Break-fix is straightforward: something breaks, you call, you are billed for the time. No monthly commitment. For a small business watching every line item, that is a genuinely appealing proposition.
It is also a model with an incentive structure that works directly against you, and that is the part worth understanding before comparing prices.
The incentive problem
Under break-fix, the provider is paid when things go wrong. Under managed service, the provider is paid a fixed fee regardless — which means every hour they spend fixing something is an hour of margin lost.
This is not a claim about anyone's ethics. It is simply that the two models reward opposite outcomes. Break-fix revenue rises with failures. Managed service revenue is protected by preventing them. Given a stable environment, a break-fix provider earns less, and a managed provider earns more.
Over a few years, that difference compounds into two very different environments.
What break-fix structurally cannot include
The billing model makes certain work impossible to deliver, because nobody is paying for it:
- Proactive monitoring. Nobody bills for watching a system that has not broken.
- Patch management. Routine, invisible, and produces no billable event.
- Backup verification. The test restore that proves your backups work is unbillable maintenance.
- Documentation. Every engagement starts by rediscovering your environment, at your expense.
- Lifecycle planning. Nobody is tracking which machines are about to fail.
- Security posture. Configuration review, access audits, MFA enforcement — all continuous, none of it triggered by a fault.
The consequence is that a break-fix environment accumulates unaddressed risk continuously. Each individual incident gets resolved competently; the conditions producing incidents never get addressed.
The arithmetic that actually matters
Comparing break-fix hourly rates against a monthly managed fee is the wrong comparison, because it counts only the invoice and ignores the outage.
The real cost of an incident is the provider's time plus the productivity lost while it is unresolved. A four-hour outage affecting twenty staff is eighty person-hours of disruption — which will dwarf the repair invoice in virtually every case.
Break-fix also has an inherent response delay. You are not on a retainer, so you enter a queue behind clients who are. That queue position is precisely the difference between a two-hour outage and a two-day one.
The security argument has become decisive
Break-fix was a defensible choice when the main risk was hardware failure. It is much harder to defend now, for a reason that has nothing to do with technology preference: insurance.
Cyber insurance applications now ask direct questions about MFA coverage, endpoint detection and response, patch cadence, and backup testing. These are continuous controls. A break-fix arrangement does not produce them, which means a business on break-fix is frequently answering those questions inaccurately — and an inaccurate answer can void the policy at claim time.
That risk is not theoretical, and it is not priced into the hourly rate.
When break-fix is still the right answer
It remains legitimate in specific situations:
- Very small businesses. Under roughly five people with simple cloud-based needs, a managed agreement can genuinely be more than required.
- Genuinely simple environments. No servers, no compliance obligations, everything in one cloud platform, low tolerance for complexity.
- Supplementary specialist work. Even businesses with managed IT sometimes engage specialists for defined projects.
- Businesses where downtime is genuinely cheap. Rare, but they exist.
The test is not company size but consequence. If a day without systems is an inconvenience, break-fix may be rational. If it stops revenue, it is not.
The transition point
Most businesses cross the line somewhere between 10 and 20 employees, or at the first of these events: the first compliance obligation, the first cyber insurance renewal with conditions attached, the first outage that costs a full day, or the first client security questionnaire.
The signal to watch is not the total spend. It is the pattern of it. Reactive spending is lumpy and unpredictable — a quiet quarter followed by an expensive month. Flat-rate is not cheaper in every individual month; it is predictable, and predictability is most of what a business is actually buying.