spend.
The standard I'd use
Here’s the test I’ve landed on after being the operator in a few different companies: the seat should be able to produce roughly four times its grossed-up cost in value. Below that, the hire is probably early.
Four times two hundred thousand is eight hundred thousand, which sounds absurd until you break it into where it actually comes from. There are four sources and they are not equally knowable.
Owner time freed, and you can size this before you sign. What is an hour of the owner’s attention worth pointed at selling, strategy, or the relationships nobody else can hold? At this size, five hundred dollars an hour is conservative. Free forty hours a month and you can write the number down in advance, because you already know both terms.
Instruments, also sizable. Most firms this size decide on lagging, partial or simply wrong numbers, and the cost shows up as leaks: work never billed, a service line losing money quietly, pricing that hasn’t moved in three years. Size it against the leak you actually have. I’ve seen making the numbers visible be worth ten points of margin — not on its own, but as the thing that showed where the ten points were.
Those two alone, on the model firm, come to about two and a half times the grossed-up cost of the seat — forty hours a month at five hundred dollars is two hundred and forty thousand, and ten points of margin on two and a half million is another two hundred and fifty. Four hundred and ninety thousand against a two hundred thousand dollar seat, before either of the sources you can’t put a number on. Which is the point: the case has to close on the half you can compute.
Capacity and delivery quality. More and better work with less labor, showing up as retention, repeat business, referrals. Real, and I’ve watched it happen. You can’t put a number on it in advance without inventing one, so it stays out of the tally.
Opportunities that surface. Once there are systems and the owner is thinking strategically, things appear: a service that should exist, pricing that should change, a segment nobody was serving. Also real, also not forecastable.
Clear the standard on the two you can compute. Treat the other two as why the estimate is conservative, not as numbers you’re counting on.
Whether there's anything there to claim
None of those four doors open unless the business has something behind them.
That’s what owners miss when a revenue floor looks arbitrary. It isn’t about affording the salary. It’s whether there’s enough surface area to recover anything: enough revenue that a few points of margin is real money, enough people that systems and accountability change output, enough transactions that pricing and leaks matter.
An eight hundred thousand dollar firm with two staff doesn’t have that surface, and no operator, however good, can manufacture it. There’s nothing to plug because there’s nothing leaking at scale. At two and a half million there usually is — enough revenue and enough people that the slack exists, and the job is finding it.
That’s the real test behind the floor. Not can you afford the seat. Is there enough loose value in the business for the seat to go and get.
Early or wrong
Assume you’ve cleared the standard and made the hire. The distinction that matters next is the one owners can’t make in the moment.
Early looks like: the margin is down and the metrics underneath it are moving. Collections up. The unprofitable service line identified and priced. Reporting where there wasn’t any. The owner doing less production work. None of it has reached the bottom line yet, because these things take two to four quarters.
Wrong looks like: the margin is down and nothing underneath it moved. More process, more meetings, better-looking documents, same numbers.
You can only tell these apart if you decided in advance what “moving” would look like. Without that, month nine is a mood, and moods in a compressed-margin business run toward panic.
I’d put it more plainly. If you can’t name the two or three operating metrics the hire is supposed to move, and where they stood the week they started, you haven’t hired an operator. You’ve bought relief, and relief has no measurement attached to it.
Structuring the deal so both sides can see
When the seat costs more margin than you’re comfortable with, the temptation is to shift pay into variable. Lower base, bigger upside. It usually doesn’t work, for three reasons.
Variable comp needs a floor. An operator whose base doesn’t cover their life optimizes for survival, and survival optimization is short-term by definition.
Thresholds set from the budget don’t pay. If the trigger is a level the business hasn’t actually reached recently, the variable component is decorative, and everyone knows it within two quarters. Set thresholds against demonstrated performance, then step them up.
Define the pool in writing before the period starts. What counts, what’s deducted first, who calculates it. A pool that can move after the work is done is a discretionary bonus with extra steps, and an experienced operator prices that in on day one by asking for more base.
Those three are what make the variable component a signal rather than a discount.
The thing that has to be true first
Every number above assumes you know your current margin, monthly, on a consistent basis, and that you know which two or three operating metrics actually drive it.
Most owner-led firms this size don’t. They know the annual figure and they know whether the account balance feels healthy. That’s not enough to run this calculation, which means it’s not enough to make the hire well, which means the hire gets made on feel and then evaluated on feel nine months later.
This is the first step of the method I use, the TAG method — See the Truth. Implement Actions. Achieve your Goals. Applied to a hire it’s unusually concrete. The truth is your loaded cost as a share of revenue and the operating baseline the day the person starts. The goal is what the seat has to give back and by when. The action is the hire itself, which is the part everyone starts with.
Do it in that order and the first year is legible. Do it in the other order and you’ll be reading a margin decline with no way to tell what it means.