• Do you feel that your IT systems are under performing?
• Are you losing sales because your IT systems are not working properly?
• Does your IT provider spend their time picking up the pieces after things have gone wrong rather than preventing problems from happening in the first place?

Many people think that if their outsourced IT provider responds quickly to their requests they are getting a good service. While a rapid response is certainly a basic requirement for any outsourced IT provider, it’s not where the service starts.

Fixing IT problems should be seen as a last resort. Ideally, all potential problems should be detected before they become a problem and prevented from happening in the first place. You are then free to use your IT provider to talk about adding value to the business.

The cost of not being pro-active

Have you ever stopped to work out the cost of downtime? To quantify downtime there are two primary factors to consider: productivity losses and business losses. Calculating both reveals wasted expenses and lost revenue.

Consider this real-world example which illustrates the effect of downtime: a recruitment company with 20 staff wasn’t using any form of pro-active server monitoring. The week before they implemented pro-active server monitoring, they had an outage of their main email server which lasted approx. 1.5 days. During that time they estimate they lost potentially up to £3,500 in new business, due to not being able to receive job specs or contact potential candidates by email. That would have been recurring revenue every week.

The problem was caused by low free disk space and when they tried to send an outbound marketing email campaign the server disk filled up and the server stopped working. If they had used a simple tool to pro-actively monitor the server this problem would have been prevented because they would have been alerted to the low disk space and would have had the opportunity to take remedial action before it became critical.

The importance of checking and testing backups

Most organisations we talk to are reasonably happy that they have robust backup systems in place. But when asked for evidence they start to get worried. The likelihood of failure in any single backup system is so high it’s almost guaranteed.

First of all you need to design resilient backup systems. Never rely on a single backup system. In this modern age of Virtualisation and Cloud Computing, many organisations have not two but three different types of backups, both on site and off site.

We’ll go into backup design in more detail in a future article, but suffice to say it’s vital that backups are managed actively on a daily basis. All backup systems should be configured to email logs to a central mailbox that’s either monitored continuously by a human or flows into a service desk. Log files should be checked for errors on a daily basis. Any errors found should be entered into an issue tracking system and fixed by a qualified technician. Issues should be reviewed periodically for trends.

Humans to monitor the monitoring systems

If you have automated systems in place to look for exceptions that’s great – we encourage that. But automated systems can go wrong and for that reason a human being should manually check the monitoring systems daily. The human doing the checks should also sign to verify they have checked the monitoring system and the backup log files.

But even that’s not enough. Log files do not always tell the full story. It’s perfectly plausible that a backup log file will not report any errors and yet the backup still failed or got corrupted. For this reason, all backup systems should be tested on a monthly basis. The only way to guarantee a backup has worked is to do a test restore of a random set of files to an alternate location. This doesn’t have to be a time consuming activity but it does need doing at least once per month.

If your IT department of outsourced provider does not do all of the above its time to review your strategies. I have lost count of the number of new clients that have come to D2NA for support because their backup systems were not being managed pro-actively and when needed they were not available. Under these circumstances the best case scenario is that data can be recovered but it takes an unnecessarily long time to do so. The worst case scenario is that data is lost irrecoverably.

Maintaining a robust backup system is not rocket science, but its time-consuming and there needs to be robust processes in place to support the technical systems. If you think this sounds expensive, ask yourself: what is the cost of losing my data or of extended periods of downtime?

Pro-active support - your minimum expectations from your IT department or outsourced service provider

• Pro-active monitoring of servers and network infrastructure, with exceptions generating console alerts on a dedicated screen that’s highly visible to the support team
• Monitoring systems should also generate emails alerts, delivered to a mailbox that’s monitored continuously during your agreed support hours
• Daily checks of automated monitoring systems by a human
• Daily checks of backup logs by a human
• Monthly test restore of a random set of files to an alternate location
• Monthly service reports
• Quarterly meetings with your Account Manager

Conclusion: when calculating the cost of your IT support provider, first calculate the cost to your business of downtime or irrecoverable loss of data.