Many service delivery organizations use their help desk systems as a primary source of KPI data relating to outages of core IT infrastructure. Whilst this is a useful ancillary function of the helpdesk system, it has several important shortcomings.
Potential problems include:
- Outages may only become evident when the customer raises a ticket; which is too late for any proactive service management.
- The time taken to resolve a raised ticket may include administrative overhead time and not reflect the actual time that an application or server was unavailable.
- It’s important to ensure that multiple tickets related to the same outage are correctly consolidated so as to not result in double (or worse) counting of unavailability.
- Without accurate assignment of the ticket to the correct configuration item/business service, an issue may not get the appropriate level of attention.
- Producing accurate and meaningful reports of the availability of key business services may require the manipulation of many tickets raised over a reporting period.
- Turning an array of resolved tickets into accurate business service availability reporting will require a great deal of manual work.
Using ticket receipt and resolution times as measures of availability, and relating individual calls to business services does not support easily “reacting” or reporting before SLAs are potentially breached.
By using data collected through regular monitoring however, it is possible to avoid most of the above problems:
- By receiving alerts about degrading events in the monitored estate, it becomes possible to address issues before they become service impacting.
- By creating a service hierarchy based on all the monitored components that contribute to the smooth running of a business service, it is possible to gain “real-time” health status of a business service, as well as producing an “availability to date” report against agreed service objectives.
- Because the availability calculations are made based on monitoring of each key element in the infrastructure, there is no requirement to concern yourself with how many, if any, affected end-users raised tickets due to the same outage.
- By assigning an impact and priority to each configured business service, any IT issues can be addressed according to their effect on services.
- As the historical status data is held with the reporting system, there is greater flexibility to rapidly produce service availability reports for time periods of your choosing.
SLA’s/ SLO’s are often linked to service credits for missed targets. A monitoring and reporting system that offers the ability to track performance against an agreed objective is essential in allowing you to maximize customer satisfaction whilst minimizing exposure to financial penalties.