Formula Reference

How to Calculate the True Cost of Downtime

Updated April 2026 · Sources: ITIC 2024, Uptime Institute 2025, Ponemon 2022

The Formula

Downtime Cost =
  Lost Revenue
  + Lost Productivity
  + Recovery Cost
  + Reputation / Churn Cost
  (x Regulatory Multiplier for Finance / Healthcare)

Quick Estimate (30 seconds)

If you just want a rough number:

Company SizePer MinutePer HourSource
Small Business (<50 staff)$427$25,620Ponemon 2022
Mid-Size (50-500 staff)$9,000$540,000Ponemon 2022
Large Enterprise (500+ staff)$23,750$1,425,000ITIC 2024

Multiply your per-minute estimate by outage duration. For an accurate number with your own inputs, use the calculator.

1Lost Revenue

Lost Revenue = (Annual Revenue / Hours Per Year) x Industry At-Risk % x Outage Duration (hrs)

Hours per year is 8,760 for 24/7 businesses or 2,080 for business-hours-only operations. The industry at-risk percentage accounts for what fraction of revenue is genuinely lost (vs delayed) during an outage - ecommerce loses nearly 100%, SaaS loses 60-70% because subscription revenue is mostly delayed rather than lost outright.

Caveats to consider: Subscription revenue is often delayed, not lost - a SaaS company may recover most deferred revenue within 30 days. Partial degradation (slow site vs full outage) reduces the at-risk percentage. Cached content or offline modes can reduce impact for content-heavy sites.

WORKED EXAMPLE

Acme SaaS: $50M ARR / 8,760 hrs = $5,708/hr x 65% at-risk x 4 hrs = $14,841 lost revenue

2Lost Productivity

Lost Productivity = Employees x Fully-Loaded Hourly Rate x Outage Hours x Productivity Loss %

The productivity loss percentage varies significantly by role. Manufacturing workers on a halted line lose 100% - there is literally nothing to do. Knowledge workers lose 40-60% because they can handle emails, meetings, and offline tasks. Customer support agents lose 80-90% if the ticketing system or telephony is down.

Role TypeProductivity Loss %Reason
Manufacturing / Production100%Line-down, no alternative work
Customer Support80-90%System-dependent roles
Sales / Business Dev70-80%CRM and comm tools down
Finance / Operations60-70%ERP/accounting access lost
Engineering / Dev40-60%Can use local tools; CI/CD blocked
Management / Strategy20-40%Mostly async; in-person meetings continue

WORKED EXAMPLE

Acme SaaS: 200 employees x $85/hr (fully-loaded) x 4 hrs x 80% = $54,400 productivity loss

3Recovery Cost

Recovery costs are easy to underestimate. They include overtime pay for engineers working the incident, external vendor escalation fees (premium support tiers), infrastructure rebuild or rollback costs, customer communication effort, and post-mortem labor (typically 4-8 engineer-days for a serious incident).

A useful heuristic: recovery cost is typically 8-12% of the combined lost revenue and lost productivity. For a $100,000 direct-cost incident, expect $8,000-$12,000 in recovery overhead.

WORKED EXAMPLE

Acme SaaS: ($14,841 + $54,400) x 8% = $5,540 recovery cost

4Reputation / Churn Cost

This is the squishiest component - and the one most calculators skip. The methodology: estimate the churn rate uplift caused by the outage (typically 1-5% of affected customers for a public outage), multiply by affected customer count, multiply by average customer LTV.

For B2B SaaS with high LTV, a 1% churn uplift on 1,000 customers at $10,000 LTV each = $100,000 in reputation cost. For high-volume B2C (ecommerce), brand damage is less about per-customer value and more about search visibility and PR recovery cost.

Key caveat: Reputation cost is the hardest to attribute and the first CFOs challenge. Model it conservatively and cite the Ponemon brand-damage research for credibility.

WORKED EXAMPLE (Acme SaaS)

$14,841 lost revenue x 18% brand-damage rate (B2B SaaS) x 4 hrs = $10,686 reputation cost

Full Worked Example: Acme SaaS

$50M ARR, 200 employees, 4-hour outage during business hours, SLA 99.95%, B2B SaaS vertical.

ComponentCalculationAmount
Lost Revenue$50M / 8,760h x 65% x 4h$14,841
Lost Productivity200 x $85/h x 4h x 80%$54,400
Recovery Cost8% of direct costs$5,540
Reputation / Churn$14,841 x 18% x 4h$10,686
Total$85,467

What is surprising: Productivity loss ($54,400) is 4x the revenue loss ($14,841) for a knowledge-worker-heavy company. Most CFOs only look at revenue loss. The productivity number is what actually stings.

What Most Calculators Get Wrong

x

They use 3 inputs (revenue, employees, hours). Missing SLA tier, industry, customer count, and churn model.

x

They use the Gartner 2014 $5,600/min figure as a benchmark without noting it is 12 years old and survey-based.

x

They skip reputation cost entirely - the component that compounds over 6-12 months post-outage.

x

They ignore SLA credit asymmetry - the cloud SLA credit is a rounding error compared to your actual loss. See /sla-credits.

x

They assume 100% revenue loss regardless of industry - SaaS and subscription businesses often recover most revenue, just delayed.

Frequently Asked

What is the formula for calculating cost of downtime?
Cost of Downtime = Lost Revenue + Lost Productivity + Recovery Cost + Reputation/Churn Cost. Each component is calculated separately and then multiplied by an industry compliance factor for regulated industries like finance and healthcare.
How do you calculate lost productivity during an outage?
Lost Productivity = Number of Affected Employees x Fully-Loaded Hourly Rate x Outage Duration in Hours x Productivity Loss Percentage. The productivity loss percentage varies by role: manufacturing workers lose 100%, knowledge workers 40-60%, customer support 80-90%.
What is a fully-loaded hourly rate?
The fully-loaded hourly rate is an employee's total cost to the employer, including salary, benefits, payroll taxes, office space allocation, and equipment. It is typically 1.25x to 1.4x the base salary hourly rate. For a $100,000/year employee, the fully-loaded cost is $60-$70 per hour ($50/hr base + 25-40% burden).
What do most downtime calculators get wrong?
Most calculators use only 3 inputs and ignore reputation/churn cost entirely. They use a one-size-fits-all productivity loss percentage, forget recovery cost (overtime, vendor escalation, post-mortem labor), and use the Gartner 2014 benchmark without noting it is 12 years old and likely understates current costs.