You’ve got more work than you can do alone. The guy you’ve been calling for overflow is available more often now, and you’re starting to wonder whether he should just be on the payroll.
That question gets answered badly all the time, usually on cost alone, and usually in favor of keeping everyone a sub because a 1099 looks cheaper. It sometimes is. But the number on the invoice is only one of three things you’re deciding, and it’s not the one that gets people in trouble.
The cost comparison people run, and the one they should
Here’s the version most contractors do in their head. The sub charges $45 an hour. A guy on payroll would take $28 an hour. Sub is more expensive, employee is cheaper, done.
That comparison is wrong because it puts a fully loaded number next to a raw one.
The sub’s $45 already includes his truck, his tools, his insurance, his overhead, his downtime between jobs, and his profit. When the job is done, that number stops. There is no cost on the days you have nothing for him.
The $28 employee is a wage, not a cost. On top of it you carry the employer side of payroll taxes, unemployment insurance, and workers compensation. Comp is the wild card, and it swings hard by trade and by state. Roofing and structural work can run several times the rate of light interior work. Add whatever you provide in tools, truck, phone, and any benefits.
The realistic loaded cost of a $28 wage typically lands somewhere in the mid to high $30s per hour, and higher in expensive comp classes. Call it $36 for the sake of argument.
Now compare $45 to $36 and it looks closer. But you’re still not done, because $36 is only what he costs while he’s producing. Employees get paid for drive time, for the morning where the supply house has your order wrong, for the rained-out afternoon, and for the week in February where you’ve got two days of work and five days of payroll.
Divide by billable hours instead of clock hours and the picture flips again. If your employee is billable 75 percent of his clock hours, your effective cost is closer to $48. At 85 percent it’s about $42.
The honest conclusion: employees are cheaper per productive hour when you can keep them busy, and more expensive than subs when you can’t. Utilization is the whole variable. Everything else is noise around it.
Control is the thing you’re actually buying
The cost math is close enough that it rarely decides the question on its own. What decides it is how much control you need over how the work gets done.
With an employee, you set the schedule. You say be at this address at 7, wear this shirt, use this method, do it in this order, and go to that other job when this one wraps early. You train him your way. He answers to you all day.
With a sub, you’re buying a result, not a day. You define the scope and the deadline. He decides how and when to hit it. If you find yourself telling a sub exactly what hours to work, requiring him to use your tools and your process, and treating him as though he’s yours to schedule, you have an employee with the wrong paperwork.
That last sentence is not a philosophical point. It’s the classification test.
Classification is not up to you
Whether someone is a sub or an employee is a determination based on the actual relationship, not on what the two of you agreed to call it. A signed independent contractor agreement does not make someone a contractor if the day to day arrangement looks like employment.
The factors that get weighed, in plain terms:
Behavioral control. Do you direct how the work is performed, or just what the outcome should be? Set hours, required methods, and close supervision all point toward employment.
Financial control. Does he have his own tools, his own truck, real business expenses, and the ability to make or lose money on the arrangement? Does he work for other people? A sub who bills only you, uses only your equipment, and takes no risk looks like an employee.
Relationship. Is it project-based with an end, or ongoing and indefinite? Is the work he does the core of what your business sells?
Rules vary by state, and some states are considerably stricter than the federal standard. California’s ABC test is the well-known example, and several other states use versions of it. Construction sometimes gets carve-outs, sometimes gets extra scrutiny. Check your own state before you decide, because this is one area where getting it wrong is expensive: back taxes, penalties, and potential liability for an uninsured injury on your job site.
The insurance question that ends arguments
Ask any sub for a current certificate of insurance, general liability and workers compensation, before he sets foot on your site. Then verify it’s in force instead of filing the PDF and forgetting it.
Here’s why this matters more than the tax question. If an uninsured sub gets hurt on your job, there’s a real chance your policy ends up treating him as your employee for that claim. Your general liability carrier will also ask for sub certificates at audit, and uncovered subs typically get charged back to you at your rates. Contractors get surprised by that audit bill every year.
The rule that keeps this simple: no COI, no work. It feels rigid the first time you have to say it to someone you like. It stays cheap forever after.
A decision framework you can actually use
Run these four questions in order.
Can you fill 35 or more hours a week, most weeks, for the next six months? If no, use subs. Payroll during a dry stretch is the fastest way to burn through your cushion.
Do you need to direct the work hour by hour? If yes, hire. You cannot legally get that level of control from a sub, and the workaround is exactly what gets reclassified.
Is this the core work you sell, or a specialty you touch occasionally? Core, repeatable work belongs in-house over time. Electrical, HVAC, and licensed specialty work usually stays with subs regardless of volume.
Do you have the cash to carry a bad month? Employees need to be paid whether or not the customer has paid you. Two to three months of payroll in reserve is a reasonable floor before you take that on.
The middle path most crews actually take
The choice isn’t binary, and the honest answer for most growing crews is both.
A common structure: one or two employees who form the reliable core, handling the work you always have, plus a bench of trusted subs for overflow and for specialties. You get scheduling control where you need it and elasticity where the volume is unpredictable.
Expect the payroll side to grow slowly and deliberately. Hire when you’ve turned down work for three straight months, not when you have one busy week. And when you do hire, be aware that the loaded cost is only part of it. Onboarding, training, and the productivity dip while somebody learns your standards are real, and they show up in your margin before the new capacity does.
The tracking burden is where this gets ugly in practice, since managing a mixed crew means keeping sub agreements, certificates, hours, and payouts straight per job while still running the jobs. Punchlist tracks sub assignments and insurance expiration dates alongside the job itself, so you’re not finding out a certificate lapsed when the audit letter arrives. Join the waitlist to get it when it ships.