The Dispatch

A Practical Framework for IT Reliability: Getting the Operate Stage Right

Written by John Ohlwiler | 11/16/26, 12:45 PM

If your company runs on EOS, you know the moment. It’s the Issues section of the weekly Level 10 meeting, and someone adds “warehouse Wi-Fi” to the list. Again. The team spends ten minutes on it, someone promises to call IT, and a few weeks later it’s back, sitting next to “shared drive is slow” and “new hire’s laptop wasn’t ready.”

None of those issues is a crisis on its own. Together, they tell you something worth hearing: your technology isn’t failing because of bad luck. It’s failing because nobody has built a system for keeping it reliable.

What is an IT reliability framework? An IT reliability framework is a repeatable set of practices that keeps technology available and predictable: knowing what you own, monitoring it, keeping it updated, proving you can recover, and documenting how work gets done. It moves IT from reacting to breakdowns to preventing them.

At Sentry, we call that foundation the Operate stage of the Technology Maturity Model. Here’s how to get it right.

Why Does IT Keep Breaking in Growing Companies?

Most second-stage businesses didn’t design their technology. They accumulated it. A server bought during a growth spurt, a firewall a former employee configured, a backup someone set up years ago and nobody has looked at since. What worked at 20 employees quietly strains at 80, and the cracks show up as recurring tickets.

The cost of those cracks is real. A 2025 study from ITIC and Calyptix found that many small and midsize businesses lose $25,000 or more for every hour of downtime.1 Larger organizations face even steeper numbers: in Uptime Institute’s 2025 survey, 57% of respondents said their most recent major outage cost more than $100,000.2 (If you’ve never put a number on your own exposure, our guide to calculating the real cost of downtime walks through it.)

Here’s the encouraging part. Uptime Institute found that 85% of outages caused by human error trace back to staff not following procedures or to flaws in the procedures themselves.3 In other words, a large share of downtime is a process problem. And process is something you can fix.

That’s the frustrating truth for many leaders: you’ve built disciplined processes for sales, finance, and operations, yet technology still runs on memory and goodwill. It shouldn’t have to.

What Is the Operate Stage of the Technology Maturity Model?

Sentry’s Technology Maturity Model (TMM) moves businesses through four stages: Operate, Secure, Integrate, and Innovate. Operate comes first because everything else depends on it. Security controls are only as strong as the systems they protect. Integrations break when the underlying platforms are unstable. And AI projects, as we covered in how AI workflows are changing businesses, amplify whatever is already in place, good or bad.

Getting Operate right doesn’t mean buying new tools. It means putting five practices in place and keeping them running.

How Do You Make IT More Reliable? Five Operate-Stage Practices

1. Know exactly what you have

You can’t keep something reliable if you don’t know it exists. Start with a living inventory of every device, application, license, and vendor, including warranty and end-of-support dates. Then build a lifecycle plan that replaces equipment on a schedule instead of after it fails. The payoff is a technology budget that reads like a forecast, not a series of surprises.

2. Find problems before your people do

If your team is the alert system, you’re already behind. Remote monitoring and management (RMM) tools watch servers, computers, and network equipment around the clock and flag warning signs like failing drives, full storage, or offline backups. The key is routing every alert to someone who owns the response, so warnings turn into fixes rather than noise.

3. Patch on a schedule, not when you remember

Updates close the gaps attackers and bugs exploit, but only when they actually get installed. Verizon’s 2025 Data Breach Investigations Report found that only 54% of vulnerabilities in edge devices like firewalls and VPNs were fully remediated, with a median of 32 days to do it.4 Set a monthly patching cadence, prioritize internet-facing equipment, and track which devices fall behind.

4. Prove you can recover

A backup you’ve never restored is a hope, not a plan. Define two numbers with your leadership team: your recovery time objective (how long a system can be down) and your recovery point objective (how much recent data you can afford to lose). Then test restores at least quarterly against those targets. Florida businesses got a firsthand lesson in this during hurricane season, which we unpacked in what storms teach us about IT resilience.

5. Write it down, then follow it

Documentation is the least glamorous practice and often the most valuable. Network diagrams, admin credentials stored in a secure vault, onboarding and offboarding checklists, vendor contacts, and a simple change process (who approves changes and when they happen) turn tribal knowledge into something the whole team can rely on. Given how many outages trace back to procedures, this is where reliability is often won or lost.

How Do You Measure IT Reliability?

If you run on EOS or Scaling Up, you already believe in managing by numbers. Technology deserves a few seats on your Scorecard. These starting targets are a reasonable baseline for most second-stage companies; adjust them to your business.

Metric What it tells you Starting target
Unplanned downtime How often critical systems are unavailable Trending down quarter over quarter
Time to resolve priority issues How quickly business-stopping problems get fixed Defined and measured against an agreed target
Repeat issues Whether problems get fixed or just patched over No repeats among your top five issues
Patch compliance Share of devices current on updates 95% or more within 30 days
Backup restore tests Whether recovery actually works Successful test every quarter
Aging equipment Devices past warranty or end of support Under 10% of the fleet

Pick two or three to start. The goal isn’t a perfect dashboard. It’s making technology health visible in the same meeting where you review every other part of the business.

What Changes When the Operate Stage Is Solid?

The difference shows up in places you might not expect. The Issues List gets shorter. New hires have working equipment on day one. Your technology budget stops generating surprise invoices. Leadership time that went to firefighting goes back to growing the business.

You also earn the right to move forward. With a stable foundation, the Secure stage becomes far easier, because insurers and auditors ask the same questions your Operate practices already answer. And when the conversation turns to AI and automation, you’re building on systems you trust.

Sentry has guided growing companies through this transition for more than a decade, and the pattern is consistent: the businesses that scale smoothly aren’t the ones with the fanciest tools. They’re the ones that took the fundamentals seriously. If you’re weighing who should own that roadmap, our comparison of a fractional CIO vs. a full-time CIO is a useful next read.

Frequently Asked Questions

What’s the difference between IT reliability and IT security?

Reliability keeps systems available and performing; security protects them from threats. They overlap (patching and backups serve both), but they’re distinct goals. In the Technology Maturity Model, reliability is the Operate stage and security is the Secure stage that builds on it.

How much uptime should a growing business target?

Many growing businesses aim for 99.9% availability on critical systems, which works out to about 8.8 hours of unplanned downtime per year. Businesses with real-time revenue, like e-commerce or healthcare, often need more. The right target depends on what an hour of downtime costs you.

Can an in-house IT person handle the Operate stage?

Often, yes, with support. One person can’t monitor systems around the clock, patch every device, and document everything while also answering daily requests. Many companies pair internal staff with a managed IT partner so each focuses on what they do best.

Is break/fix IT support enough for a growing company?

Break/fix support, where you pay someone only when something breaks, is reactive by design. It can work for very small teams, but it gives no one an incentive to prevent problems. Once technology becomes central to how you deliver for customers, proactive management usually costs less than the downtime it prevents.

How long does it take to get the Operate stage right?

It depends on your starting point, but most of the work is project-shaped: an inventory, a monitoring rollout, a patching cadence, a tested recovery plan, and core documentation. Most companies see the biggest gains early, then refine over the following quarters.

Know Where You Stand

Reliable IT doesn’t start with a purchase. It starts with an honest look at where your business is today and which practices are missing. That’s exactly what a TMM Audit is built to show you, with a clear, prioritized path forward.

Curious where your business falls on the Technology Maturity Model? Request a free TMM Audit. [PLACEHOLDER LINK]

References

  1. ITIC and Calyptix. 2025 SMB Security and Hourly Cost of Downtime Survey, January 2025. https://www.calyptix.com/wp-content/uploads/Calyptix-ITIC-SMB-Security-and-Hourly-Cost-of-Downtime-Survey-Results-1.2025.pdf

  2. Uptime Institute. “Uptime Announces Annual Outage Analysis Report 2026,” May 13, 2026 (citing Uptime’s 2025 Annual Survey). https://www.businesswire.com/news/home/20260513289344/en

  3. Uptime Institute. “Uptime Announces Annual Outage Analysis Report 2025,” May 6, 2025. https://www.businesswire.com/news/home/20250506733553/en/Uptime-Announces-Annual-Outage-Analysis-Report-2025

  4. Verizon. 2025 Data Breach Investigations Report, as reported by Infosecurity Magazine, April 23, 2025. https://www.infosecurity-magazine.com/news/verizon-dbir-jump-vulnerability/

Editor’s notes: (1) Uptime Institute surveys skew toward data center and larger IT operators; the post frames the 57% figure as “larger organizations” for that reason. (2) The 8.8 hours in the FAQ is arithmetic (0.1% of 8,760 hours), not a cited statistic. (3) Scorecard starting targets are Sentry guidance, not published benchmarks; confirm the team is comfortable with them.