Here's an eight-letter word, twice as bad as any four-letter word, that no business leader wants to hear: downtime. Today's businesses are far more dependent on IT services than ever, and that's true whether you're talking about internal IT services (like ERP) used to drive strategic operations, or external IT services used to satisfy client and customer demand.
Among the more daunting long-term potential consequences of downtime to the organization are these:
• Lost revenues because business couldn't be transacted. A recent Ponemon study tells us the average cost of downtime for US-based organizations is a stunning $5600/minute.
• Diminished brand strength, because the company is seen as unreliable. The same study suggests the average length of downtime was 90 minutes, leading to roughly $500K of costs per incident.
• Evaporating market share, because unhappy customers go to competitors.
Quite a mess in short. And it's a mess that's rapidly getting bigger. Aberdeen Group found that between 2010 and 2012, the cost per hour of downtime climbed an average of 65%!
Now, all of this is obvious in my own area of application performance monitoring (APM), and while downtime is a problem, there is a bigger issue, more subtle lurking just beneath the surface. When we believe everything is running smoothly but don't know something is wrong, that’s when the most damage happens.
For instance, suppose a BizTalk-based service is up and running in a holistic sense, but operating in a subtly inconsistent manner — difficult to detect — that leads to lost transactions from time to time. By this I mean occasionally lost e-mails, lost database entries, lost purchase orders, etc. Time spent by customers resending email and clients waiting or employees spending time looking for an invoice that has not come through – all of these cause more loss in productivity and reputation over a longer period of time.
According to Pricewaterhouse Coopers, the average organization, spends $120 searching for a lost document and wastes 25 hours recreating each lost document.* But what we don't know is how much productivity is lost through not knowing when a problem exists, searching for a file that is not lost at all. Waiting on document resends, searching for "missing" invoices or customer relationships that need to be repaired due to an apparent miscommunication because information is not flowing smoothly in a system all impact company efficiency and costs.
Application performance, and service uptime, can be affected by myriad factors — some as subtle as a gradual shortage of key computational resources. That's why it's important to find a way to granularly monitor your system. To have control and visibility over the problems that are happening so you can decide which ones to tackle is key.
Over time, I think we're going to see that kind of granular insight play a larger and larger role in APM as a field. But in the meantime, it's important to have a clear view of the information that flows throughout your organization so you can see any problem lurking out of sight.
Ivar Sagemo is CEO of AIMS Innovation.
Related Links:
* DocuSense Blog: How Much are Lost Documents Costing You?