Ask a small-business owner what an hour of IT downtime costs and most will shrug. That shrug is the problem. Downtime feels like an occasional nuisance, a slow morning while something gets fixed, rather than a number on a spreadsheet. But it is a number, and once you work it out honestly it tends to change how you think about backups, hardware quality and support. This piece walks through how to calculate your own figure and why the true cost is almost always larger, and arrives sooner, than people expect.
Downtime is a number, not a feeling
The reason downtime is underestimated is that its cost is diffuse. When systems are down, nobody sends you an invoice; the loss is spread across idle staff, stalled sales, missed deadlines and frustrated customers, none of which shows up as a single line. Because it never arrives as a bill, it never gets the attention a bill would. The first step to taking it seriously is to convert that vague dread into a concrete pounds-per-hour figure for your own business.
There is a simple, defensible way to estimate the direct part. Take your number of affected staff, multiply by their average fully-loaded hourly cost, and add the gross profit you would normally earn in that hour. That gives you a conservative cost of one hour with the systems down. For most small businesses the result is sobering, and it is deliberately conservative, because it ignores everything harder to measure that we come to next.
- •Direct cost per hour = affected staff x loaded hourly cost + lost gross profit for that hour
- •It feels minor because no single invoice ever arrives - the loss is spread and invisible
- •Even the conservative direct figure surprises most owners the first time they calculate it
The hidden costs that dwarf the obvious ones
The wages-and-sales figure is only the visible tip. Underneath sit costs that are larger and longer-lasting. There is the recovery itself: emergency support, overtime and the scramble to restore systems, often at premium rates because it is urgent. There is the backlog: work does not vanish during an outage, it piles up and has to be cleared afterwards, so productivity stays depressed for days beyond the outage itself.
Then there is the damage that does not appear on any ledger. A customer who cannot reach you, place an order or get served may quietly go elsewhere and not come back. Reputation, once dented, is slow and expensive to rebuild. And depending on the cause, an outage can carry regulatory weight: if downtime stems from a security incident involving personal data, you may face reporting obligations and the costs that follow. These soft costs routinely exceed the hard ones.
What actually causes it
Downtime has a small number of usual suspects, and most are preventable. Hardware failure is the classic: a single ageing server or a cheap laptop with no redundancy fails, and whatever depended on it stops. Human error and failed updates account for a surprising share. And cyber incidents, particularly ransomware, have become a leading cause, capable of taking an entire business offline at once rather than one component at a time.
The common thread is that the cheap-looking choice often causes the expensive outage. A bargain machine with no warranty cover, a server kept years past its sensible life, an untested backup that turns out not to restore, these are the false economies that produce the worst days. We pull this thread in the hidden cost of cheap business laptops, and the logic applies equally to servers and infrastructure.
Spending a little to avoid losing a lot
Once you have your pounds-per-hour figure, prevention stops looking like a cost and starts looking like insurance with a calculable return. A tested backup and disaster-recovery plan, hardware with proper warranty and redundancy, and responsive support are not luxuries; they are cheaper than the downtime they prevent, often by a wide margin. The question is no longer whether to invest but how much, and your downtime figure answers it.
Practical priorities: make sure backups exist and are actually tested by restoring from them, not just assumed to work; keep critical hardware within support and replace it before it becomes the weak link; and have a clear, fast route to help when something breaks. A managed support relationship turns a crisis into a phone call, which is one of the strongest arguments in our outsource versus in-house piece. Start with backup and disaster recovery and keep ageing kit covered with hardware maintenance.