How to stop embarrassing enterprise website downtime

Enterprise website glitches and downtime are embarrassing and unprofessional. They can and should be avoided

Imagine visiting your favourite shop during normal opening hours, only to find all the lights out, nobody inside and a sheet of paper taped to the door saying, “Closed unexpectedly due to too much custom. Please come back later, though we’re not sure when.”

If you were a particularly loyal customer you might go away and come back later. More likely you’d take your business elsewhere, especially if the problem kept occurring. Why waste your time on an unreliable organisation that clearly doesn’t value your custom enough to keep its doors open?

This is the choice facing visitors to your enterprise website when it becomes unreliable or unavailable. The chances are that this is happening more often than you realise, because there are so many potential points of failure. That’s because a modern corporate website depends, at minimum, on the following:

  • Reliable internet connections between clients and the web server
  • Dependable hosting with high percentage server uptime
  • Timely renewal of security certificates and other credentials
  • Sane use of development servers for thorough testing, with total separation from live servers
  • Sufficient CPU and data bandwidth to handle spikes in popularity and demand
  • A web admin team that understands the business need for website stability and continuity
  • Content providers (either within the company or as consultants) who supply content appropriate to the structure of the site
  • Fast, good quality web code accessible on a wide range of mobile and desktop browsers
  • Timely updates to the server software and any web applications running on top of it, to prevent malware and bug-related exploits
  • Web designers who can balance presentation against server footprint, i.e. an understanding that bandwidth-heavy designs have performance penalties when scaled up to tens of thousands of users.
  • Database servers capable of scaling to handle high numbers of requests
  • Testers/QAs who know what to look for and what to do when they find it

Reading that list, which isn’t even comprehensive, makes one wonder how any websites stay up as reliably as they do. Yet when sites do become unavailable, more often than not it’s the human part of the equation that’s at fault.

There’s no technical reason for any enterprise site to become unavailable, yet that still happens today. For example, “bandwidth exceeded” messages are disappointingly common. Website facelifts that break mobile browsers now occur so frequently that they’re barely worth mentioning. Usually they’re rolled back within a couple of days, once the enraged customer feedback reaches the right ears. A couple of weeks later, a much-scaled-back facelift will be launched, with far less fanfare. This time it might work, or at least break fewer devices.

To continue reading this article register now

7 secrets of successful remote IT teams