by Craig McLellan, CEO
There are numerous reasons why companies must create disaster recovery plans:
Shareholders and other stakeholders want to ensure their assets are protected
Employees want to ensure continued employment
These are very clear ramifications. What is often less well understood is clinical disasters’ manifestation over time.
Imagine a scenario where a mid-sized distributor experiences a 10-minute server outage due to human error. Deemed the “best” resolution to the problem, the outage’s solution was executed, understanding that a small amount of data would be lost.
Unfortunately, the acceptable data loss allowed orders to be processed erroneously for two days. Customers call with reports of incorrect shipments. Assuming the customer base is somewhat consistent with ordering patterns, this company experiences a 10 per cent error rate for the current month’s orders (2 days in 20 business days).
This error rate would prompt at least 10 per cent of customers to look for an alternate supplier. Imagine if half of these “shoppers” found it attractive to switch and never came back!
In less than 30 days, company revenues would fall five per cent. Since it’s a distribution company, margins are thin (< 20%) so that revenue drop is equal to a 25 per cent profit reduction.
In 90 days, the company’s revenues have stabilized at the new lower level and management looks for other ways to elevate profits to pre-failure levels. One of their few remaining levers is reducing labour costs.
Assuming the company, like most in business today, has optimized labour costs, there isn’t room for a 10 per cent staff cut without negative customer satisfaction.
This is how a company’s failure spiral begins. An event triggers a five per cent revenue drop, which management offsets with double the reduction in labour. Customer satisfaction drops, which ultimately cause more to leave — and the loop rewinds and begins again.
It’s scary but at least two in five companies recovering from disasters cease to exist within 18 months. However, these business-ending disasters are not well known. The example above is an illustration highlighting the impact of a rash decision on a small event and its long- term effect on the company.
Understanding how people architect application infrastructure is becoming more important than its recovery plan. Good design means incorporating best-in-class technology and leveraging cost effective products such as Server Replication Service instead of using a one-size-fits-all traditional disaster recovery plan.