Calculating the Costs of Downtime

As companies become increasingly dependent on technology, the probability of experiencing system failures or unavailability also increases. Downtime translates into real costs and losses and, depending on the nature of the company, these costs can be quite significant. Therefore, it is important for all organizations to calculate the cost of downtime in order to understand its implications on the business, realize the importance of system upgrades and maintenance programs, and be able to justify the associated costs.

Breaking Down the Costs of Downtime

Downtime can be defined as any incident of system unavailability that inhibits a user's access to data, or their ability to use a system's functions to perform business duties. Downtime can be caused by a range of reasons such as power outages, hacker attacks, computer viruses, and employee sabotage. The degree to which downtime affects a company varies substantially based the reach of the outage (individual, departmental, across the entire enterprise), complexity of the system, the users' dependency on the system, contingency plans in place, and business scale. Due to such variability, companies find it difficult to quantify the costs associated with system failures.

Total downtime costs can be broken down into the following (see Figure 1):

• Productivity costs. During a system outage, costs arise due to the time employees lose performing their regular duties. Additionally, costs arise because of the extra time employees need to make-up for their lost productivity. The extent of the latter cost depends on the effect on productivity factor applied to the calculation. For example, one downtime hour may only translate into half an hour of catch-up time.
• Business costs. Lost revenue opportunities during an outage have large impacts on total downtime costs. Not only does downtime affect one-time transactions, but it also has negative effects on future sales if, during downtime, customers leave to competitors and then remain loyal to them. However, not all transactions are lost during an outage as customers may return upon system recovery, so estimations should be made accordingly.
• Recovery costs. Recovery costs include the price paid to repair the system, IT staff overtime, and perhaps the use of third party consultants or technicians to restore services.
• Indirect costs. Several costs arise from system failures due to ripple effects such as late charges incurred because of missed deadlines, extra fees for phone calls or mailings, and customer apology compensations. Other costs such as damaged reputation or low employee morale are harder to quantify but should be considered and included if a reasonable estimate is possible.

Figure 1. Total Downtime Costs
Source: Info-Tech Research Group

The maximum downtime costs for an organization can range anywhere from the tens of thousands of dollars to even billions. A recent example was Amazon.com's two hour system outage in June, 2008 during which the company was estimated to have lost up to $31,000 per minute - over a million dollars an hour.

Recommendations

1. Refer to historical data. By examining records of costs, duration times, and other effects that resulted from previous system failures, more accurate estimations can be made about downtime costs. The value of this data emphasizes the importance of capturing and recording all activities following future system outages.
2. Calculate total costs of downtime. Use the Info-Tech Advisor “Downtime Cost Calculator” to determine the impact of all downtime costs: productivity, business, recovery, and indirect.
3. Perform a high-low analysis. Estimate the minimum and maximum impacts of a system outage to provide a reasonable range. Typically, low estimates account for failures that affect only a limited number of employees and require little repair time. However, high estimates account for worst-case scenarios such as day-long, enterprise-wide outages. By building conservatism into the calculation, companies can prepare a better contingency plan.
4. Project the total costs. By projecting the hourly downtime costs to a year, the expected annual cost can be determined. This figure is often much higher than managers expect, but is valuable to understand the implications of downtime.
5. Calculate the payback of downtime-prevention investments. Perform a payback analysis using the annual downtime cost figure to make a case for system upgrades, maintenance programs, and other contingency strategies (note: the tool assumes the investment solution will eliminate all downtime costs, which is not always the case). Although these investments also have significant costs associated with them, the payback analysis will help determine the worthiness of the spend.

Bottom Line