Service Desk Performance Measurement

Service desk performance should be measured with a balanced set of indicators, because most service desk metrics are interlinked and a focus on one may negatively affect another. John Sansbury suggests a set covering resolution, service levels, staffing, call handling and satisfaction.

• Resolution and service level measures include first-line incident fix rate, the percentage of incidents and service requests completed within service level by priority, and average incident duration by priority, measured from incident occurrence until normal service restoration rather than from logging to closure.
• Average cost per major incident should include support effort, lost user time, business impact and reputational damage; it is effective at justifying investment in problem management.
• People and call measures include unplanned absence per analyst, analysts moved into other internal roles or leaving within 12 months, incidents fixed per analyst, call wait time, call duration, abandonment rate, time to abandon and user satisfaction.
• Each measure needs a specific interval: the longer the baseline, the easier the target is to meet, so better providers use shorter baselines. Measures should also reflect the customer experience.
• Where much contact is by phone, plot analysts against call volumes with average speed of answer and abandonment rate, daily from 15 or 30-minute observations. This reveals analyst-to-call fit far better than reporting the percentage of calls handled within service level.