Post 07 of 12
Which metrics explain last-mile cost and service, and how to build a dashboard around them.
Last-mile delivery is the most expensive part of many supply chains. Capgemini Research Institute estimated it at 41% of total supply chain costs. It is also the part customers see. Yet many operators manage it with a few network-wide averages, such as on-time rate, cost per stop and vehicle utilisation. Those show what happened, but not why. This article covers where last-mile data comes from, the metrics that explain cost, and who needs to see what.
A parcel or distribution network already records most of what it needs. The difficulty is that each system records a different part of the delivery, with its own identifiers and timestamps.
Dispatch and route planning: the planned route, the stops and the delivery windows.
Telematics: where the vehicle went and how long it stood at each stop.
Driver app: proof of delivery, failed attempts and the reason recorded.
Customer portal and notifications: what the customer was told, and when.
Until these are joined in one data model, with one definition of a successful delivery, every metric shows only part of the picture.
Industry research gives a sense of the scale.
41%
of total supply chain costs attributable to last-mile delivery
Source: Capgemini Research Institute, 2019 ↗84%
of consumers said they were unlikely to return to a retailer after one poor delivery experience
Source: Convey consumer survey, 201897%
of organisations surveyed said their current last-mile delivery models are not sustainable at full scale
Source: Capgemini Research Institute, 2019 ↗Network-wide averages hide most of the useful detail. The metrics below are most useful broken down by zone, route, time window and carrier.
This is the share of deliveries completed on the first attempt. Each failed attempt means paying twice for driver time, vehicle cost and handling, and a second delivery has to be planned.
Break it down by zone, time window, carrier and delivery type. Zone data shows access problems. Time-window data shows when customers are not at home.
Compare zones where customers receive a narrow delivery-window notice with zones where they are only given the day.
Use the history in planning: schedule high-failure zones in the windows where more deliveries succeed.
Stop density is the number of successful deliveries per route kilometre or per hour. Time at stop, compared with the planned time, shows where parking, access or building entry slow drivers down.
Map demand by postcode and time of day to find routes that could be combined.
Compare planned and actual stops per route. Routes that always miss their plan usually have a fixed cause, such as a loading restriction or a hard-to-access business park.
Re-plan using measured stop times instead of a standard allowance per stop.
Track the cost of re-deliveries and returns as a share of route cost, by customer and product category. This puts a money value on failed deliveries, which makes it easier to justify fixes such as pick-up points, better address data or customer notifications.
Each role needs a short view of the metrics it can act on.
Operations managers: today's routes, open exceptions, estimated finish time per route and failed first attempts, updated through the day.
Fleet managers: vehicle use per shift, maintenance cost per kilometre and driver scores over a rolling four weeks.
Customer service and sales: on-time rate per customer, re-delivery rate by product category and complaint reasons by zone.
Finance and management: cost per successful delivery by zone and carrier, the trend against the previous period, and forecast against actual volume.
Keep each view short. A dashboard with forty metrics tends to be opened once and then ignored.
The bottom line
Last-mile analytics pays off because the costs are large and most of their causes can be measured. Join dispatch, telematics, driver-app and customer data in one model. Track first-attempt rate, stop density and re-delivery cost by zone, and give each team a short view of the metrics it can act on. Delta BI built this kind of analytics layer for a UK parcel carrier (FastTrack Logistics, name changed), cutting average delivery time by 30% and driver overtime by 22%.
How Delta BI can help
Transport and logistics BI: unify TMS, WMS, ERP and telematics data into live dashboards for route optimisation, carrier performance and on-time delivery.
See how it worksStart with the decision
Tell us about your challenge. We will reply to arrange an assessment and identify a practical route to delivery.
Request a data assessment