A weekly fleet review should end with a decision: which delay to fix, which truck needs attention, or which lane needs a different price. Six measures are enough to start if every number has a definition, an owner and a reliable source.
Build the scorecard from completed work
Choose one week and include trips completed in that week. Keep a separate list of trips still in progress. For each measure, show its numerator, denominator, exclusions and last update. Compare against the previous few weeks only after the definitions stay consistent. A spreadsheet is sufficient for a first version.
Separate an operational review from a financial close. Delivery and delay data can be reviewed quickly; fuel receipts, workshop invoices and driver settlements may arrive later. Mark a cost figure provisional until those records are complete.
Six measures worth reviewing each week
1. On-time delivery
Trips delivered by the agreed deadline ÷ completed trips with a deadline.Keep customer-requested changes and missing timestamps visible. Use the delivery event, not the time a dispatcher later closed the job. A late delivery needs a reason the team can act on: loading delay, traffic disruption, breakdown or an unrealistic promise.
2. Turnaround time at loading and delivery sites
Release time − arrival time for each visit, then compare the median and the slowest visits by site. Averages alone can hide a repeated three-hour queue. Split queue, loading or unloading, and document-release time when the timestamps exist. The truck turnaround guide shows a six-timestamp visit record.
3. Unplanned downtime
Unplanned unavailable hours ÷ scheduled vehicle-hours. Count hours only inside the period when that vehicle was expected to work, and distinguish a planned service from a breakdown. Keep the vehicle, fault, workshop start and release times with the total. The downtime guideexplains how to collect these records without treating all workshop time alike.
4. Empty running
Empty kilometres ÷ all measured kilometres for vehicles whose load status and distance are known. An empty return is still a real cost. Define what counts as empty on mixed-load and repositioning legs. If the trip changes trucks, attribute each leg to the truck that drove it.
5. Fuel economy
Measured kilometres ÷ litres consumed gives km/L. A fuel purchase is not necessarily fuel consumed during that week, so use a consistent full-to-full test or reconciled tank data before comparing vehicles. Group by vehicle class, route and load; investigate data gaps before blaming a driver. See the fuel consumption method.
6. Direct cost per kilometre
(Fuel + workshop spend + driver settlement) ÷ measured kilometres for the same vehicle and cost window. Label this as a direct cost measure, since finance, insurance, tolls and overhead may be missing. Compare it with revenue per kilometre only when both use the same distance basis. The worked KES example shows the arithmetic and the effect of empty returns.
Put data quality beside the headline number
A clean-looking percentage can be misleading if half the trips have no recorded arrival time. Add a coverage line to each measure: how many trips or vehicle-days qualified out of all relevant records. Opening odometer readings belong to the moment the driver starts a trip, and closing readings to completion. A planned route length should remain a plan, not quietly become measured distance.
For example, a 90% on-time rate from 9 of 10 documented deliveries says less about the fleet if another 20 completed deliveries have no reliable deadline or arrival timestamp. Review the missing 20 before celebrating the percentage.
Make the weekly meeting short and useful
- Check missing readings, timestamps and receipts first; assign someone to resolve them.
- Look at the trend by vehicle, route and customer rather than one fleet-wide average.
- Choose one exception with a clear owner and a next action.
- Record the decision and revisit it the following week.
A useful scorecard is a record of decisions as well as numbers. If turnaround worsens at one depot, update the dispatch plan or speak to the site. If cost per kilometre rises for one truck, check fuel data, repair history and missing distance before changing its rate.