Certain infrastructure applications should truly be core to running an operation, whether its internal IT support or running a SaaS or IaaS business. When implemented correctly, monitoring and ticketing apps give the organization tremendous insight and transparency to how the operation is working.
The analytics which can be derived can pinpoint trends which need addressing before outages occur or can tell you what is running well or what areas of the environment are struggling. It also can take the pulse of the customer experience using your services, whether its your help desk or your service offering to paying customers. So what’s the problem then?
In our experience we rarely have seen either one of these tools used to their potential. There are many reasons for this so, to keep this relatively short, I will just key in on a few areas:
App Selection: When companies select business applications, most organizations look at the business processes and then select an application which is appropriate for how they are going to run or change their processes for how they want to run their business. In most cases this is not the selection process for monitoring and ticketing systems. We see the majority of infrastructure organizations not looking at these systems as applications so the selection is “buy first then make it fit.” The end result is frustration and more investment in more monitoring because the app didn’t meet the needs of the organization.
Monitoring Application Evolution: Historically, an organization will buy the tool, get feedback from the different domains, overload the system with alerts and turn the service up. Someone will loosely be titled “the administrator” who will then keep the system patched. The problem? There is no process to keep the alerts current and eventually the monitoring is creating more false alerts than notifications that something truly needs attention. The teams know the false alerts but ignores them which puts the business at risk because, eventually, the alert may be real and nothing happens. The solution? First, the application needs to be continually updated. The best way is to tie the management of the system into your change control process. Changes to the environment affect how you want your world alerted. Very few companies tie these two together and it’s the main reason for “aging alerts.” Second, each domain leader should be accountable for ensuring the alerting has value and work with the administrator to keep alerts current. Ask the question “does this alert create an action on our part?” or “did someone want to see it but there wasn’t a good reason to set the alert up in the first place?” We have seen some monitoring services where 50,000 alerts are occurring a month making the app useless and a lost investment.
Ticket Applications: This application can be an amazing analytics tool as mentioned earlier. Defining how the app will work, the appropriate fields and information to populate a ticket, etc. needs to be thought out. It is our experience when working with many clients that the most common problem field is called “miscellaneous.” Miscellaneous does not help dig deep into issues and review trends in your world. It skews the data to the point where most of it is useless. In addition, incomplete ticket data such as close time and date and root cause fields not being completed doesn’t help with figuring out what needs attention in your world. One way to avoid this problem is utilize the ITIL framework when developing the functionality of the app. Take the time to figure out what data you want from the system, just as you would with a business application .
Tie the 2 Apps Together: We see many organizations running these apps as separate entities when in fact integrating a clean monitoring app with the ticketing service creates a very powerful data repository for analysis. When most of the ticket updating is left purely to people remembering to enter a ticket, you rarely get complete information so the analytics will be suspect. When these two run separately, not tying the data together is missing the power of the investment which was made in the first place.
In summary, take the time to figure out what problems you are trying to solve and what data is important to you in running your operation. Align this process with your application selection. Take the time to re-engineer your internal operating processes so you can get the most out of these applications and have accountability within the group to ensure the information is current and has value.