Infonetics Says Midrange Businesses Lose Lots of Dough from Network Downtime

It’s a story we hear time and time again from server makers, the software providers who make high availability clustering software, or the companies that provide disaster recovery products and services: downtime, even a little bit, costs companies a lot of money. They talk about the servers, but they often don’t talk about the broader network that links them together, to end users, and to the outside world.

Often, the statistics that these vendors cite is based on statistics culled from the largest enterprises in the world, which obviously have more complex IT infrastructures and more money at stake for each passing second on the clock when downtime hits the servers. That’s why Infonetics Research, a market research firm based in San Jose, California, recently did a survey of medium-sized businesses–those with more than 100 and fewer than 1,000 employees–and focusing on the broader network, not just servers, to quantify the amount of money these businesses are losing to downtime.

The company’s study, called The Costs of Downtime: North American Medium Businesses 2006, indicates that the average midrange company loses $867,000, a sum equal to about 1 percent of their annual revenues, because of IT system and network downtime, which comes to 140 hours per year. So even if you have your servers clustered and highly available at five nines (99.999 percent available), having such terrible uptime for the broader network–that 140 hours works out to about 98 percent availability–is very disruptive. According to Infonetics, 56 percent of that outage time is due to pure outages–a network device or a server being offline because of a failure, configuration issue because some operator or administrator did something wrong, or whatever. Application degradation and failure account for a total average of $213,000 in annual costs, with outages having about 65 percent of that cost and degradation being 35 percent. Infonetics says that midrange companies do not usually have the tools to identify network performance degradation. Service provider outages are a big part of the problem, too, says Infonetics, and outages for external networks or software services are not under their control. This all makes ensuring uptime a tricky business across complex networks.

“There isn’t a single problem area that organizations need to focus on, which would be a simpler fix,” says Jeff Wilson, principal analyst at Infonetics Research. “Every decision is critical, from hardware selection, to product setup, and from employee training to SLAs with service providers. Human error is the most troubling, because fixes for human error are elusive and require process changes and retraining, which can take a long time and be very expensive.”