The Gross Margin That Could Not Survive a Hire

The Engagement

Client: Founder-led professional services firm. Small team, a combination of employees and contractors.

Focus area: Gross margin and offer economics.

Timeframe: Three months.

How It Came Up

She wanted to hire, and she wanted to pay herself more.

Neither one was going to happen first. She had drawn on her cash reserve to cover a gap, and rebuilding that runway came before any decision that added ongoing cost. That is the order I work in. Cash determines whether a business can carry a new commitment at all. Gross margin determines what it can sustain, and whether a hire will ever pay for itself.

There was a longer reason too. She is building toward a business that does not require her to deliver the work. That is the direction she has chosen, and it means the margin question had to be answered eventually no matter what. The hire only made it urgent.

So we rebuilt cash first. While we did, her pipeline grew, and the projects coming through were the largest she had sold. The gross margin on her profit and loss looked strong enough to support both the hire and the raise.

That margin included none of her own time. She delivers most of the client work herself, and her pay sits in operating expense as owner's salary rather than in cost of sales, which means her hours never enter the cost of delivery. Every hour she worked came through as margin. The bookkeeping was correct. It could not tell her whether the work would generate enough profit once someone else was doing it.

She had been pricing her proposals on what she believed the work was worth. There was no calculation underneath it. That is common, and it is exactly why the number had never been tested.

What We Did First

We did not reprice anything. We tested it.

She picked one live project, two phases. Her whole team tracked hours against it, including her. That last part is what made the test work, because her hours were the missing input and no estimate would have been credible enough to price from.

Then we had to determine what her time was worth to the business, which is a different question from what she pays herself. What we needed was the cost of having someone else do her client-facing work. We worked it out together, against what she had been paid at that level before.

Then I presented the margin two ways. The first was gross margin as it appears on her profit and loss, which counts only what she pays other people. The second was the effective gross margin, the same project with her own hours costed in. I put them side by side and let her look at them.

What The Test Showed

On the profit and loss, the project carried a gross margin of roughly 80 percent.

On the effective gross margin, phase one came in at 52 percent. Phase two came in at 47 percent as currently staffed. Then we added the senior hire she was planning, and phase two fell to 25.7 percent.

For a service business I hold 60 percent effective gross margin as the floor. Not the target. The floor. Every one of those numbers was under it, and the gap between 80 and 47 is the entire distance between what her financials could show her and what was actually true about the work.

Phase two had been underpriced from the day it was sold. The same test set a ceiling on what she could pay a senior hire before the work stopped covering its own delivery, which answered a question she had been stuck on for weeks.

What We Changed

She now prices on the effective gross margin rather than the one on her profit and loss. The reported margin still runs the cash conversation, because that is the money that actually moves.

Time tracking became permanent. Hours are budgeted by role at the time of bid rather than reconciled after the fact, which gives her team clear boundaries and gives her a way to see whether people are being properly utilized.

We also agreed that the training period on a new hire would run below the floor on purpose, and that this is an investment rather than a pricing failure. Naming that in advance mattered, because otherwise the first project with a new person on it looks like a mistake.

What We Built

She was not going to call me every time she priced a project, so I built her a custom pricing calculator. She enters the phases, the hours by role, and the rates, and it returns both margins side by side with the cash timing underneath.

I built it with AI, which is why it took days instead of a quarter. I host it, and it is hers for as long as she is my client. It is not a template. It was built for her business, her cost structure, and where she is trying to go.

One decision inside that build is worth explaining. The calculator lets her model a proposal across a range of margins rather than only at my 60 percent floor. She was not ready to jump straight there. Pricing at 60 felt like a number that would cost her the work. So I gave her the ability to see the same proposal at 40, knowing what she would find, which is that the difference between the two numbers was smaller than the fear attached to it. Seeing that gap for herself is what moves someone to price at 60. Being told to price at 60 does not.

That is the line I hold on AI. AI can read your P&L. It can't sit in the seat.

Hand it her financials and it will tell you her gross margin is 80 percent. It will be right, and it will be useless. Effective gross margin is not in the data. There is no account for it, no line to read, nothing to pull. It exists only because someone decided the founder's own hours belonged in the cost of delivery and then went and collected them. AI would not have decided that. It would not have raised it as a possibility. It would have produced a clean and accurate analysis of a number that could not answer her question, and she would have hired against it.

What AI did was build the tool, after the thinking was done. That is the right division of labor. My interpretation is not automated. My availability is.

Where It Stands

She priced a corporate proposal from the calculator, defended the number on explicit hours by role, and won the work.

The hiring search is proceeding with a rate ceiling she can defend, and she now knows before she signs a proposal whether the work will pay for itself.

Client details have been changed. The financial mechanics and the sequence are as they occurred.