When a system goes down, the first thing everyone looks at is how long it took to come back. What almost nobody calculates is how much that hour actually cost — and without that number, it's impossible to seriously justify an investment in redundancy, monitoring, or a continuity plan. You don't need a consulting firm to run this math. You need the right method and your own data.

The three real cost components

The cost of an hour of downtime isn't a single figure — it's the sum of three things that rarely end up on the same spreadsheet:

  • Idle labor: the people who can't work while the system is down, but who still get paid for that hour. If ten people earn the equivalent of $15/hour on average and can't make progress on anything during the outage, you've already lost $150 in wages with nothing produced in return.
  • Lost sales or revenue: if the system that's down is the one that processes orders, takes bookings, or serves customers, every hour of downtime has a direct opportunity cost — it's not idle salary, it's revenue that never came in and probably won't come back.
  • Recovery cost: the outage doesn't end when the system comes back. Lost data needs to be re-entered, half-completed transactions need to be reviewed, and sometimes overtime needs to be paid to catch up on the backlog that built up. This is the component most consistently underestimated, and it often ends up costing more than the outage itself.

The method: how to build the math

The base formula is simple, and it's the same one our own downtime calculator uses:

Cost per hour of downtime = (affected people × average hourly cost) + average hourly revenue that depends on the system that's down.

To get those two numbers, you need data you already have, not market estimates:

  • How many people depend on the system to work, and what's the average hourly cost of those positions (salary + burden, divided by hours worked per month).
  • How much your company bills on average per hour of normal operation — your monthly revenue divided by business hours in the month is a reasonable starting point, though it's worth adjusting for the specific system that's down (not all your revenue depends on a single system).

Multiply that hourly cost by however many hours of downtime you had over the last year — check support tickets, your ISP's logs, or simply the team's memory if you don't keep a formal record — and that's your real annual cost, before adding recovery cost.

A simple example

Consider a hypothetical SMB with 8 people, an average hourly cost of $12 per person, and revenue that depends on the down system of $40/hour. The cost per hour of downtime is (8 × $12) + $40 = $136. If that company had 15 hours of unplanned downtime over the last year — split across internet outages, server crashes, and the occasional power failure — the direct cost was $2,040. Add recovery time on top: if each outage generated an average of one extra hour of work to catch up, that's another 15 hours of labor, in this case $1,440 more. Total: around $3,480 a year, for a small SMB alone. These numbers are illustrative — the real math depends entirely on your own costs and your own outage history.

What to do with this number once you have it

The annual cost of downtime isn't a figure to file away — it's the ceiling for what makes sense to spend on preventing it. A backup link, a properly sized UPS, or monitoring that catches anomalies before they turn into an outage all have a known cost upfront. Comparing that cost against what you're already losing turns the investment decision into math, not intuition.

Why this helps you decide, not just diagnose

Once you have the real annual cost of downtime, the question "is it worth investing in a backup link, a better UPS, or proactive monitoring?" stops being a gut feeling and becomes a comparison of numbers: if the investment costs less than what you lose in downtime per year, it pays for itself — and probably faster than you think, because recovery cost is almost never included in the initial calculation.

This also explains why the same outage costs wildly different amounts at different companies. A professional services firm where nobody can bill during an outage has a very different cost profile than a hotel that keeps taking walk-ins with a paper backup while the booking system is down. The formula doesn't change — but the inputs you plug into it should reflect how your specific operation actually degrades, not a generic assumption borrowed from an industry report.

We built a downtime cost calculator that runs this math with your own numbers in under a minute. And if you want the full picture — how much your manual operation costs beyond downtime, and how much gets recovered with managed infrastructure and automation —, the ROI calculator adds it all up in one estimate.