Last month, I introduced our fall Disaster Recovery Education initiative and promised to provide you with more valuable information. This week, we’ll talk about length of downtime (period without access to some or all of your computing resources) and its cost to your organization.
A decade ago, many companies considered a week or more of downtime―in the wake of a disaster―to be acceptable. Small and medium businesses (SMBs) were becoming increasingly computerized and reliant on the Internet, but the hunger for instantaneous communications was nowhere near as ubiquitous as it is now.
Today, many large enterprises measure acceptable downtime in hours (if not minutes) and cannot imagine going for a week or more without access to their email, corporate files, customer records, business processes and other important revenue drivers. SMBs need to adopt this same strategy to compete effectively―and it’s becoming increasingly affordable to do so.
Let’s consider the cost of downtime. It varies by the business, industry and business application, of course, but it is considerable in every instance. Analytical software developer Alinean ran some numbers and determined that the average cost per minute of infrastructure downtime is $700. Having a business application down can cost far more.
Alinean placed the cost per minute of messaging outages at $1,000; with e-commerce pegging at an estimated $10,000 per minute. Supply-chain management software, where system downtime can literally disrupt your entire supply chain for a week or more following restoration, tops the charts at $11,000 per minute.
For SMBs that don’t conduct e-commerce, rely on complicated supply chains or have legions of salespeople in the field, these numbers may seem excessive. To help you get a handle on your own cost of downtime, I’ve provided some practical calculations you can run.
Quantifying downtime, at its most fundamental level, requires calculating two primary losses: reduction to individual or workgroup productivity and loss of transactions. Calculating both helps you pinpoint wasted expenses and lost revenue.
For the first number, you’ll run the “CPA” calculation:
C = average employee cost (salaries or wages + benefits)
P = number of people affected
A = amount of time they are affected
As of September 2013, the average U.S. average “burdened” salary (compensation plus taxes and benefits) per employee was $29.11 per hour, per the U.S. Department of Labor. (Calculating your real number would be more accurate, of course).
Multiply this figure by the number of employees in your operation, then by the amount of downtime you consider acceptable in the event of an emergency. That figure equals your cost of lost productivity.
For business impact, the calculations become more difficult. While you can calculate this per employee, it’s likely easiest to multiply your transactions per hour or day by the average profit per transaction and then by the downtime duration.
These calculations do not consider lost customer good will (or even lost customers), nor do they consider any overtime you will authorize to bring the firm back up to speed in the event of a disaster. These numbers are hard to gauge, but a good figure would be to increase the “people” and “business” numbers by a percentage that increases with the length of downtime (5% for durations of a few days; 25-30% for downtime of a month or more.)
Once you’ve calculated your downtime costs for a variety of disaster scenarios (server crash―1-3 day of disruption; fire or major weather event; possibly weeks or months of disruption), compare it against the cost to reduce your downtime exposure.
If it still seems expensive to implement a near-instant recovery plan, give me a call. Our IT experts can show you how affordable it is to never lose access to your data―even for a minute.