Gatling downtime cost calculator

WHAT DOES ONE OUTAGE REALLY COST?

Tell us about the service, the outage, and who works on the fix.

We calculate lost revenue, lost productivity, recovery and reputational impact, and show what companies in your industry report.

Sourced benchmarks, labelled assumptions

Downtime cost calculator

Learn how to start high-scale testing

The service

Revenue generated by this service

I don't know this figure

Only the revenue that depends on this service

Who does the outage affect?

Industry

The outage

Downtime duration

People working on the recovery

Engineers, SREs, support and managers on the incident

Employees who rely on the tool

Staff blocked by the outage, not the responders

How badly does the outage block their work?

No published benchmark exists for this. These two levels are our assumption.

Cost rates

Sourced defaults, in US dollars. Replace them with your own rates if you know them.

Engineer cost per hour

Employee cost per hour

How we calculate this

Lost revenue (customer-facing services) = revenue per hour × downtime hours. Yearly revenue is divided by 8,760 hours, monthly by 730, daily by 24. Internal services count as $0. If you don't know the revenue, we don't guess it and show your industry's benchmarks instead.
Lost productivity = responders × engineer cost per hour × downtime hours, for engineering time pulled off planned work. For internal tools we add: employees who rely on the tool × employee cost per hour × downtime hours × productivity loss (50% or 100%).
Recovery = responders × engineer cost per hour × half the downtime, for post-incident work after the outage ends. For banking and travel, an optional 10% of lost revenue covers customer compensation and reimbursements.
Customer churn and reputation = lost revenue × an industry multiplier, from 0.3 for media to 1.0 for banking. These multipliers are our estimate, set higher for regulated sectors where outages have led to fines and compensation.
Likely range applies the low and high end of each assumption: engineer cost, post-incident effort (0 to 1 × the downtime), compensation (0 to 45% of lost revenue), the churn multiplier (half to double), and productivity loss (25% to 100%).

Our approach

Lost revenue is your service's revenue spread evenly over the year, multiplied by the length of the outage. If you don't enter revenue, we don't guess it: we show what companies in your industry report in recent surveys by ITIC, Enterprise Management Associates, New Relic and Splunk/Oxford Economics. Several of these are sponsored by monitoring vendors and rely on executives' own estimates, so treat them as context, not a forecast. Lost productivity and recovery are based on the time of the people involved, priced with official labour-cost statistics (US Bureau of Labor Statistics, Eurostat) and the Stack Overflow Developer Survey, including employer benefits and social charges. Customer churn and reputational impact are the least certain part. No independent study measures them by industry, so we apply a conservative multiplier to lost revenue. Every figure is listed with its source and year, and you can change the rates.

Sources

This calculator provides an estimate. Survey figures lean toward large enterprises, and studies sponsored by monitoring vendors come from companies with an interest in high downtime costs. Spreading revenue over 8,760 hours assumes steady demand, so an outage at peak time costs more, and some revenue may be recovered later.