Your Subcontractors Are Costing You More Than Employees

The Math You're Not Running on Subs

March 2020. Card table in the garage. Hank's two, climbing my leg. Dad's in the hospital with a knee replacement that turned into six weeks of complications. Dispatch is me, a cell phone, and a legal pad.

I call three subs I'd been leaning on. First one's gone — he picked up emergency generator calls when COVID hit. Second one didn't answer. Third one called back four days later. Every one of them went where the fastest money was. Made sense for them. I get it. Nearly sank me.

The story I'd been telling myself about subs — lower risk, lower commitment, lower overhead — fell apart in about two weeks.

Here's the thing about Crestwood. I worked there from 2012 to 2018. Thirty trucks, Brecksville operation, real money. They used subs and it worked. But they had a contracts manager. A QC checklist on every sub job. A dedicated person who did nothing but manage sub relationships and inspection timelines. The reason it worked at Crestwood is exactly what a three-truck shop can't afford to build. I came home to three trucks and kept running their playbook without their infrastructure. That's not a plan. That's a mistake wearing a plan's clothes.

The math on subs looks clean until you actually run it. Invoice comes in, you apply your margin, you move on. What you're not running: the markup bleed on their materials. The scheduling gap you can't bill. The callback you eat at your own labor rate because the customer called you — not the sub. You.

Those three costs don't show up on the sub invoice. They show up in your margin at the end of the month when you're wondering where it went.

---

Where the Markup Bleed Actually Happens

The sub buys the materials. At his cost, with his markup to you. You apply your margin on top. If your sub is buying a $400 expansion tank and charging you $520, and you're marking to $680, your customer's paying $280 in combined markup on a part that cost $400.

I dropped American Home Shield and Choice Home Warranty in 2022. Revenue dropped that year. Net went up. I've written about this before — the number on the invoice is not the number that matters. Same logic on sub jobs. The invoice looks bigger. What you keep is smaller.

The callback math is worse. I had a job on West 117th last year — older home, slow drain call, turned into a sewer scope, turned into a real conversation about a collapsed clay line sixty feet under the side yard. I had a sub do the camera work. He came, he scoped, he left. The lady called me to talk through what the scope found. Called me when she had questions. Called me when she was scared of the repair number. Called me when she wanted to schedule.

I ate the diagnosis conversation, the estimate conversation, two follow-up calls. Two hours of my time on a job where I was already splitting margin with a sub. Her relationship was with me whether I sent my own guy or not. What I didn't have was any control over what the sub did or how he explained it to her.

---

The Scheduling Problem Is Yours, Not Theirs

When you book a sub for Tuesday, your customer thinks your company is showing up Tuesday.

When the sub's other job runs long and he's not there Tuesday, you're the one who gets the one-star review. On Nextdoor. Which in Parma moves faster than anything else I've seen — faster than Yelp, faster than Google. One no-show on a street over here and three neighbors have seen the post before I do. I can't get ahead of it. I can't explain that the guy who didn't show wasn't technically my employee. I made the promise. I own the no-show.

Becca runs the office two days a week. What that actually means is she holds dispatch together. She knows which jobs are running long. She knows when Dom's going to be late because she talked to him at 8 a.m. She catches the scheduling problem before it becomes a customer problem.

When a sub operates outside that system — and they always do, because they have their own dispatch and their own schedule — the whole chain breaks. Becca's making calls she can't answer. She doesn't know where the sub is. She doesn't know if he's going to make it. She's good at her job. I put her in a bad spot every time I drop a sub into a customer-facing day without her having any visibility into what that sub is doing.

---

Subs Are a Startup Move, Not a Long-Term Plan

When I came back in 2019, the shop was three trucks and limping. Payroll was tight. I couldn't carry a new W-2 tech and benefits. Subs kept trucks rolling without fixed cost I couldn't cover. That made sense then. I'm not saying it was wrong.

My dad built this shop starting in 1984. One truck, phone in the kitchen. Built it on personality and a Rolodex. The whole thing ran through his phone number. When that number stopped getting answered in spring 2020 — because he was in the hospital — the shop almost died. Not because we lacked good customers or good work. Because we'd built the whole thing on one man's availability.

Heavy sub reliance is the same problem at a smaller scale. Their phone goes to voicemail, you've got no one. You can't train around that. You can't schedule around it.

I'd tell my own kid not to take over a family business without a written operating agreement and a real buyout. I did it on a handshake. I love my dad. The handshake was a mistake. A handshake with a sub you trust works the same way — fine right up until he takes a bigger contract, or gets hurt, or calls you on a Thursday to say his rate's going up 30% on Monday. When that happens, you've got a customer on the phone and no answer for them.

Subs made sense when the shop was limping. They stopped making sense once I had recurring work, customers who expected to see our truck, and a name on the side of that truck that's mine.

---

When the Math Actually Flips — A Real Story

About two years ago, Old Brooklyn job. Near-west side Cleveland, late-1920s block, clay pipe neighborhood — the kind of street I know well enough that I can tell you what's underground before I scope it. Came in as a routine scope-and-quote. Camera found a bad section forty-two feet out. Partial collapse, roots in a hairline crack. Standard spot repair. Permit required, inspection required.

I had a sub lined up who'd done three or four jobs for me. Good work. Hard to schedule.

He started Tuesday. By Thursday he hadn't pulled the permit. I found out because the inspector called me — the sub had filled out the application with my company's license number and hadn't actually pulled it yet. Job was mid-dig. I'm not stopping a mid-dig job. Homeowner's got an open sewer lateral and a hole in her yard.

I pulled the permit. I sent my own guy Friday because the sub had another job and wasn't available. I paid the sub for what he'd done. Paid my guy for the rest. Paid the permit fee. Scheduled the inspection. Made the callback when she had questions about yard restoration.

That job cost me sub labor plus my own labor plus permit plus two hours of my time on the phone.

If I'd sent my own guy from day one, I've got one labor cost and full control of the timeline. That's the whole math problem in one job.

On the permit thing — I want to be clear because I feel strongly about this. Guys who skip permits are handing that risk to the homeowner. I know they know it. When the sub working under your license number skips it, it's not his license on the line in Ohio. It's mine. I found that out on a Thursday with a hole in somebody's yard. I don't work behind shops that do it.

---

What to Do Monday Morning

Pull your last six months of sub invoices. Sort them by job type.

Now find every callback on a sub job. Any ticket where you sent someone after the fact. Any call you personally made to smooth a customer. Any hour you or your guys spent finishing or fixing what a sub started. Put that labor cost on the original job ticket.

Then run that total against what a part-time W-2 tech would have cost you over the same six months. Part-time — 20 to 25 hours a week. Factor in FICA, Ohio workers' comp for plumbing, unemployment. Call your accountant and get the actual number. Don't guess it.

I'd bet the gap is smaller than you think. In shops I've talked to, some of it's already flipped the wrong way. The sub invoices feel like variable cost so they feel manageable. The W-2 feels like a commitment so it feels scary. But the variable cost is adding up in places you're not tracking yet.

If Becca ran this audit, she'd pull every job where the margin came in ugliest and work backward to figure out why. She'd want to see whether sub labor and materials were eating the job before we ever touched it. That list — job numbers, margin column, sub involvement yes or no — is your starting point. Not a meeting about it. Just the number.

If you've got more than eight or ten low-margin sub jobs in six months, you probably already have your answer. Get the number from your accountant. Then decide. But get the number before you decide anything.

---

FAQ

Is it actually cheaper to hire a W-2 employee than use a sub once you factor in taxes and workers' comp?

Closer than most owners think, and I'd rather you get your own accountant to run it for your shop than trust a number I throw out here. What I'll tell you from my own experience: when you add up FICA, Ohio workers' comp, unemployment, and basic benefits, the W-2 cost is meaningfully higher than base wages. But then you have to stack that against the sub's materials markup, the callbacks you eat, and the scheduling control you give up. On recurring work at decent volume — say, six or more similar jobs a month — the W-2 often wins. On true one-off specialty work you'd do twice a year, the sub still makes sense. The answer depends on your job mix.

What job types make the most sense to keep subbing versus bringing in-house first?

Keep subbing specialty work you don't have the tools or license for — HVAC, electrical, concrete. Bring in-house anything you're doing repeatedly, anything customer-facing, anything where your scheduling integrity is on the line. If a sub no-show on a given job type would get you a bad Nextdoor post before you even know what happened, that work shouldn't be subbed out.

How do I know if my sub relationship is working or if I'm just used to it?

Can your CSR tell you right now where the sub is on any active job? If no, the sub is operating outside your system. Also: when's the last time a sub job came in on schedule, under your cost estimate, with no callback and no customer call to you afterward? If you can't name a recent one, you're probably used to it, not satisfied with it.

What do I do about a sub who's good but chronically late or hard to schedule?

Stop booking them on customer-facing work. If they do good work, use them where the timeline is flexible — rough-in phases, non-urgent installs, anything where you haven't made a specific promise to a homeowner. Any job where a customer is waiting on a commitment, you need someone you can reach at 7 a.m. Good work doesn't fix a review about a no-show.

If I'm a one- or two-truck shop, does any of this apply to me yet?

Yes, but the volume threshold is lower. At one truck, you're really asking whether to hire your first employee, not your second. The callback math and the scheduling math still apply. What changes is the minimum volume where the W-2 pencils out — probably closer to four times a month on a specific job type rather than six. Run the same audit. Just adjust the bar.

How do I handle the conversation with a sub I've used for years when I decide to bring that work in-house?

Straight and early. Call them. Tell them you're bringing the work in-house because the volume justifies it. Thank them for what they've done. If they ask whether they did something wrong, be honest — it's not about their work, it's about your shop's structure. Most subs who've been around understand this. What they don't forgive is hearing it through someone else. Tell them before you post the job listing.