What an IT hire really costs once benefits, tooling, and coverage gaps are counted — plus the co-managed model most businesses land on, and a straightforward framework for deciding by company size.
Somebody on staff has become the computer person. It was never their job. They were good with technology, so people started asking, and now they spend six hours a week resetting passwords and fighting a printer instead of doing the work you actually hired them for. Eventually somebody says out loud: maybe we should just hire an IT guy.
That is the right instinct and often the wrong first move. The comparison most businesses run is a salary against a monthly fee, which makes the hire look expensive and stops there. The comparison that matters is broader in both directions — the loaded cost of employment on one side, and the things an employee gives you that no outside provider fully replaces on the other.
We are an outsourced IT provider, so treat our conclusions accordingly. But we have also helped businesses write the job description, sat in on the interviews, and then supported the person they hired. The genuinely honest answer for most companies between 25 and 100 people is neither column of this comparison — it is both, split deliberately. That is the part of the conversation the industry tends to skip.
The Number Behind the Number
A realistic build-up for one mid-level IT generalist in western Kentucky. Figures are illustrative regional estimates — your benefit load and recruiting costs will differ — but the structure holds everywhere.
| Cost line | Annual | Notes |
|---|---|---|
| Base salary — mid-level IT generalist, western Kentucky | $62,000 | Desktop, network, and light server work. Security depth costs more. |
| Employer payroll taxes (FICA, FUTA, SUTA) | $5,300 | Roughly 7.65% federal plus state unemployment |
| Health, dental, retirement match | $11,000 | Commonly 15–25% of salary for a small-group plan |
| Workers' comp, payroll admin, HR overhead | $1,800 | Small but real, and it scales with every hire |
| Recruiting and onboarding, amortized | $3,000 | Higher if you use a recruiter; IT roles sit vacant a long time here |
| Tooling: RMM, backup, ticketing, documentation, security stack | $7,200 | One-seat pricing is brutal — MSPs buy this at volume |
| Training, certifications, conference time | $2,500 | Skip it and the skills go stale in about three years |
| Workstation, phone, vehicle allowance, desk | $2,200 | Annualized over the useful life of the equipment |
| Fully loaded, year one | ~$95,000 | Before any hardware, licensing, or project spend |
The salary is about 65% of the total. Everything else is the part that gets left out of the mental comparison — and the tooling line is the one that surprises people most. Remote monitoring, patch management, backup, ticketing, documentation, and a security stack cost far more per seat at one seat than they do at a thousand. A managed provider spreads that across every client; a single employee cannot, so in practice most small in-house IT departments simply go without several of those tools.
Note also what $95,000 does not buy: coverage outside working hours, expertise in every discipline, or a backup plan for the two weeks that person is away. Those are not budget items. They are structural limits of having one of something.
The Structural Problem
None of these are about the quality of the person you hire. They apply equally to an excellent one.
Even before vacation and sick time, a single full-time hire is present for under a quarter of the calendar. Failures do not observe business hours, and neither do ransomware operators — the majority of encryption events start on a Friday evening or a holiday weekend precisely because nobody is watching.
Most small businesses have no plan for the week their IT person is at the lake. The honest version of the plan is usually "hope nothing happens," and it works right up until the server does not come back after a storm.
Networking, Windows Server, Microsoft 365 administration, firewall and VPN configuration, endpoint security, backup architecture, and your line-of-business application are seven specialties. A good generalist is genuinely strong in two or three and is guessing on the rest — which is not a criticism of them, it is arithmetic.
Undocumented environments are the norm, not the exception, because documentation is the first thing to slip when one person is also the helpdesk. The cost of that shows up on the day of the resignation, when nobody knows what the firewall password is or which backup is the real one.
The Other Direction
An outsourced provider should be honest about this rather than pretending the gap does not exist. Somebody who sits in your building has advantages that no service agreement fully reproduces.
They notice things. The printer that has been making a noise, the machine in shipping that takes four minutes to boot, the person in accounting who has started writing passwords on a sticky note. None of those generate a ticket, and all of them become tickets eventually. They also absorb the two-minute question — the one nobody would formally submit but everybody needs answered — and that lowers the friction of asking for help, which is worth more than it sounds.
And they carry context. They know that the 30th is month-end close and the accounting server does not get rebooted that day. They know which supervisor needs a call before their department is touched. They know what was tried two years ago and why it did not work. A good MSP builds much of that through documentation, a consistent lead technician, and regular on-site time, but it takes deliberate effort and it never becomes automatic in the way presence is.
This is precisely why the hybrid model exists, and why we recommend it more often than we recommend replacing an internal person with us.
Side by Side
| In-House Hire | Outsourced MSP | |
|---|---|---|
| Annual cost | $85,000–$115,000 fully loaded for one mid-level hire | A flat monthly fee, typically a fraction of one salary |
| Coverage | Business hours, minus vacation, sick days, and training | Monitoring around the clock; contracted response windows |
| Breadth of skill | Strong in two or three areas, learning the rest on your time | A bench with different specialties to escalate across |
| On-site presence | In the building, every day, noticing things before tickets exist | Scheduled visits plus dispatch — real, but not constant |
| Business context | Deep. Knows your processes, your people, your busy season | Built over time through documentation and account continuity |
| Single point of failure | Yes — illness, vacation, or resignation exposes everything | No — the team and the documentation outlast any one technician |
| Tooling | Single-seat licensing, expensive per unit and often skipped | Enterprise stack bought at volume, included in the fee |
| Scaling up or down | Hiring cycles measured in months; layoffs are painful | Adjust seats on the agreement as headcount moves |
| Loyalty and priority | You are the only client, all day, every day | You are one of several, managed through a queue and an SLA |
Notice that neither column wins outright. The last three rows are exactly where the co-managed model splits the difference.
The Answer Most Businesses Land On
Your internal hire keeps the context and the relationships. The provider supplies the tooling, the coverage, and the specialist depth one salary cannot buy. The arrangement only fails when the split is left vague, so it gets written down before anything starts.
There is a version of this conversation where the internal person hears “we hired an MSP” as “we are replacing you,” and it poisons the arrangement from day one. Handle that directly. In a well-run co-managed relationship the internal person is the one who benefits most: they stop being on call every weekend, they get somebody to call at 11pm, and they get access to a tooling stack their budget would never have approved.
A Straightforward Framework
Headcount is a proxy, not a rule. Complexity moves you up a band: servers, compliance obligations, multiple locations, or specialized software all push toward more internal capability sooner.
A full-time hire is unjustifiable at this size and would spend most of the week without meaningful work. A managed agreement or even hourly support covers everything you need. Put the salary into the business instead.
The pressure to hire starts here, usually because somebody on staff has quietly become "the computer person" and is drowning. Give that person their real job back and outsource the IT. You are still well below the workload that justifies a salary.
A single hire becomes defensible, and immediately becomes a single point of failure. The strongest arrangement is one internal person owning user support and business context, with an MSP owning infrastructure, security, tooling, and after-hours coverage.
Two or three internal staff can cover the day-to-day, but security, compliance, cloud architecture, and project work still benefit from outside depth. Co-managed remains the norm at this size rather than the exception.
At this scale a real internal department makes sense. Outsourcing shifts from general support toward specific functions — security operations, compliance audit support, and large project delivery.
FAQ
The salary is roughly two thirds of it. A competent IT generalist in western Kentucky lands somewhere around $55,000 to $80,000 depending on experience and whether they can handle servers and security or only desktops. On top of that come employer payroll taxes at about 7.65%, health and retirement benefits that commonly add 15% to 25%, recruiting costs, and the tools that person needs to do the job — remote support software, monitoring, backup, ticketing, documentation, and training or certifications. Fully loaded, a single mid-level hire usually lands between $85,000 and $115,000 a year before you have bought a single piece of hardware.
The rough industry rule is one IT person per 50 to 75 employees for general support, and that holds up reasonably well in practice. Below about 25 employees, a full-time hire is almost always underused for their salary and overexposed on the skills you actually need. Between 25 and 75, a single hire becomes defensible but creates a single point of failure. Above 75, an internal team plus outside specialists is usually the strongest arrangement. The employee count matters less than complexity, though — a 20-person engineering firm with three servers and CAD workstations may genuinely need someone on site, while a 60-person operation running entirely in the cloud may not.
Co-managed IT means you keep your internal person or team and bring in an MSP alongside them rather than instead of them. The provider supplies the tooling stack — monitoring, patching, backup, security, documentation — plus after-hours coverage, vacation and sick backup, and escalation to specialists your one hire cannot reasonably be. Your person keeps the business context, the relationships, and the day-to-day work they are best positioned to do. It is the right answer far more often than either side of the debate admits, particularly for businesses between 25 and 100 employees that already made a hire and are feeling the coverage gaps.
Concentration. One person cannot cover 8,760 hours a year, cannot be expert in networking and security and cloud and your line-of-business application simultaneously, and cannot be in two places when two things break. They also accumulate knowledge that lives nowhere else, so a resignation, an illness, or a two-week vacation exposes the whole organization at once. Most businesses discover this the first time that person leaves and nobody knows the firewall password. Documentation discipline and a standing outside relationship are the two mitigations, and the second one is much easier to actually maintain.
Presence and context, and both are real. Someone who sits in your building learns which machine the shipping manager uses, notices the printer problem before a ticket exists, and understands why the month-end process cannot be interrupted on the 30th. They are available for the two-minute question that nobody would open a ticket for, and they build relationships with staff that make people ask for help earlier. An MSP works through a queue and a schedule. A good one closes much of that gap with dedicated technicians, regular on-site visits, and real documentation, but it would be dishonest to claim it disappears entirely.
Yes, and that is a large share of the work we do. The usual arrangement is that we own the infrastructure layer — monitoring, patching, backup, security tooling, and after-hours alerting — while your person owns user support, the line-of-business applications, and the internal relationships. They gain a team to escalate to, real vacation coverage, and a tooling stack no single-person budget could justify. The arrangement fails only when it is set up ambiguously, so the scope split gets written down before anything starts: who owns what, who is called first, and who has final say on changes.
If outsourcing is on the table, the next two questions are what it costs and which model you are buying. See IT support pricing for the models and market ranges, and managed IT vs. break-fix for why a monthly agreement behaves differently from hourly support. Already have a provider you are unhappy with? Read switching IT providers.
A free assessment gives you an honest read on your actual workload, risk, and complexity — including whether you would be better served by hiring. We have told businesses to hire before, and we will again.
Murray, KY 42071 • Serving Calloway County and western Kentucky
Ready to Transform Your Technology?
Let’s discuss how we can help streamline your IT infrastructure, improve security, and accelerate your business growth with our comprehensive technology solutions.
Ready to discuss your technology needs? Our team of experts is here to help you find the perfect solution for your business. Contact us today for a free consultation.
Phone
(270) 807-1000Service Area
Murray, KY & Western Kentucky — Remote Support Available Nationwide
Support Hours
24/7 Emergency Support
For urgent technical issues or emergency support, our team is available 24/7 to ensure your systems stay operational.