Hiring an IT Person vs. Outsourcing: The Real Math for a 40-Person Company
The fully-loaded cost of an IT hire, what one person structurally can't cover, and the cases where hiring is still the right answer.
For a company of roughly 40 employees, the salary you pay an IT person is about half what that person actually costs you — and the question that matters isn't which is cheaper per month, it's what's covered. The honest comparison has to include the fully-loaded cost of the hire and the structural limits of one person, and it has to concede the cases where hiring is still the right answer. This article does that math for general business IT. Regulated environments — pharmaceutical, life sciences, anything carrying validation, audit or GxP obligations — change the calculation entirely and are scoped separately; see what drives pharma IT cost for that side of the ledger.
What does an IT hire actually cost?
See how your IT actually stacks up.
Twelve questions, about four minutes, and your score appears on screen immediately — no sales call required.
Start with salary, then add everything that sits on top of it. The U.S. Bureau of Labor Statistics reports a median annual wage of $109,470 for network and computer systems administrators (May 2025 data). A mid-level IT generalist in the U.S. — the profile most 40-person companies hire — sits in a band of roughly $90,000 to $120,000 depending on region and scope. Call the midpoint $105,000.
On top of base salary, the commonly used fully-loaded-cost multiplier is 1.25 to 1.4 times base — a figure cited by both Vena and Intuit's QuickBooks, covering employer payroll taxes, benefits and overhead. At $105,000, that puts recurring cost at roughly $131,000 to $147,000 a year before the one-time costs of actually getting the person in the door.
Building it up by line item:
| Line item | Low | High |
|---|---|---|
| Base salary | $90,000 | $120,000 |
| Employer payroll taxes (FICA 7.65% + unemployment) | $7,700 | $11,000 |
| Health, retirement, paid leave (~20–30% of salary) | $18,000 | $36,000 |
| Equipment and software for the role | $2,000 | $4,000 |
| Training and certifications | $1,000 | $3,000 |
| Recruiting (one-time, year 1) | $4,000 | $15,000 |
| Management and HR overhead | $3,000 | $8,000 |
| Total, year 1 fully-loaded | ~$125,700 | ~$197,000 |
| Monthly equivalent | ~$10,475 | ~$16,400 |
Cross-check: the multiplier range (1.25–1.4× base) gives $112,500 to $168,000 recurring, which sits inside the table once you strip out the one-time recruiting and first-year equipment costs. Either way, the salary is about half the number; the rest is real money leaving the business every month.
What does one person actually cover?
One person works roughly 2,000 hours a year. Subtract vacation, sick leave, training and the non-hands-on parts of the job — meetings, vendor calls, purchasing — and you're at something closer to 1,500 productive hours. That is a single body of coverage in an environment that runs 8,760 hours a year.
The structural limits follow directly. One person cannot cover nights, weekends or holidays without being on call continuously, which has its own cost in burnout and attrition. They cannot be in two places during an outage — the server room and the boardroom at the same time. And their knowledge is broad rather than deep: competent at most things, expert at few. Networking, security, cloud, identity, backup and endpoint are each their own discipline; one person is an informed generalist across all of them and a specialist in none.
What happens when they leave?
This is the risk nobody prices in, and it's the single strongest argument for outsourcing — though it isn't a financial one.
Undocumented environments are the norm, not the exception. Credentials live in one person's browser. Architecture decisions live in their head. When they resign, the replacement takes 60 to 90 days to recruit and onboard, and during that period the company has no coverage at all. The outgoing person's last two weeks are spent wrapping up, not transferring knowledge — because knowledge transfer requires two people and time, and by the time you're doing it, you have neither.
The cost of that gap — a single prolonged outage during the uncovered weeks — can exceed an entire year of the savings you thought you had by hiring.
What does the outsourced model actually buy?
Be precise rather than promotional. A retainer buys a team rather than a person, which is the only structural answer to the coverage problem above. It buys 24/7 monitoring and response, because a team rotates. It buys specialist depth across security, cloud, network and compliance, because a provider staffs those disciplines in depth rather than asking one person to approximate them.
It also buys documented environments — not because outsourced providers are more virtuous, but because they cannot function without documentation. Every ticket, every change, every system has to be written down or the next engineer cannot pick it up. And it buys continuity when someone leaves: their engineer leaves, and another one steps in with the same documentation. Yours leaves, and you start over.
The trade-off is control and proximity. The provider is not in the building. For most general business IT that's fine; for some companies it isn't, which is the subject of the next section.
When hiring is the right answer
Hiring genuinely is the better answer in four situations, and pretending otherwise would make this a sales argument rather than an analysis.
First, if you run heavy custom internal software — bespoke line-of-business applications, integrations nobody else understands, a codebase that lives in the building — an internal person who owns it day to day is hard to replace with a retainer.
Second, if you have a large facility with constant physical needs: wiring closets, a manufacturing floor, lab equipment, frequent desk moves. A remote team can do a lot, but someone on-site is faster.
Third, if you have unusual security requirements that demand someone cleared and on-premises — government work, defense, certain financial environments — the control model favors an internal hire.
Fourth, if you're large enough to staff a real internal team with depth across disciplines, the economics flip and the coverage problem solves itself.
There's also a non-financial case: some companies simply want someone in the building — a face, a name, a person who knows where the printer is. That preference is legitimate, and it's worth the premium to the companies that hold it.
What about doing both?
Most companies above 50 employees eventually land on a hybrid: an internal IT manager or systems owner who runs the relationship, sets priorities and handles the in-building work, with an outsourced team providing specialist depth and after-hours coverage behind them. The internal person becomes the single point of accountability and the provider becomes their bench.
This is usually the endpoint rather than the starting point. A company hires one IT person, hits the coverage wall, and adds the retainer behind them. It's the most defensible model for the companies that can afford both layers, and it's where the comparison in this article usually resolves.
How to actually compare two proposals
Before you compare a salary to a retainer, normalize them. The monthly numbers only mean something side by side once you've defined the same six things on both sides:
- Coverage hours — business hours vs. 24/7. An internal hire is not 24/7 at any price.
- Response-time definitions — what "critical" means in minutes, and what the tiers below it are.
- What "unlimited" excludes — most retainers say unlimited help desk; the exclusions are where the limits hide.
- Who owns licenses and tooling — whose tenant the software sits in, and what happens to it if the relationship ends.
- What's billed separately — hardware, projects, migrations, validation work.
- Continuity — what happens during a two-week vacation, or the day the only IT person resigns.
| Dimension | Internal hire | Outsourced | Hybrid |
|---|---|---|---|
| Coverage hours | ~1,500 productive hours/year; no nights/weekends | 24/7 monitoring and response | 24/7 from the team; in-building during business hours |
| Depth | Broad, not deep — one generalist | Specialist depth across security, cloud, network, compliance | Generalist on-site + specialist bench behind |
| Continuity risk | High — single point of failure | Low — provider absorbs departures | Medium — internal person is still one, but backed |
| During a two-week vacation | Largely uncovered | Unchanged | Handled by the outsourced team |
| During an outage | One person, one place | Multiple engineers, rotation | Internal triage + team escalation |
For reference, the general managed IT figures: a company of about 10 employees typically starts at around $2,000 a month, and companies with 15–50 employees typically invest $3,000–$8,000 a month. Scope determines where in that range a given company lands, and regulated environments sit outside it entirely — see what drives pharma IT cost. The detail on what's included is on our managed IT services page, and the small-business angle is on the small business industry page.
The decision
Run the table for your own company and the answer usually becomes obvious. If the fully-loaded cost of a hire is in the same neighborhood as a retainer and the retainer buys a team, 24/7 coverage and continuity, the retainer is the better allocation of the same money. If you have custom software, a heavy physical footprint, unusual clearance needs, or enough scale to staff a real team, hiring is the better answer. And if you can afford both, the hybrid is where most companies end up.
The question isn't which is cheaper per month. It's what happens at 2 a.m., during a two-week vacation, or the day your only IT person resigns. That's the question that decides this, and it isn't on the salary line of any budget.
If you operate in a regulated environment — pharmaceutical, life sciences, anything with validation, audit or GxP obligations — this whole comparison is the wrong frame, because the scope and the obligations change both sides of the math. That calculation is in what drives pharma IT cost.
Hero photo: "A balance scale with spheres on a marble pedestal against a grey background" by Myles Bloomfield on Unsplash.
Frequently asked questions
Don't see your question? Email info@isectra.com and we'll answer it directly.
Roni Banerjee is Chief Investment Officer at iSectra, where he evaluates the build-versus-buy decisions behind the firm's managed IT engagements.
Related reading
Get the next post in your inbox.
Signal, once or twice a month. No spam — just the posts worth reading.
Let's see what your IT could feel like.
A senior engineer will review your environment and respond within 24 hours. No sales script, no obligation.
