7 delivery fleet metrics that actually matter
You do not need a dashboard with forty charts. You need seven numbers that tell you whether the week was healthy and where to look next.
1. Earnings per active hour
This is the number drivers feel most directly. Divide total earnings by active hours, not by clock hours. A driver who is online for ten hours but only active for six has an earnings-per-hour figure based on six.
Compare each driver with your own fleet median before comparing them with an external benchmark. City density, vehicle type, and platform mix change the number too much for a single national figure to be meaningful.
2. Pay per order
Pay per order exposes the effect of order mix. Two drivers can earn the same amount while one completes 120 short orders and the other completes 80 long ones. If pay per order falls while order volume rises, the fleet may be absorbing more low-value work.
Use it as a diagnostic, not a ranking. A driver with a high pay per order may simply have received better orders.
3. Orders per hour
Orders per hour measures throughput. It is sensitive to batching, restaurant wait times, and distance, so compare it within a city or zone rather than across the whole fleet.
A sudden drop is usually an operational signal: longer waits, more traffic, or a change in dispatch. A sudden rise can mean short orders and lower pay per order.
4. Completion rate
Completion rate is the share of accepted orders that were delivered. It belongs in every payout review because it is both a quality signal and, on some platforms, a pay modifier.
A completion rate below the fleet's normal band deserves a conversation, not an automatic penalty. Look at the driver's cancellations and the reasons before changing pay.
5. Tip share
Tip share is tips divided by total earnings. It shows how much of a driver's compensation depends on customer generosity, which is less predictable than base pay.
A very high tip share is not automatically good: it can hide a low base rate. A very low tip share can be normal in markets where tipping is uncommon. Read it alongside pay per order.
6. Distance per order
Distance per order tells you how much driving each delivery requires. Rising distance per order with flat pay per order means the effective hourly rate is under pressure.
It is also a data-quality check. A value above roughly 200 km per order usually means the distance column is counting something other than delivery distance.
7. Payout concentration
Concentration is the share of the total payout that goes to the top three drivers. It is a risk measure: high concentration means the fleet depends on a small number of people, and a single absence changes the week.
There is no universal target. Track your own trend and ask whether the concentration is caused by performance, availability, or dispatch bias.
Put the seven numbers to work
Do not review all seven every week. Pick the two or three that match the decision in front of you: a pay review, a retention conversation, a dispatch change, or a data-quality check. A metric is only useful if it changes a decision.
Questions
What is a good earnings per hour for a delivery driver?
There is no universal number. It depends on the city, vehicle, platform, and whether hours are gross or active. Compare against your own fleet median and track the trend rather than aiming for an external benchmark.
Should distance be part of the performance score?
DeliverPayout shows distance but does not score it, because more distance is not automatically better performance. Distance is a cost and workload input, not a quality signal.
Why do two drivers with the same earnings have different scores?
The balanced score combines earnings, order volume, pay per order, and quality. A driver with the same earnings but fewer orders and a higher rating can rank differently from one with more orders and a lower rating.