Cost of downtime: What an outage really costs you
Sponsored content. This article was supplied by a third party and Business Money has been paid to publish it.
Credit: Pexels
How much is downtime really costing your business?
Most businesses can tell you what an hour of downtime feels like. Far fewer can tell you what it costs. That gap matters, because the cost of downtime is a financial exposure like any other, and it behaves very differently from the rough figure most owners carry in their heads.
Downtime is also more common than most think. The UK Business Data Survey 2026 found that 20% of businesses storing or processing data away from their premises had been hit by a server or cloud outage in the previous 12 months. Databarracks’ Data Health Check 2026, based on 500 UK IT professionals, put cyber incidents as the single biggest cause of IT downtime for the fourth consecutive year.
The instinct is to think about lost sales: revenue divided by trading hours, multiplied by hours lost. It is a start, but it captures only the money that walked out of the door. It ignores the money still walking out while nothing is being produced, sold or shipped.
Key takeaways
- Lost revenue is only part of the picture: idle wages, recovery overtime, scrapped stock, expedited freight, and contractual penalties all add to the real cost of downtime.
- A more accurate production downtime formula adds lost gross margin per hour, idle labour costs, recovery and remediation, and any penalties or written-off stock, then divides by hours lost.
- The per-hour cost almost always rises with the length of an outage, because short interruptions get absorbed while long ones break delivery promises.
- Cyber incidents were the single biggest cause of IT downtime for the fourth consecutive year, according to Databarracks’ Data Health Check 2026, and one in four organisations experienced a cyber incident originating with a supplier or third party.
- Power failure takes out machinery, refrigeration, servers, payments, and access control simultaneously, yet it is typically the cause planned for last.
- For a food business with an estimated downtime cost of £1,500 an hour, a 10-hour outage alone represents £15,000 of disruption before any stock write-off or customer impact is counted.
The costs that never reach the sales report
Payroll keeps running when the line stops. If 30 people are stood down for four hours at a fully loaded cost of £22 an hour, that is £2,640 of wages bought with no output, before a single lost order is counted. Salaried staff cost the same whether the systems are up or not.
Then come the recovery costs, which are usually larger and always less visible:
- Overtime or weekend shifts to clear the backlog, often at premium rates
- Expedited freight to protect a delivery date that has already slipped
- Scrapped work in progress, spoiled stock and chilled or frozen product written off
- Equipment damage from an abrupt stop, plus unplanned engineer callouts
- Contractual penalties, service credits or rebates where a service level was missed
- IT time spent on restoration, data reconciliation and checking what was actually lost
- A customer service queue that takes days to clear and pulls staff off other work
Further down the chain there are effects no spreadsheet catches neatly. A missed slot with a major retailer can cost more in future volume than the day itself cost in output. A supplier who reschedules you to the back of the queue adds weeks. Databarracks also found that one in four organisations had experienced a cyber incident originating with a supplier or third party, which is a reminder that downtime risk is rarely contained inside your own four walls.
Why nobody can quote you a standard hourly figure
Published averages for the cost of unplanned downtime in manufacturing tend to be enormous, because they are dominated by automotive plants and semiconductor fabs where an idle hour runs into six figures. Those numbers are close to useless for a 40-person engineering firm or a regional food producer.
A more honest production downtime formula works from your own figures: lost gross margin per hour, plus idle labour cost per hour, plus recovery and remediation, plus any penalties or written-off stock, divided by the hours lost. Run it for a one-hour stoppage and again for an eight-hour one. The per-hour figure almost always climbs with duration, because short interruptions get absorbed and long ones break delivery promises.
Do it once and you have a number you can actually use when weighing up resilience spending, insurance excesses or whether a maintenance window is worth the disruption.
The causes that get planned for last
Power failure is the obvious one and the one most often left until last. A loss of mains supply takes out everything at once: machinery, refrigeration, servers, card payments, access control, lighting and heating. The causes are mundane, which is part of the problem. A failed substation, a storm, a contractor’s digger through a cable, a fault in an ageing distribution board, or planned works by the network operator.
The relevant question is not how likely it is, but how long it lasts. A UPS will hold servers up long enough to shut down cleanly, which protects data but does not keep you trading. Standby generators for hire load banks and distribution kit and can carry critical circuits (cold storage, key production lines, comms rooms, alarms) while a permanent fix is arranged. Generator hire for events, projects and installations will typically cover business continuity needs from the same diesel and hybrid fleets.
Cyber incidents and IT failure sit at the other end of the spectrum but need the same discipline. As the Databarracks figures show, they remain the leading cause of IT downtime, so the planning points are an incident response plan people have actually read, backups that have been tested by restoring from them, and an agreed recovery time objective signed off before anything happens rather than argued over during the outage.
Third-party and supply chain failure is the cause most often left off the risk register entirely, yet one in four organisations in that same survey experienced a cyber incident originating with a supplier or third party. Your continuity plan is only as good as the weakest link in it. Map which suppliers are genuinely single-source, identify alternatives before you need them, and check what their own recovery commitments are.
Finally, there is the key person risk. Losing access to the one individual who understands a critical system, machine or process is its own form of operational downtime, and it arrives without warning through illness, resignation or a badly timed holiday. Documented procedures and cross-training are unglamorous, but they are cheap insurance against a stoppage nobody can fix.
The cost of inaction, in round numbers
Take a food business with £8,000 of stock in chilled storage and a downtime cost of roughly £1,500 an hour once wages, lost margin and recovery are included. A 10-hour outage is £15,000 of disruption plus a potential stock write-off, and that is before any customer is told their order is not coming.
The businesses that handle outages well are rarely the ones with the biggest budgets. They are the ones who have already worked out their hourly number, identified which three systems must stay live, and made the calls before they needed to.