The problem
A family-owned group in the Philippines ran 4,000 hectares of farmland on Excel, paper and phone calls. Ask about a tractor's condition and the answer described it as it was two weeks earlier, because that was how long the report took to arrive. Ask where it was, and the best anyone could say was "it should be heading somewhere."
What we built
A single system shared by the office and the field. Managers and dispatchers plan staff and equipment from it, and the crews who take equipment out record what they used and what condition it is in, from their phones. Before designing anything, I spent time on the estate with those crews to see how the work actually happened. Internet in the fields was unreliable. A full offline mode was outside the project's scope, so we agreed a simpler routine with the client: crews sync before they head out and again when they return. A system the crews do not trust gets worked around, so I built it around what they told me.
What changed
The estate could see its equipment as it stands today: where each machine was last recorded, what condition it is in and how it is being used. That record made equipment analytics possible for the first time. The system is still running, now maintained by the client's own IT team.
Technical notes
A mobile-friendly web app for the field crews, with desktop dashboards for admins and managers covering operations and analytics. React front end; Node.js and MySQL (Sequelize) back end, self-hosted on the client's own Unix servers. Offline sync was scoped and set aside in favour of a sync routine at the start and end of each field trip. GPS tracking was discussed as a next step.