Connected label search
Search by product, crop or pest in any order. Compatible choices narrow as you go, with exact-use aerial checks before label-derived rates are available.
AUSTRALIAN AERIAL SPRAY PLANNING
From label.
To load plan.
With a clear trail.
Bring product selection, mixing calculations and job records into one practical workspace. Keep the source behind the numbers in view.
Aerial-MVP preview · Australian scope · No APK required
THE WORKFLOWSOURCE → PLAN
01
Find the exact use
Product + crop + target
02
Build the mix
Rate, area, carrier & components
03
Keep the record
Save, revisit & print
Your field decisions.
A traceable planning record.
875app-ready products*
15,379released use paths*
One workflowselection → calculation → record
*Packaged dataset snapshot: 29 September 2026. Not a live registration check or blanket approval for every use.
BUILT AROUND THE JOB
Less disconnected work.
More useful context.
Start with a product, crop or target. Carry the selected evidence through to the job—not just a number copied into a calculator.
Mixing with context
Plan quantities from your area and chosen rate. Retain carrier rules, required components and source-linked conditions instead of flattening them into a single rate.
Saved jobs & templates
Save jobs locally, reuse mix templates and transfer job records through JSON export/import. Restored selections require fresh label-rate authority.
Field-ready records
Prepare printable mix sheets and client field reports. Keep source references and incomplete-information notices visible in the output.
Quotes & invoices
Create multi-line quotes and invoices, keep document numbers and statuses, and print or save through your browser’s PDF controls.
An installable web app
Responsive layouts, light/dark themes and Learning Mode. Offline workflows are available after a successful initial cache; test your device before field use.
EVIDENCE, NOT ASSUMPTION
The source matters
as much as the answer.
Agdron’s released paths are built from product-specific evidence, with source identity, rates, carrier requirements and mapping checks. An unresolved path stays held.
Explore the evidence model →- 01
Exact use, not ingredient similarity
One product’s label does not authorise another product’s use.
- 02
Unknown stays unknown
Missing or conflicting values are not filled with guesses.
- 03
The current label remains essential
Structured warning coverage is incomplete. Read all applicable label directions and restrictions.
GET STARTED
Open online. Prepare before the field.
Use a normal browser session on a reliable connection. The current database download is approximately 692 MB—Wi-Fi is recommended.
Allow the initial database and offline cache to finish. Browser storage is local, not cloud backup; export important job records.
Install using your browser’s install or home-screen option where supported. Test an offline restart and your job workflow before relying on it.
Know the boundaries
This is an Australian aerial-MVP preview, not an automatic compliance assessment. Warning backfill is unfinished, external label links need internet, and physical phone/tablet field acceptance remains pending. Do not use missing warnings as evidence that no restriction applies.
YOUR NEXT JOB, IN ONE PLACE
Meet Agdron Mixer.
OUR STORY
A business plan.
An idea worth building.
The idea for Agdron Mixer came while developing the business plan for Agdron’s drone-spraying business. Working through that plan highlighted an opportunity: make spray-mix planning easier and less prone to costly mistakes.
Make one part of the job easier
The starting point was not to replace every part of running a spray business. It was to improve one important part: working out the mix. Bringing product information, rates, carrier requirements, calculations and job records together could make the process easier to follow and help reduce opportunities for costly mistakes. That is the aim—not a promise that software can eliminate error.
Learning how to work with AI
Building the app has also meant learning how to give AI useful direction: breaking a large task into smaller jobs, explaining the rules, reviewing results and correcting mistakes. “Teaching” AI here means refining instructions, examples and checks—not training a new AI model from scratch.
Give the repetition to the tools
AI-assisted coding and scripts help with the time-consuming work: organising label evidence, extracting structured information, building screens and running repeatable tests. The value is in getting through that work systematically, while keeping the source behind each result visible.
AI helps build it. Evidence still has to support it.
A confident AI answer is not proof. The build process uses source references, validation rules, tests and explicit holds for unresolved information. It has involved iteration and rework, not a single prompt that produced a finished app. Warning coverage and physical-device testing remain unfinished; AI assistance does not replace the current label or the operator’s judgement.
The journey is still underway: turn an idea into something useful, learn from testing, and be honest about what is ready and what still needs work.