A dashboard can look reassuring while leaving the next decision unclear. A service team may report that most requests were handled on time, yet the manager still cannot tell which service is struggling, whether important requests were missed or what should change. A useful key performance indicator begins with the decision it is meant to support.
Start with the service outcome
Ask what the service needs to achieve for its users. A helpdesk may need to restore access to an essential service, while a cleaning team needs to meet defined conditions in specified spaces. Those outcomes lead to different measures. Counting calls answered might help manage helpdesk capacity, but it does not establish that the underlying problem was solved.
Write the management question in one sentence: “Are priority requests receiving the agreed response?” or “Are inspected areas meeting the agreed condition?” If nobody can explain what decision would change when the result changes, reconsider whether the measure belongs on the main dashboard.
Give the number a definition
A percentage needs a numerator, a denominator and a reporting period. A time measure needs a defined start and stop. Record the source of the data, the person responsible for checking it and the treatment of incomplete records. A count of completed jobs also needs a clear meaning for “completed”: work carried out, inspected, accepted or administratively closed can describe different events.
Definitions make comparisons possible. If one team measures first response and another measures final resolution, their figures should not appear under the same label. Changes to measurement rules should be visible in the report so that readers can distinguish a real trend from a change in calculation.
Separate the measure from the target
The indicator describes performance. The target expresses the level being sought. For example, the proportion of eligible requests receiving an on-time response is a measure; the agreed percentage to achieve is its target. A target without a stable definition is difficult to interpret, and a measure without context can encourage unhelpful comparisons.
Pair the result with an explanation
A short review should state the result, the evidence quality, the main exception and the proposed response. Include volume when it matters: a percentage based on a handful of cases deserves different interpretation from the same percentage based on a large, stable population. Investigate unusual results before assigning a cause or changing the service.
At the next review, choose one indicator and trace it back to individual records. Then ask whether the action taken last time addressed the actual service problem. This turns reporting into a practical conversation about performance.
Reading and further learning
Concept reference: Brian Atkin and Adrian Brooks, Total Facility Management, fourth edition, Wiley, 2015, printed pp. 199–201. The book discusses performance measures, their connection to objectives and the distinction between indicators and targets. This article and its examples are independently written.
Explore M13 — SLA & KPI Development and Performance Monitoring for a structured learning pathway. Course suitability, delivery and arrangements are confirmed before booking.