The honest answer to in-house maintenance vs outsourcing is that no door count flips the switch. The breakeven is a volume-and-mix question: a full-time technician earns their keep when the routine, schedulable, unlicensed work your portfolio generates each month costs more to buy from vendors than it costs to employ someone to do it. Most of the models I see go wrong in the same two places. They compare a tech's wage to a vendor's hourly rate instead of comparing fully loaded cost to fully loaded invoice, and they assume a tech is productive for every paid hour. Fix those two and the answer falls out of your own work order history in an afternoon.
Set the comparison up correctly
Three inputs, and only three:
- The fully loaded monthly cost of one technician. Not wage. Everything.
- Your effective vendor cost per completed work order in the same category. Not the quoted hourly rate - invoice total divided by work orders closed.
- Your monthly volume of addressable work orders. Addressable is the word doing the heavy lifting.
Addressable means work a general maintenance technician can legally and competently finish: turns, punch lists, drywall and paint, faucets and toilets, disposals, appliance swaps, door and lock hardware, screens and blinds, detector checks, filter changes, seasonal preventive maintenance. It does not mean panel work, gas lines, refrigerant, roofing, structural repair, or anything needing a permit or a licensed contractor of record in your state. Pull twelve months of work orders and tag each one addressable or not. Most portfolios are surprised how much of their spend sits in the non-addressable pile, and that portion stays outsourced no matter what the math says.
What an in-house technician actually costs
Base wage is the smallest surprise in the number. The lines people forget:
- Employer payroll taxes: the 7.65% FICA match, plus federal and state unemployment at your own experience rating
- Workers' compensation, usually one of the larger add-ons for a maintenance class code and wildly variable by state
- Health insurance contribution and any retirement match
- Paid time off, holidays and sick leave - hours you pay for and do not get
- Vehicle: purchase or lease, commercial auto insurance, fuel and maintenance, or a mileage reimbursement policy plus a hired and non-owned auto endorsement
- Tools, ladder, meters, phone, uniforms, background check
- Certifications the work requires: EPA Section 608 for refrigerant handling, EPA RRP certification for renovation in pre-1978 housing
- Supervision and admin, because someone dispatches, reviews and approves this person
Whatever multiple those lines add to base wage, it will not be 1.0. Build your own factor instead of borrowing one.
The utilization discount nobody applies
Paid hours are not productive hours. Subtract PTO, holidays and sick days, then subtract drive time, supply house runs, waiting on access, callbacks and paperwork. Track it for one month with honest time entries. On a scattered-site portfolio the gap is brutal; on a single 200-unit property it is small. That one variable moves the breakeven more than wage does.
What outsourced vendors actually cost
The quoted rate is not the cost. Per completed work order you also pay for:
- The trip charge, or a one- or two-hour minimum on a twenty-minute job
- Parts at a markup rather than your supply house account price
- After-hours, weekend and holiday premiums, using their definition of after-hours
- The diagnostic visit followed by a second billed visit once the part arrives
- Scheduling latency, which is a tenant-experience cost even when it never appears on an invoice
The offsets are real too. With vendors you pay nothing in a quiet month, carry no workers' comp exposure for that labor, own no truck, absorb no idle time, and buy licensed capability on demand instead of hoping your tech qualifies.
The breakeven, worked through
Breakeven work orders per month = loaded monthly tech cost divided by (vendor cost per addressable work order minus in-house cost per addressable work order). Then sanity-check it against capacity: productive hours per month divided by average hours per work order.
Illustrative example. These are placeholder figures chosen to show the shape of the math - not market rates, not a quote, not a customer result. Replace every number with your own.
- Base wage $25/hour x 2,080 hours = $52,000. Apply a loading factor of 1.4 for taxes, workers' comp, benefits, vehicle and tools: $72,800 a year, about $6,070 a month.
- Paid hours less 200 hours of PTO, holidays and sick leave leaves 1,880. At 65% productive time that is roughly 1,222 productive hours a year, or 102 a month.
- In-house cost per productive hour: $72,800 / 1,222 = about $60.
- The average addressable work order takes 1.2 hours, so in-house cost is about $72. Vendor invoices average $175 for the same work once trip charges and parts markup are in. Difference: $103.
- Breakeven: $6,070 / $103 = about 59 addressable work orders a month.
- Capacity check: 102 productive hours / 1.2 = about 85 work orders a month.
Breakeven sits comfortably below capacity, so at 59-plus addressable work orders a month the hire pencils out with headroom. At 30 a month it does not, and enthusiasm will not fix it - look at a part-time tech, a tech shared across the management company, or a preferred vendor with negotiated volume terms.
Five things the spreadsheet does not price
- Licensing. State rules on what an employee may perform on managed property vary, and minor-repair exemptions do not extend to licensed trades. Confirm scope with your state licensing board before you budget it.
- Worker classification. Paying the tech on a 1099 to dodge the loading costs is the most expensive shortcut available in this decision.
- Response time. Repair-timeline and habitability requirements vary by state and city, and speed shows up in renewals rather than in the maintenance line.
- Coverage. One tech is one point of failure. Vacations, illness and resignations happen, and you still need a vendor bench on the day they do.
- Institutional knowledge. A tech who knows which building has the galvanized supply lines diagnoses faster every year. Real, and unbudgeted.
The split most small portfolios land on
Almost everyone who runs the numbers honestly ends up hybrid rather than pure:
- In-house: turns, punch lists, preventive maintenance rounds, cosmetic repairs, appliance swaps, hardware, minor plumbing and fixture work inside license limits, and first-response triage on everything.
- Vendors: HVAC and refrigerant, panel and circuit work, gas appliances, sewer and main lines, roofing, permitted work, work at height, capital projects, and after-hours coverage.
Triage is the underrated in-house win. A tech who can tell you in fifteen minutes whether it is a tripped GFCI or a genuine electrical fault prevents a lot of $175 vendor visits.
Run one system, whichever side of the line you land on
The failure mode of the hybrid model is that it quietly becomes two systems: in-house work in a texting thread, vendor work in email, and no single view of what a job actually cost. Plan@Job is built around the operator - the business that manages clients, job delivery, subcontractors and its own operatives, which is exactly the seat a property manager occupies when they run maintenance.
In project management, a job splits into internal stages, each with scope, dates, status and cost, each delivered in-house or subcontracted. The client sees the overall agreed scope and price plus the evidence you choose to share, while internal stage costs and subcontractor identity stay internal. That is what makes the comparison above measurable every month instead of a one-off exercise.
The AI staff sit behind it. Sophie Bennett checks job briefs for missing information and asks for clarification, handing off to a person after two attempts, which is how you stop a vendor arriving without a unit number or an access code. Daniel Reed supports eligible sharing of work with your existing vetted network, follows up on quotes and proposed dates, and produces a morning brief. Maya Collins prepares deposits and overdue payment actions for your approval. None of them awards work, sets a price or moves money on its own - those stay a person's decision, as they should when a vendor relationship is on the line. Useful for bench-building: in supported workflows subcontractors can price work before vetting completes, so due diligence sequencing does not block you from getting comparable numbers.
That is the goal for a small portfolio - in-house responsiveness at outsourced prices, with the same brief quality, follow-up discipline and stage-level cost visibility whether the work goes to your own tech or to a vendor. Plan@Job is live in the UK, and US property managers can register interest through the US property managers page. For the cost-modeling side of trade businesses generally, our sister site Construction Arbitrage covers the operator's version of the same math.
One next step: tag twelve months of work orders addressable or not, divide the addressable spend by the count, and you will have the two numbers that decide this.
FAQ
At what portfolio size does an in-house maintenance tech make sense?
There is no universal door count, and anyone quoting one is guessing about your portfolio. Density matters more than size: 150 units across three buildings produces far more productive hours per paid hour than 150 scattered single-family homes across a metro. Unit age, local vendor rates and turnover move it as well. Run the formula on your own twelve months.
Can I hire a maintenance technician as a 1099 contractor instead?
Only if the relationship is genuinely independent, which it usually is not. The IRS common law test weighs behavioral control, financial control and the nature of the relationship. If you set the schedule, supply the van and tools and direct the work, that is an employee whatever the agreement says. Several states apply stricter ABC tests, including California, Massachusetts and New Jersey. Misclassification exposes you to back taxes, penalties and an uninsured injury. A genuinely independent handyman with their own tools, insurance and other clients is a vendor - price them on the vendor side of the model.
How do I keep vendor pricing honest without in-house labor as leverage?
Three levers. Write the scope before the visit so quotes are comparable. Require more than one priced option above a dollar threshold you set. Track cost per completed work order by vendor and by trade over time rather than comparing headline hourly rates, because trip charges and parts markup are where the variance hides. Negotiate the trip charge, the minimum, the after-hours window and a parts markup cap in writing before the first job, not after a disputed invoice.
Published September 17, 2026.