They Weren't Losing Bids. They Were Winning the Wrong Ones.

Company

Cypress Fabrication Services

Location

Lafayette, Louisiana

Shop

10,000 sq ft on 30 acres

Certifications

ASME, ABS, DNV, AISC

Engagement

Fractional CFO

Focus

Costing, pricing, cash flow, concentration risk

The Challenge

Founded in 2014, Cypress Fabrication Services builds ASME pressure vessels, modular skid packages, and structural steel out of a 10,000-square-foot coded shop in Lafayette, Louisiana. The company had grown on oil and gas work — and when that activity slowed after COVID, the concentration became a liability. Revenue contracted, cash tightened, and the line of credit balance climbed and stayed high.

But the revenue drop wasn't the real problem; it was what the drop exposed. With less volume in the shop, no cushion was left to hide the fact that nobody could say which jobs actually made money.

What We Found

Material cost was never the issue. Cypress buys steel against specific projects, so those costs were clean and traceable.

Labor and overhead were a different story — and in a shop like this one, that's where the money hides. Cypress runs 27 welding machines, two overhead cranes with 200,000 pounds of combined capacity, and an automatic pipe welder, plus certifications — ASME, ABS, DNV, AISC, NBBI — that each carry ongoing cost in procedure qualification, third-party inspection, and traceability documentation.

None of that is project-specific. All of it has to be recovered through the labor rate. And the rate Cypress was bidding with hadn't kept pace with what the shop had become.

They weren't losing work. They were winning work that couldn't pay for itself.

When we rebuilt the rate, the real cost per shop hour came in meaningfully higher than what was being quoted. Every job won at the old rate quietly ate margin. And because the error was baked into the bid rather than showing up as a cost overrun, it read as a market slump instead of one mistake repeating.

Underneath that sat a timing problem. Cypress's larger skid and turnkey packages bill on milestones, so the shop bought steel and burned hours for weeks before cash arrived — financing its customers' projects on its own line of credit.

What We Did

1. Rebuilt the rate from history and forward growth

We reconstructed the cost picture from historical financials, then layered in where the business was heading. That second part matters: a shop rate is only as good as the hours you assume you'll sell, so a rate built on last year's volume misprices a company that intends to grow.

Then we taught it. A rate you're handed expires the moment your costs change. A rate you understand, you can rebuild yourself — and defend in a negotiation.

2. Rebuilt the bidding process around that rate

An ASME vessel doesn't carry the same labor and inspection content as a structural package or a skid build. Bidding all of it off one blended assumption guarantees some work subsidizes the rest. With accurate rates, bidding became a calculation with known margin — and unprofitable jobs get caught before the quote goes out.

3. Built a rolling 13-week cash flow model

Built around Cypress's real payment behavior — milestone dates, material lead times, payroll cycles, line-of-credit activity — the forecast let leadership see a squeeze forming weeks out, early enough to act.

4. Made milestone structure a bidding decision

Milestone structure isn't a contract detail to settle at the end of a negotiation — it's a financing decision. We trained leadership to map a project's cash curve during the bid, so milestones keep the job cash-flow positive instead of leaving the shop underwater until delivery.

5. Treated industry concentration as a financial risk

Oil and gas fluctuates wildly, and Cypress's dependence on it was why a downturn became an existential problem instead of a soft quarter. We advised deliberate diversification into other end markets — which the corrected rate made possible, because you can't price work in an unfamiliar industry until you know what an hour in your own shop costs.

The Hard Part

Knowing the right price and charging it are two different problems.

The owner was afraid that raising prices would cost him work — that he'd price himself out of contracts he needed, in a market that was already soft. It's the most common objection we hear, and it's a reasonable fear.

The opposite happened. The customers he lost were the ones who had been draining his margins: accounts that only ever made sense at a price that didn't cover his cost. What replaced them were customers who evaluated him on capability, code compliance, and reliability rather than on being the cheapest quote — and who treated him as a long-term partner instead of a vendor to squeeze.

He didn't lose the good customers. He lost the ones who were costing him money.

Raising prices didn't shrink the business. It grew it.

The Outcome

Cypress now bids from a rate it can stand behind, with margin engineered into the quote rather than hoped for. Profitability is visible at the job level, and milestones are structured so projects fund themselves rather than drawing down the line of credit.

The diversification advice held, too. Cypress today serves water treatment, maritime, utilities, renewable energy, food and beverage, and agriculture alongside oil and gas — so a slowdown in any one market is now a soft quarter rather than a threat to the business.

And the capability stayed in the building: leadership understands the labor rate, the overhead, and the cash position, and can work all three without us in the room.

The Takeaway

A demand shock in your end market will expose whatever your financials were already hiding. In a coded shop, what's usually hiding is a labor rate that hasn't been rebuilt since the last time you added equipment or earned a certification.

If your materials tie neatly to projects but your margins still don't make sense, look at labor and overhead — and look in the bid, not on the shop floor.

And if you suspect your prices are too low but you're more afraid of raising them, that fear is worth testing. The customers you lose are rarely the ones you wanted to keep.