Every proposal for monitoring, redundancy or backup eventually meets the same question: is this worth it? The honest answer requires knowing what an hour of downtime actually costs your business, and most businesses have never calculated it. They either wave at a large intuitive number or dismiss the risk entirely.
The calculation is not difficult and it makes every subsequent decision easier.
The four components
Lost revenue. For businesses that transact online, this is the most direct figure: annual online revenue divided by trading hours gives an average hourly rate. Weight it — an hour on a Tuesday morning is not an hour on a Friday afternoon, and if you have a seasonal peak, model the outage occurring during it.
For businesses that do not sell online, this is enquiries rather than sales. Take your monthly enquiry volume, your conversion rate and your average job value, and calculate what an hour of unreachable phones or a dead contact form is worth in pipeline.
Idle staff cost. Take the fully loaded hourly cost of staff who cannot work during the outage. If twelve people at an average of $60 an hour cannot do their jobs, that is $720 an hour going out regardless of what comes in. This is the component people most often forget, and in service businesses it usually exceeds the lost revenue figure.
Recovery cost. The direct expense of fixing it — emergency IT support at after-hours rates, any hardware, and the overtime spent catching up afterwards. Backlogs are real: a four-hour outage frequently costs eight hours of recovery work.
Reputation and churn. The hardest to quantify and often the largest. Rather than guessing, take an existing figure you trust — your customer lifetime value — and estimate how many customers an outage of a given length would plausibly cost you. Even a conservative estimate makes the point.
Work through an example
A twenty-person professional services firm. Fifteen staff cannot work, at an average loaded cost of $65 an hour: $975. Enquiries worth roughly $400 an hour in expected pipeline value. Emergency support at $250 an hour. That is approximately $1,625 per hour before touching reputation.
A single full-day outage, which is entirely realistic for a hardware failure without a tested restore, is therefore around $13,000. Against that figure, a failover internet connection and a monitored backup arrangement stop being a discretionary spend.
It is worth modelling the longer scenario too. The multi-day outages that genuinely damage businesses are now more often caused by ransomware than by hardware, and they run for days rather than hours. If you have not assessed where you sit on that risk, a cyber security self-assessment will give you a clearer input to the same calculation.
Use it to set targets, not just to justify spending
The more valuable use of this number is setting your recovery objectives deliberately. If an hour costs $1,600, you can decide rationally how much recovery time you are willing to tolerate and how much you will pay to reduce it. Recovery time and recovery point objectives become business decisions rather than technical ones.
It also tells you where to spend. If most of your cost is idle staff, the priority is restoring access quickly — which favours failover and rapid recovery. If most of it is lost online revenue, the priority is keeping the customer-facing path up, which favours redundancy at the front end.
Detection is part of the duration
One detail changes the arithmetic more than anything else. Total downtime is detection time plus response time plus repair time. Businesses focus almost entirely on the third and ignore the first, yet outages discovered by a customer phone call routinely run for hours before anyone starts working on them.
Reducing detection from two hours to two minutes removes nearly two hours from every outage you have, at a cost that is trivial next to the hourly figure you just calculated. It is usually the highest-return item on the list, and the one that gets bought last.