Managed IT vs. break-fix: what's actually right for your business

Break-fix looks cheaper until you price the hour your team spends waiting. Here's the math, the tipping point, and the three questions to ask any provider.

Manuel RuizFounder, CEO & CISO
September 24, 2026 · 7 min read
New York street with a yellow cab in motion

Break-fix is the IT model most small companies start with because it looks like the model with no commitment: something breaks, you call someone, you pay for the hours. Managed IT is a flat monthly fee for a provider to own the environment and keep it from breaking in the first place. The decision looks like a pricing question. It isn’t. It’s a question about who carries the risk of downtime.

The hidden invoice in break-fix

When a server goes down under break-fix, the provider’s clock starts when they arrive. Your clock started when the first employee couldn’t log in. Gartner puts mid-market downtime at roughly $5,600 per minute; even if your number is a tenth of that, a four-hour outage costs more than a year of managed services for a 20-person team. Break-fix providers are also structurally incentivized to let small problems become billable ones — nobody gets paid for the patch that prevented the outage.

What managed IT changes

  • Monitoring runs continuously, so the provider usually knows before you do.
  • Patching, backups, and security baselines happen on a schedule, not when someone remembers.
  • Response time is a contractual number (ours is 2 minutes) rather than “when a tech is free.”
  • Costs are predictable: one line item, no surprise invoices after a bad month.

The tipping point

In our experience the crossover is around 10–15 employees, or earlier if you handle regulated data. Below that, break-fix with a good local tech can work. Above it, the arithmetic flips: the cost of one engineer-hour of your own staff waiting exceeds the per-user managed fee, and compliance obligations (HIPAA, SOC 2 questionnaires from customers, cyber-insurance applications) start demanding evidence that only continuous management produces.

Three questions to ask any provider

  • What is your measured average response time, and will you put it in the contract with credits?
  • What percentage of tickets do you resolve without a human, and how do you prove it?
  • If I leave, how fast can I get my documentation and admin credentials back?

If the answers are vague, you’re buying break-fix with a subscription attached. If they’re specific, you’re buying an IT department.

Want this applied to your environment?

Thirty minutes with an engineer and it's a plan, not a blog post.