The true cost of downtime (and how to put a number on it)
Downtime costs more than lost sales during the outage. Here's a simple way to calculate what an hour down really costs you — and why it usually justifies monitoring many times over.
“We can’t really afford monitoring right now” is a sentence that only makes sense until you calculate what a single outage costs. Downtime is rarely just the revenue you miss while the site is down — and once you put a number on the full cost, the math for prevention gets very easy.
The direct cost is only the start
The obvious line is lost revenue: if you make $10,000 a day, an hour of full downtime during business hours is a few hundred dollars, more at peak. Simple enough. But that’s the smallest part of the bill.
Add the costs that don’t show up on the dashboard
- Recovery labor — engineers pulled off roadmap work to firefight, plus the context-switching tax on everyone around them.
- Support load — the flood of tickets and messages, and the team-hours to answer them.
- Churn and trust — customers who hit an error at the wrong moment and don’t come back, or who quietly lower their opinion of your reliability.
- Reputation — a public outage that lands on social media or a status-page subscriber list costs credibility you spent years building.
- SLA penalties — if you’ve promised uptime contractually, a breach can mean service credits or refunds.
Most of these dwarf the direct revenue loss.
A back-of-envelope formula
You don’t need a spreadsheet with twenty tabs. Start here:
Cost per hour of downtime ≈ (hourly revenue) + (people involved × hourly cost × hours spent) + (a churn/reputation factor).
Plug in your own numbers. Even a modest SaaS often lands in the hundreds to low thousands per hour once labor and churn are included — and a single bad outage can exceed a year of monitoring spend.
Why detection speed is the multiplier
The cost scales with duration, and duration starts at detection. If you find out about an outage 30 minutes late because you were checking every 5 minutes from one location — or worse, from a customer email — you’ve added 30 minutes of full cost before anyone even started fixing it. Faster, multi-region detection with instant alerts is the cheapest lever you have on the total.
The bottom line
Downtime costs far more than the revenue missed during the outage — recovery, support, churn, reputation, and penalties usually cost more. Put a rough number on your cost-per-hour, multiply by how long detection alone adds, and monitoring stops looking like an expense and starts looking like insurance that pays for itself the first time something breaks.
Reliability that pays for itself. See pricing →
