Whether you detect downtime on your own or are alerted to it with the help of a monitoring service, you must take immediate action to correct the problem. You'll be better prepared to do so if you understand the possible causes of your site crash. Going through the process of troubleshooting downtime can be frustrating and sometimes costly, but look on the bright side: After going through this process once, you'll be better able to cope with - and ideally prevent - such crises in the future.

Pinpoint Server Problems
Downtime is often a product of a malfunctioning server. If you have your own Web server, you'll need to consult the particular hardware and software specifications involved in order to pinpoint the problem. If you're not technically inclined, make sure you have an expert in server technology on your team, or at least ready access to a consultant. Keep in mind that neither of these comes cheap.

If you outsource your site to a host, use any and all support channels to alert them to the problem and ensure that they fix it quickly. Timely response in such a crisis is a key criterion for a good Web host, so if your service doesn't come through for you, it might be time to consider switching.

Prepare for Traffic Fluctuations
A spike in traffic - a sudden increase in the number of people accessing your site - is the most publicized cause of downtime. If you've found no obvious problems in the server arena, it's time to look at visitor statistics. Have you just posted a hot promotional offer on your site, or announced a long-awaited redesign? If you didn't prepare for a corresponding increase in traffic, your efforts to win customers might ultimately have the opposite effect. Other causes of dramatic traffic increases include seasonal demand (holidays) and widespread unexpected publicity.