How do you run 920 site visits a year without a single system?

What they had

The information behind those visits lived in five systems: a spreadsheet, a job-management package, Google Maps and the accounts package, plus the re-typing that held the four of them together. The MD scheduled every job himself, typing the same visit into each in turn. Setting the week's schedule took most of a morning, and every re-type was another chance for an error to creep in — a wrong date, a wrong address, a price that didn't match the quote.

None of the systems was bad on its own. The problem was that none of them was the whole picture, so the MD was.

What we built

One system, built around how they already worked.

• A rules-based scheduler. The MD's rules, turned into an algorithm: eight-hour crew days, surveys go to the dedicated surveyor, similar-distance jobs grouped together. It schedules a week of visits in seconds and refuses a move that breaks the rules, instead of letting the error through.
• A quote engine. A completed survey becomes a priced quote, costed on visits, hours, miles and materials. Anything over the sign-off threshold waits for the MD; everything under it goes straight out.
• Offline field capture. Crews complete visits at sites with no reception. The record saves on the phone and syncs the moment there's signal. Nothing is lost and nothing needs re-typing.
• Role-based screens. The office sees the schedule board. The crew sees today's run. The MD sees the week's numbers and the decisions waiting on him, and nothing he doesn't need.

What changed

The MD got roughly 90% of his scheduling time back. The re-typing went, and the re-typing errors went with it. When a crew completes a visit, the office sees it immediately, ready to invoice — not sitting on a phone waiting to be typed in.

The demo on our homepage runs this system's actual logic, with the client's data swapped for demo data. Move a job around it and you'll see the rules at work.