Reporting fleet numbers: explain what was counted and over what period

A statement that a fleet is growing leaves several questions open. Is the number about owned vehicles, vehicles allocated to a programme, or vehicles ready for a particular shift? These quantities can move differently. Reporting becomes more useful when a metric describes a clear event or condition and the reader can see the boundary around it.

Name the unit and the period

Before quoting a figure, ask for its unit, coverage and reference period. A count of vehicles is different from vehicle-hours. A monthly total is different from a typical day. A citywide number may include several programmes with different operating windows. Record whether the figure is a snapshot, a cumulative total or an average.

The same discipline applies to percentages. Identify the starting and ending values, the denominator and whether the underlying group changed. Growth in the number of vehicles says little about available capacity if the service area or operating schedule expanded at the same time. For example, if ready vehicle-hours stay the same while required hours double, the share of coverage falls even when the vehicle count is unchanged. Report ready vehicle-hours against required vehicle-hours for the same place and period, and state both definitions.

Separate readiness, activity and results

A vehicle can be ready without receiving a trip, and a completed trip does not reveal every interruption that preceded it. Sparrow’s Ride Hailing operating context helps explain why preparation and delivered activity are separate questions. Ask which part of the process a reported number describes before using it as evidence of the whole service.

The UK government’s Magenta Book evaluation guidance distinguishes delivery measures from outcomes and treats causal impact as a separate evaluation question. In a mobility story, more completed work may be an observed result; attributing that change to one intervention needs additional evidence about what else changed.

Show what the records leave out

Ask whether the figures cover every shift or only records available at the reporting deadline. Missing entries should be described as missing. They should not silently become zero activity or be excluded from the comparison without explanation. Check whether a failed trip, duplicate record or late correction is handled consistently.

Canada’s guidance on data quality treats completeness, interpretability and timeliness as distinct considerations. This is useful reporting context: a recent dashboard can still omit important records, and a complete dataset can still be difficult to interpret. State provisional status where later reconciliation may change the result.

Make the comparison repeatable

A reader should be able to follow the comparison without access to private operational data. Publish the definition, period, geography, inclusion rules and material limitations. Where commercially or personally sensitive records cannot be shared, explain the method and the level at which the claim can be checked.

Sparrow’s Press & Media page is the starting point for reporting enquiries. A useful request identifies the exact metric and proposed comparison rather than asking only for a headline number. Consistent definitions make the eventual story clearer and help a later update explain whether the service changed or the measurement did.