Replacing the whiteboard: dispatch, yard stock and customer ETAs in one system
A bespoke dispatch platform for a bulk landscaping supplier — drag-and-drop truck scheduling with conflict detection, a live visual yard map across 50+ bays, AI-estimated travel times and automated SMS ETAs to customers.
- Client
- Gympie Landscape Supplies
- Location
- Gympie, Queensland
- Services
- AI & Automation,Custom application development,Cloud & Infrastructure
Project stage. Built and in daily use. The efficiency figure below is the business's own assessment rather than a measurement we took independently — treat it as their view, not an audited result.
30%
efficiency gain in dispatch
Reported by the business. Less time on manual calculations and answering ETA phone calls.
50+
yard bays tracked live
Stock deducted automatically on delivery completion, with low-stock bays flagged visually.
Automatic
customer ETA notifications
SMS sent on dispatch with an AI-calculated arrival window, replacing inbound ETA calls.
Overview
Gympie Landscape Supplies runs a bulk aggregate and landscaping yard — soil, gravel, mulch, sand — dispatching trucks to customers across the region. Nexcentric built Gympie Dispatch, a bespoke cloud application that runs the yard's scheduling, stock and customer communications in one place.
This one is close to home: both businesses are in Gympie. We could stand in the yard and watch the problem happen, which is a meaningfully different starting point to reading a requirements document.
The problem: the whiteboard didn't scale
The yard ran on whiteboards, paper dockets and manual stock counts. That works until it doesn't — and as volume grew, four things started costing real money:
- Scheduling errors — rescheduling a delivery meant physically rubbing out and rewriting a whiteboard, with transcription mistakes and double-bookings the predictable result
- Inventory blind spots — tracking levels across 50+ bays, some measured in tonnes and some in square metres, was reactive rather than proactive
- Constant ETA calls — customers had no visibility of their delivery, so they rang the office, and someone had to stop and find out
- Manual conversions — dispatchers calculating round-trip travel times and converting weights to volumes by hand, with different factors for top soil versus gravel
Why bulk materials make this harder than normal logistics
A parcel is a parcel. Bulk landscaping materials are not: the same order might be quoted in tonnes, stored by volume, and delivered by truck capacity, with a different conversion factor per material. Get the conversion wrong and you either send a truck out underloaded or promise a customer more than fits.
Add travel time that varies with load and destination, and a yard where stock physically moves between bays, and you have a scheduling problem that a whiteboard can represent but not actually solve.
How the system works
The design brief was a digital nervous system for the yard — one place where the schedule, the stock and the customer's expectations all stay in sync without anyone re-keying anything.
- 1
Orders land on an interactive schedule
A CSS Grid Gantt chart shows every truck and every delivery slot. Rescheduling is drag-and-drop, and the system detects conflicts as they're created rather than when the truck turns up.
- 2
Travel time is estimated, not guessed
Google Maps and Genkit AI calculate round-trip travel time including loading buffers, so a slot reflects how long the run will actually take rather than someone's recollection of it.
- 3
Unscheduled orders wait somewhere visible
Orders that can't be slotted yet sit in a dedicated space rather than on a sticky note, and get placed when a truck frees up. Nothing gets forgotten because it had nowhere to live.
- 4
The yard map shows real stock levels
A custom SVG map of the yard renders each bay with its current fill level and flags low stock automatically. A dispatcher can see what's actually available without walking out to look.
- 5
Units convert themselves
The data model handles tonnes, square metres and units, with per-material conversion factors, so volumes are calculated rather than worked out on a calculator.
- 6
Customers are told automatically
When a load is marked dispatched, Twilio sends the customer an SMS with the AI-calculated ETA. They're ready when the truck arrives, and they stop ringing the office to ask.
- 7
Stock deducts on completion
Completing a delivery deducts the material from the bay, so levels stay current and reordering can happen before something runs out rather than after.
What the application does
- Drag-and-drop Gantt scheduling with automatic conflict detection
- AI-estimated round-trip travel times including loading buffers
- Interactive SVG yard map with per-bay fill levels and low-stock flags
- Multi-unit inventory across tonnes, square metres and units
- Per-material conversion factors for automatic volume calculation
- Holding area for unscheduled orders awaiting a truck
- Automated customer SMS on dispatch with calculated ETA
- Automatic stock deduction on delivery completion
- Printable and PDF delivery dockets with digital timestamps
- Mobile-responsive for use in the yard, not just at the desk
What changed in the yard
The system took the repetitive, error-prone parts of dispatching and made them the computer's job:
- Rescheduling takes seconds and can't silently double-book a truck
- Stock levels are known rather than estimated, so reordering happens in time
- Customers get told when their delivery is coming, without anyone making a call
- Fewer wasted trips where nobody was on site to receive the load
- Conversions and travel times calculated consistently rather than per-dispatcher
- Professional dockets replacing handwritten paper
What it's built on
Included for the technically-minded — and because the stack is part of why it could be delivered quickly and run cheaply.
Framework
- Next.js (App Router)
Backend
- Firebase Firestore
- Firebase Auth
AI
- Google Genkit
- Google Maps API
Frontend
- Tailwind CSS
- shadcn/ui
- CSS Grid Gantt
State
- Zustand
- Immer
Integrations
- Twilio (SMS)
- PDF docket generation
Automate the boring parts, keep the local knowledge
A dispatcher who has worked a yard for years knows things a system doesn't — which customers need a phone call, which driveways a truck won't fit down, when weather is about to change the day's plan.
The system doesn't try to replace that. It removes the arithmetic, the transcription and the ETA phone calls, so the person doing the job spends their attention on the decisions that actually need judgement.
That's the same line we hold on every project: automate the administration, leave the expertise where it is.
Outcome
Gympie Dispatch replaced a whiteboard, a paper docket book and a set of manual calculations with a single system that keeps the schedule, the yard stock and the customer all in sync. The business reports a 30% efficiency gain in dispatch, largely from time no longer spent on manual conversions and answering ETA calls.
It's also the clearest example of what we mean by a local IT partner. Both businesses are in Gympie, the requirements came from standing in the yard rather than a specification document, and the person who built it is a short drive away when something needs changing.
Have a process that’s eating someone’s week?
That’s how this one started. Book a free 30-minute review and we’ll tell you honestly whether it’s worth automating — and sometimes the answer is no.