Requires Free Membership to View
When you register, you'll begin receiving targeted emails from my team of award-winning editorial writers on the latest customer relationship management (CRM)and call center technology issues today. Our goal is to keep you informed on the hottest issues facing this fast-changing industry.
Hannah Smalltree, Editorial Director
There is no concise answer to this question because there are a lot of things that can happen and each vendor might prepare differently for those possibilities. For example, downtime is something that many people keep under a microscope, for good reason. Most large SaaS CRM vendors have taken the step of ensuring their facilities are up to standards that the industry has developed over the last few years. These standards insulate the physical facility as much as possible from things like earthquakes and floods. They also have uninterruptible power supplies and each has rigorous on premises security to prevent intrusion. Nevertheless, people still worry about backup and recovery as well as security and encoding their communications and a great deal more. Suffice it to say that the foreseeable problems — the ones that made some people skittish about this form of computing — have been dealt with.
Pitfalls include downtime and data loss which can take us in a lot of directions. Data loss is not something that I lose sleep over, given the redundancy of modern systems but down time can still be an issue. Downtime might not cause you to lose your data but it may prevent you from getting at your data which can be just as bad. Many of us think about SaaS in Boolean terms — it's either on or it's off and we have been conditioned to think that it's always on. But consider what happens if a primary site were to suffer a catastrophe that caused the site to restore a backup. While it is very possible that no data would be lost, it is also possible that restoring from a backup could take a while —24 to 48 hours or more.
My advice is to ask your vendor what the recovery plan is in such an event. Some vendors have mirrored data centers so that a problem in one will simply mean that another center has to take over the load. That could degrade performance temporarily but considering the alternative it might be fine for a short time.
This was first published in June 2007