The Real Cost of Downtime for a 50-Person Business

Most business owners have a rough sense that downtime is expensive. Few have ever sat down and calculated it for their own business. When they do, two things usually happen. The number is bigger than they thought, and the spend on prevention starts to look reasonable rather than expensive.

This is the working through. We use a fifty-person business as the worked example because it is large enough that the numbers are not trivial and small enough that most owners and operators recognise their own situation in it.

The simple direct cost

The simplest version of the maths is wages paid to staff who cannot do their job because the system they depend on is offline.

For a fifty-person business, average all-in employee cost in Australia, including wages, super, leave, and on-costs, sits in the range of forty-five to sixty-five dollars per hour for general professional staff, higher for senior or specialised roles. For the worked example, we will use fifty-five dollars as the blended hourly cost.

If a system outage prevents fifty people from doing meaningful work for one hour, the direct cost is fifty by fifty-five, which is two thousand seven hundred and fifty dollars. For four hours, it is eleven thousand. For a full working day, it is over twenty-two thousand.

That is just the wage cost. It assumes everyone is paid for the time and gets nothing useful done with it. In practice, a partial outage is more common, where some staff can do parts of their job and others cannot. Halve the number if you like. The order of magnitude is still meaningful.

The indirect costs

The wage cost is the visible piece. The indirect costs are usually larger and harder to capture, which is why they get under-counted.

Lost sales and missed transactions. If your e-commerce site, your booking system, or your point of sale is down, customers do not wait. They go elsewhere. For businesses where revenue is tied directly to system uptime, an hour of downtime can cost more in lost sales than in wages.

Recovery overtime and weekend work. When systems come back, the work that did not get done has to be caught up. Quotes that were due go out late. Invoices get processed in arrears. Reports get rebuilt. This often shows up as overtime, weekend work, or simply staff burning out faster than they otherwise would.

Client communication and goodwill. Every outage that affects clients triggers a wave of phone calls, apology emails, and trust repair. Senior staff, who cost more per hour, spend that hour managing communications instead of doing higher-value work. Repeat outages erode the client trust that your sales team spent years building.

Missed deadlines and contractual penalties. If you have client contracts with service-level commitments of your own, your downtime can become their reason to apply a penalty clause or, worse, to start looking at a competitor.

Compliance and reporting obligations. Some industries require you to report material outages to a regulator, an insurer, or a client. Each of those reports takes time to prepare, and some carry indirect consequences in your next renewal.

Reputation in your professional community. "Their systems went down again" is a sentence you do not want potential clients to hear from your existing ones. The cost of that reputation hit is unmeasurable, but it is not zero.

A reasonable rule of thumb, supported by enterprise downtime studies and our own experience working with SMBs, is that indirect costs run between two and five times the direct wage cost. So that one-hour outage that cost two thousand seven hundred and fifty in wages probably actually cost the business eight to fifteen thousand once everything is counted.

Three real scenarios

Abstract numbers are easy to dismiss. Specific scenarios are not.

Scenario one: a four-hour Microsoft 365 email outage during a busy week. Email is down, not just inbound and outbound but also Teams chat for many businesses that rely on it. Direct wage cost for fifty people: eleven thousand. Add quote follow-ups missed, a few client calls handled badly because nobody could check the latest email thread, and overtime to clear the backlog the next morning. Realistic total cost: thirty to fifty thousand.

Scenario two: a ransomware incident with two days of partial outage and three days of partial recovery. Direct wage cost over the worst two days, even at half productivity, sits around twenty-two thousand. Add forensic investigation, legal advice, regulator notifications if personal information was involved, the cost of the cyber insurance excess, the cost of restoring from backup, and the cost of confidence work with clients afterwards. Realistic total cost: between one hundred and fifty thousand and half a million for a fifty-person business, before any actual ransom is paid.

Scenario three: a six-hour internet outage from your primary ISP. With no failover, fifty people are largely offline for the day. Direct wage cost: sixteen and a half thousand. Add missed deliverables and recovery work. Realistic total cost: forty to seventy thousand for a single ISP failure that proper redundancy would have made invisible.

The scenarios are not exotic. We have walked clients through each of them, in real time, in the last twelve months.

What proactive monitoring catches before it becomes downtime

Most outages do not arrive without warning. They arrive after a period of degradation that proper monitoring would have surfaced.

A failing hard drive will report errors for days or weeks before it dies. A misconfigured backup will report failed jobs for nights in a row before the day you need a restore. An expiring certificate will trigger warnings for weeks before the morning your website goes offline. A misbehaving update will be caught on one device before it is rolled out to thirty.

Proactive monitoring catches these signals and acts on them. The whole point of a managed service is that someone is watching the dashboards on the days you are not. The cost of not having that watch is exactly the kind of preventable outage scenarios above.

SLA benchmarks worth asking for

If you are reviewing your IT support arrangements, the response and resolution targets in the contract matter as much as the hourly rate.

Reasonable benchmarks for a small to mid-size Australian SMB:

• Critical incident, system completely down: response within fifteen minutes, work begins immediately, regular updates every thirty minutes.
• High priority, business function impaired: response within one hour, resolution path within four business hours.
• Medium priority, individual user issues: response within four business hours, resolved within one business day.
• Low priority, request and minor issues: response within one business day.

If your current provider's SLA is silent on these, or if the response window for a critical incident is "next business day", you are paying for help that arrives after the damage is done.

What this means in practice

If you have never run these numbers for your own business, the exercise is worth an afternoon. Take your average all-in hourly cost. Multiply by your headcount. Multiply by a realistic outage duration, say four hours. Then multiply by three for indirect costs. That is your cost of one bad day.

Compare that number to the annualised cost of better monitoring, better backups, better redundancy, and a sharper SLA. The maths is rarely close.

If you would like an outside view on your downtime exposure, our free IT health check includes a continuity review. We look at your single points of failure, your monitoring posture, and your current SLA, and we tell you where the next outage is most likely to come from. Our own SLA includes a fifteen-minute response for critical incidents, twenty-four hours a day, included in the flat monthly rate.

Learn more about our managed IT services