Most of the “cost of downtime” figures you will see online are built for large enterprises and tell a small business very little. The number that matters is your own, and it is easier to work out than you might think. This guide shows how, explains what actually causes outages according to the latest UK research, and sets out what being ready to recover looks like in practice.
Why one headline figure does not help you
Search for the cost of downtime and you will find eye-catching numbers per hour. Most come from surveys of large organisations, often in US dollars, and they say little about a business with ten or fifty people. We have deliberately not quoted one here. A figure that is wrong for your business is worse than none, because it either scares you or lets you off the hook.
What does hold up is the method. Downtime costs a business in three ways, and you can put a rough figure on each.
The three costs of an outage
1. Staff time you still pay for. When systems are down, people are still on the payroll but cannot do their work. Count the number of people affected, their average hourly cost to the business (salary plus employer costs, not just take-home pay), and the hours lost.
2. Revenue you did not earn, or earned late. If you take orders, bookings, calls or payments through your systems, work out what a normal hour is worth and how much of it simply cannot happen when you are offline. Some of that is lost for good; some is only delayed. Be honest about which is which.
3. The cost of getting back up, and the costs you cannot see. Recovery has direct costs, such as engineer time, replacement equipment and overtime to catch up. Then there are the harder ones: a missed deadline, a client who loses a little confidence, a day of lost focus afterwards. These are real but hard to price, so note them even if you do not put a number on them.
- 1Staff time
People affected × hourly cost × hours lost
- 2Lost or delayed revenue
What a normal hour is worth, minus what can still happen offline
- 3Recovery and hidden costs
Engineer time, equipment and overtime, plus effects that are hard to price
A worked example (illustrative only)
The figures below are invented to show the method. They are not i90 data and not a benchmark.
Say twelve people are affected by an outage that lasts three hours, and each costs the business £25 an hour. That is 12 × £25 × 3 = £900 in idle staff time. If a normal hour brings in £400 of work that cannot happen offline, and half of it is lost for good, that is a further 3 × £400 × 50% = £600. Add £300 of engineer time and catching-up overtime, and this made-up outage costs £1,800, before counting the harder-to-price effects.
Now put your own numbers in. Even a rough answer is far more useful than a statistic from a business that looks nothing like yours.
What actually causes downtime
The most useful current UK source we have found is the Data Health Check 2026 from Databarracks, an annual survey of 500 IT professionals. It asked what caused their organisation’s IT downtime. The leading causes were:
- Cyber incidents at 29.8%, the leading cause for the fourth year running (it was 13% five years ago)
- Hardware failure at 18.6%
- Connectivity problems at 14.0%
- Upgrades and patches at 12.0%
- Cloud outages at 11.4%
Only a small share (3.2%) pointed to extreme weather or flooding, and 11% reported no downtime at all in the last year. The same research also lists theft among the causes of data loss, at 8.6%, which is a reminder that physical events count too.
Two things stand out for a small business. First, most of these causes are ordinary: a failed disk, a dropped connection, an update that did not go to plan. Second, the biggest single cause is now cyber attacks, and small businesses are not exempt. The government’s Cyber Security Breaches Survey 2025/26 found that 42% of micro businesses (1 to 9 employees) and 46% of small businesses (10 to 49 employees) had experienced a breach or attack in the previous 12 months.
The gap most businesses do not see: recovery
Knowing the cost of an outage is only half of it. The other half is how quickly you can actually get back.
The same Databarracks research found that 65% of organisations think a serious cyber attack could threaten their survival. Yet only 35% had tested recovery from a cyber attack in the previous year, and 27% fell short of their own recovery time target in their latest test or real incident. Confidence, in other words, often runs ahead of capability.
The UK picture for small firms is similar. In the Cyber Security Breaches Survey 2025/26, 29% of micro businesses and 44% of small businesses had a business continuity plan covering cyber security. For small businesses, that was down from 53% the year before.
The practical lesson is simple: a backup you have never tested restoring is a hope, not a plan. We cover one common blind spot in does Microsoft back up your 365 data.
How we approach it at i90
We cannot stop every outage, and nobody honestly can. What we can do is make them rarer, shorter and less painful. At i90 that means:
- Proactive monitoring, so problems are often spotted before your team has to report them
- Tested backup restores, so we know a backup works because we have seen it restore, not just seen it run
- Patch management, so updates are planned rather than arriving as surprise outages
- Documented recovery plans, so if something does go wrong, there is a written route back rather than improvisation
When something does go wrong, speed matters. 84% of our tickets are resolved within the hour. If you would like to see what is typically covered, read what is included in managed IT support.
When the cause is not a fault: a break-in
Not all downtime starts with a failed disk or a bad update. One of our clients experienced a break-in at their premises. Several machines, including servers, were stolen, or damaged as the thieves tried to remove them. The equipment had been secured with Kensington locks, which is a sensible precaution, but it did not stop the loss.
The business was affected at once. Staff had no machines to work on, so they could not use email, databases or do their day-to-day jobs.
We were asked to attend the site out of hours, assess the situation and come up with a recovery plan. Replacing the equipment needed authorisation from the insurers first, so in the meantime we provided a loan server and a couple of workstations to key personnel, to keep the most important work moving while approval to buy replacements came through. That was only part of the job. We were also asked to write a comprehensive report for the police and the ICO covering the data involved and what our exposure was, and to provide an asset list with descriptions for the insurers.
Four things stand out from it, and they apply to any business:
- Replacement can wait on someone else. When insurance is involved, buying new equipment needs approval first. Having a way to keep key people working in the gap, such as loan kit, is what stops a bad day becoming a bad week.
- An outage can be physical. Locks and alarms reduce the risk but do not remove it, so plan for the day equipment is simply not there.
- You need to know what you had, and what was on it. The police, the ICO and the insurers each needed something different: an account of the data involved and the exposure, and an asset list with descriptions. Having an up-to-date asset list and a clear picture of which data lives on which device ready beforehand is far easier than reconstructing it afterwards.
- The cost goes well beyond the hardware. Idle staff, out-of-hours attendance, recovery planning, loan equipment and reporting all landed at once, which is exactly why the three-cost method above is worth doing.
We have kept the details of this case general and the client unnamed.
What to do this week
You do not need a big project to improve your position. Start with these:
- Work out your own hourly cost of downtime using the three costs above.
- Ask your IT provider when your backups were last test-restored, and what was restored.
- Check that you have a written plan for who does what if systems go down, including how to contact your provider out of hours.
- Keep an up-to-date list of your equipment and know which data lives on which device.
- Make sure the basics are in place: multi-factor authentication (what is MFA) and staff who can spot phishing (how to spot a phishing email).
If your current setup cannot answer these questions, that is worth knowing. See the signs you have outgrown your IT support, and if you decide to change, what happens when you switch IT support providers explains how to do it safely.
Common questions
How much does IT downtime cost a small business?
There is no reliable single figure, and most published ones are for large organisations. Work out your own: staff time lost, revenue lost or delayed, and the cost of recovery, then add the effects you cannot easily price.
What is the most common cause of IT downtime?
In the Databarracks Data Health Check 2026, cyber incidents were the leading cause of IT downtime at 29.8%, followed by hardware failure (18.6%) and connectivity problems (14.0%).
How likely is a small business to be attacked?
In the government’s Cyber Security Breaches Survey 2025/26, 42% of micro businesses and 46% of small businesses reported a breach or attack in the previous 12 months.
What is a recovery time objective?
It is the length of time you can afford to be without a system before the business is seriously affected. It is only useful if you test it, because plans often take longer in practice than on paper.
How often should backups be tested?
There is no single rule for every business, but they should be tested regularly enough that you trust the result. Ask your IT provider when yours were last restored successfully, rather than when they last ran.
Sources
- Databarracks, Data Health Check 2026 (survey of 500 IT professionals)
- Department for Science, Innovation and Technology, Cyber Security Breaches Survey 2025/26 (fieldwork August to December 2025)
Want to know where you stand?
Get in touch and we will talk through your setup and the questions above in plain English.
Get in touch