The $5,600/Minute Figure and Why It's Misleading
Gartner's widely cited figure – that IT downtime costs an average of $5,600 per minute – gets repeated constantly in vendor marketing. It's not wrong, exactly, but it's based on enterprise-scale organisations with large revenue bases and high staff costs. For a 30-person professional services firm or a 60-seat contact centre, applying that number directly is meaningless.
The more useful way to think about downtime cost is to calculate it for your specific business. That means taking your daily revenue, dividing by your operating hours, and adding the cost of idle staff time. A business turning over $5 million per year, operating 250 days, with 40 staff at an average loaded cost of $80,000 per year, generates roughly $2,000 per hour in revenue and carries $1,600 per hour in staff cost. A four-hour outage costs that business at minimum $14,400 – and that's before you count anything except direct revenue loss and wages.
The real costs are almost always higher than the direct calculation, because they include factors that don't show up in an immediate ledger entry but absolutely affect the business's financial position over the following months.
Direct Costs: What You Can Measure Immediately
Direct costs are the ones that are easiest to quantify and the ones businesses most often use as their sole measure. They include:
- Lost transactions. If your e-commerce platform, point of sale, or booking system is down, sales that would have happened don't happen. Some of those customers come back; a meaningful percentage don't. For a business doing $50,000 per day in online revenue, a 6-hour outage doesn't just cost $12,500 – it costs whatever portion of that traffic found a competitor and converted there instead.
- Idle staff wages. Staff don't stop being paid because the systems are down. For a contact centre with 40 agents at $30–35 per hour, a 3-hour outage costs $3,600–4,200 in wages where no productive work occurs. Managers and senior staff trying to diagnose and escalate the issue add more.
- Emergency response costs. If your downtime involves a hardware failure or data corruption, you may be paying for emergency IT callout rates ($200–400/hour for after-hours work), expedited hardware replacement, or data recovery services. A single failed server with no hot spare and no monitoring contract can generate $3,000–8,000 in emergency costs before the system is restored.
- SLA penalties. If you operate under service level agreements with clients – common in professional services, outsourcing, and B2B software – downtime can trigger financial penalties. These vary widely by contract but are rarely trivial. A single SLA breach on a $500,000 annual contract might cost $15,000–25,000 in credits or rebates.
Indirect Costs: What Takes Months to Show Up
Indirect costs are where businesses consistently underestimate the true impact of downtime. They're harder to attribute directly to the outage, which makes them easy to ignore – but they're often larger than the direct costs.
Customer churn from a bad experience. Customers who couldn't reach you during an outage, whose order didn't process, or who received a poor service experience as a result of system failures, don't always call to complain. Many of them just quietly move to a competitor. For businesses where customer lifetime value runs $2,000–10,000, losing even five customers per outage adds up quickly. Research from Zendesk consistently shows that 50–60% of customers who have a bad service experience don't give a second chance.
Reputational damage on review platforms. A contact centre that goes dark during a crisis, or an e-commerce site that fails on a peak trading day, generates negative reviews. Those reviews persist and affect conversion rates for months. There's no clean way to put a number on this, but a one-star drop in Google rating for a local services business typically reduces enquiry volume by 15–20%.
Staff morale and productivity drag. Teams that regularly work around system failures – rekeying data, using manual workarounds, dealing with poor-quality information – experience fatigue and frustration that shows up in engagement scores and ultimately in turnover. The cost of replacing a trained staff member in a contact centre is typically 30–50% of their annual salary when you include recruitment, onboarding, and the productivity gap while they ramp up.
Regulatory and compliance exposure. For businesses handling personal data, financial information, or operating in regulated sectors, a downtime event that also involves data unavailability or breach notification obligations creates legal costs and potential regulatory action. Under Australia's Notifiable Data Breaches scheme, the cost of a single breach notification process – internal investigation, legal advice, OAIC reporting, customer notification – can run $50,000–200,000 before any penalty is assessed.
The Recovery Curve: It's Not Instant When Systems Come Back
One of the most consistently underestimated aspects of downtime is what happens after the systems are restored. The assumption is that recovery is immediate – systems come back, team starts working, everything returns to normal. That's rarely how it goes.
When a contact centre goes down for four hours, it comes back to a queue that has been building for four hours. Those calls don't disappear – they arrive all at once when systems restore, overwhelming the team and producing below-standard service experiences for the first two to three hours of operation. Average handle time spikes as agents work through complex, frustrated interactions. Abandonment rates remain elevated. The backlog may not clear until the following day.
For e-commerce, the equivalent is order reconciliation. When a transaction system goes down mid-process, you may have orders in partially completed states – payment taken but not fulfilled, fulfilment triggered but payment failed, inventory reserved but not committed. Untangling those takes significant manual effort, and errors during reconciliation create customer service problems that extend the disruption further.
Field service businesses face a scheduling collapse. Technicians who were dispatched based on now-inaccurate job data need to be redirected. Customers who received confirmation of appointments that didn't occur need to be called. The administrative overhead of a three-hour outage in a field service operation can consume eight to ten hours of management time to rectify.
A reasonable rule of thumb: budget the total productive time lost at 1.5–2x the duration of the actual outage, not 1x.
Case Scenarios: Three Business Types
E-commerce, $8M annual revenue, 12 staff. A five-hour outage during a promotional campaign costs approximately: $16,000 in lost transactions (assuming 60% of traffic doesn't return), $1,800 in idle staff wages, $4,000 in emergency hosting support, and an estimated $3,000 in customer attrition over the following month. Total: $24,800. The hosting contract with monitoring and failover capability that would have prevented this costs $18,000–24,000 per year.
Contact centre, 35 agents, outsourced service provider. A three-hour telephony and CRM outage during business hours costs approximately: $6,300 in wages for idle agents, $8,500 in SLA penalties across three client contracts, $2,000 in overtime to clear the backlog, and $1,500 in management time for incident documentation and client communication. Total: $18,300 – plus ongoing client relationship risk. A managed voice and CRM service with 99.9% uptime SLA costs $4,000–6,000 per month.
Professional services firm, 25 staff, revenue $4.5M. A full-day email and file server outage costs approximately: $14,400 in unproductive staff time (8 hours × 25 staff × average $72/hour loaded cost), $2,800 in emergency IT support, and delayed project deliverables that risk invoice disputes. Total: $17,200. Managed cloud infrastructure with 24/7 monitoring that would have caught the failing hardware before it failed: $3,500–4,500 per month.
What Prevention Actually Costs
The comparison that most businesses haven't done properly is the annualised cost of the outages they actually have versus the annualised cost of the infrastructure and services that would prevent them.
Most SMBs experience two to five significant IT incidents per year. "Significant" here means more than one hour of disruption affecting business operations. At $15,000–25,000 per incident (using conservative direct-cost-only estimates), that's $30,000–125,000 per year in downtime costs that are largely invisible because they're spread across the year and never aggregated.
A properly scoped managed IT services engagement – covering monitoring, patch management, backup and recovery, helpdesk, and incident response – typically costs $60,000–90,000 per year for a 30–50 person business. When you put that against $30,000–125,000 in downtime costs, the business case isn't marginal. And that comparison doesn't include the indirect costs discussed above, the regulatory risk, or the productivity drag from systems that are unreliable but not yet failed.
The conversation worth having isn't "can we afford managed IT services?" It's "what are we currently paying for downtime, and are we tracking it?" Most businesses that run that analysis change their answer.
CX Direct's managed services practice can help you run a downtime cost analysis for your specific environment and compare it against the cost of prevention. Contact us to get started.