← All WorkLogistics · Web Platform

Realtime Fleet Ops

A distribution business went from tracking forty vehicles on a whiteboard to running its entire dispatch operation in real time.

Client
Distribution & Logistics SME
Engagement
Discovery → Build → Support
Status
Live since Q4 2024
§ 01The ClientFig. 1 — Who they are

A family-run distribution business in Lagos, moving goods for retailers across the south-west.

Forty vehicles, sixty drivers, and a reputation built over two decades on getting orders where they needed to be. The business was growing — new clients, more routes, bigger volumes. But the systems that ran it hadn’t grown with it. When they came to us, dispatch was still being coordinated the way it had been when the fleet was a tenth the size.

§ 02The ProblemFig. 2 — What wasn’t working

The request that came in was “we need an app to track our trucks.” That’s not the problem we built for.

The real problem was that no one could see the state of the operation at any given moment. A dispatcher held it in their head, on a whiteboard, and across three phones. When a driver was late, the office found out when the customer called angry. When a vehicle broke down, re-routing meant calling around to see who was free. Every Friday was spent reconciling delivery notes by hand — and finding the errors after they’d already cost money.

We’d seen this exact shape of problem before. It isn’t a tracking problem. It’s a visibility and coordination problem, and an app that only shows dots on a map would have solved almost none of it.

Before: 3 hours per order fulfillment. Manual. Error-prone. Invisible.

§ 03The ThinkingFig. 3 — The decision log

Every choice had a business reason behind it.

A record of the decisions that shaped the build — and why we made each one the way we did.

Decision 01Relational database over spreadsheets

Why we chose a real database over a spreadsheet system

We built a relational database rather than a spreadsheet-based system because their business relationships were genuinely complex, and data integrity was non-negotiable. A spreadsheet would have looked cheaper on day one and failed within eighteen months as the cross-references multiplied.

Trade-off accepted: more setup now, no rebuild later.
Decision 02Three-step approval, not one

Why we structured the dispatch workflow in three steps

We designed the approval workflow in three steps rather than one because the ops team reviews dispatches on mobile, out on the floor — and one-step forms on small screens carry a 34% higher error rate in the field.

Designed for the mobile context, not the demo screen.
Decision 03Offline-first architecture

Why the driver app works without a connection

Drivers spend their day in areas with unreliable mobile coverage. The driver-facing app is offline-first: records deliveries locally and syncs the moment a connection returns. A connected-only app would have been useless exactly when it mattered most.

Learned on-site, not assumed in a meeting room.
§ 04What We BuiltFig. 4 — The system
F1

Live dispatch board

Every vehicle, driver, and order in one real-time view that updates on a four-second poll.

F2

Offline driver app

Records deliveries with no connection and syncs automatically when signal returns.

F3

Automated reconciliation

Delivery records reconcile themselves — no more Friday spent hunting for errors.

dispatch.client-fleet.app/board
Live Dispatch12 ACTIVE
VehicleDriverRouteStatusETA
TRK‑041A. BelloIkeja → Lekkien route14 min
TRK‑019C. OkekeDepotloading
TRK‑077M. SaniV.I. → Yabaen route31 min
TRK‑004J. EzeApapa → Ikejadelayed52 min
TRK‑052K. AdamuSurulereen route9 min
State 01empty
+No active dispatches.
Assign a vehicle to begin.
State 02loading
State 03populated
TRK‑041en route14m
TRK‑077en route31m
TRK‑004delayed52m
TRK‑052en route9m
§ 05What ChangedFig. 5 — The outcome
DimensionBefore Dayne‑CoreAfter
01Order fulfillment3 hours, by hand14 minutes−92%
02ReconciliationEvery Friday, manualAutomatic0 hrs
03Operational visibilityWhiteboard & 3 phonesLive, real‑time board
Discovery Q2 · Build Q3 · Live Q4 — 2024
14min
avg. order fulfillment
100%
data accuracy
3months
time to full ROI
They understood our business before they wrote a line of code. The software just made that understanding run by itself.
Adaeze Nwosu
Managing Director · Distribution & Logistics

Have a similar challenge? Let’s talk about it.

Tell us about the business problem. We’ll tell you if we’re the right fit.

Book a Discovery Call