Regional carrier
AI Consignment Desk
My three best operations people spend the day reading tracking numbers aloud.
A regional carrier running roughly 400 deliveries a day. Three operations staff are nominally responsible for exception handling — failed deliveries, damage claims, escalations — but the phone rings continuously with status enquiries, so exception work happens in the gaps and therefore late. Delays are usually discovered by the customer before the carrier notices them.
The product
AI Consignment Desk
The queue, the approval screen, the exception list. Every number is readable and every handover is visible.
Built by Koderead to show the operating system. Figures are a worked example.
Watch one record
Where the handover actually happens.
A screenshot shows a state. This follows a single record through the system, including the point where it stops and waits for a person.
Vehicle reports a fault08:14
Telematics flags it before any human notices.
Impact calculated08:15
Against the schedule, not against an estimate.
Customer told first08:16
Two minutes after the fault, before anyone rings in.
Operations works the exception08:22
Not the status call. The actual problem.
State
- Consignment
- CD-884122
- Destination
- Hull HU3
- Promised window
- 11:00–13:00
- Customer aware
- No
Delivery commitments are quoted from the operational system, never estimated
Deeper into the build
The screens where a person is still in the loop.
A queue is easy to show. What matters is what happens at the record, at the approval, and at the exception — because that is where the design either holds or does not.
| Consignment | Detected | Customer told | Channel | |
|---|---|---|---|---|
| CD-884122 | Vehicle fault, 11:04 | 11:06, revised 17:30 | SMS | Sent |
| CD-884140 | Depot delay, 08:22 | 08:24, revised window | Sent | |
| CD-884151 | Route congestion | 09:41, 40 min late | Sent | |
| CD-884124 | Refused at door | Awaiting instruction | — | Operations |
Eighteen notifications sent today before any customer asked. Each one is a call that never arrived, which is the saving that deflection rate never captures.
- Status calls per 100
- 4.1
- Median response
- 8s
- Exception work done same day
- 94%
was 22
any hour
was 51%
- Operations time on status lookups — before68%
- Operations time on status lookups — after11%
- Operations time on exceptions — after74%
The three most operationally capable people in the building stopped being a voice interface to a database.
The change
Where the work moved.
Amber marks a step the system performs. Solid ink marks one a person still performs. The counts above each column say the same thing in numbers.
Today
4 human steps- Performed by Customer
Calls to ask where it is
- Performed by A person
Operator opens tracking
Mid-exception work
- Performed by A person
Reads status aloud
- Performed by A person
Exception work resumed
Context lost
- Performed by Customer
Calls again after the delay
- Performed by A person
Delay handled reactively
After
1 human step- Performed by Customer
Asks on any channel
- Performed by Automated
Consignment identified
Reference, address or account
- Performed by Automated
Live status returned
Position, window, exception
- Performed by System
Delay detected against schedule
- Performed by Automated
Customer notified first
Revised window sent
- Performed by A person
Works genuine exceptions only
- Automated
- A person
- System
- Customer
Scope of authority
What this system is not allowed to do.
On a real engagement this is a document your operations lead signs before anyone writes code.
- Delivery commitments are quoted from the operational system, never estimated
- Dangerous goods and customs exceptions always route to qualified staff
- Damage and loss claims are acknowledged but never assessed or settled automatically
- Driver hours and statutory safety limits are hard constraints, never optimised against
Built from
The solutions behind this build.
Your version of this
This took a mapped workflow before it took any code.
If this looks like your regional carrier, the place to start is the same place we started here.