A metric measures and evaluates the effectiveness of IT service management processes.

For example, a metric could measure the effectiveness of the incident resolution process by calculating how long it takes to resolve an incident.

Sometimes a metric can be easily obtained from the data. For example, to find the number of incidents that were created today, a report will simply count the number of incidents in the incident table with a Created date of today. Often, however, metrics need to be gathered as data is updated. For example, determining how long an incident was assigned to a certain group requires collecting information about assignment changes and calculating the duration of each assignment.

The Metric plugin provides an easy, declarative way of defining metrics. Once defined, the data for the metric will be gathered, and instances of the metric will be calculated and stored. By an instance we mean a specific occurrence. For example, the "Assigned to Duration" metric measures the duration of time an incident is assigned to an individual. The metric is defined by creating a metric definition of type "Field value duration" and selecting the "Assigned to" field from the Incident table. A metric instance is then created for each incident assignment showing its duration. Reporting on the duration of incident assignments becomes easy.

Reporting on a metric is done using the database view that links the metric to the table on which it is defined.