Where field service companies lose money
Ask any owner of a service company where the leaks are and you'll hear the same list. Jobs that were done but never invoiced. Materials used but never charged. Technicians driving across town because the schedule was built without a map. Customers calling the office to ask when someone will arrive. Paper job sheets that come back a week later, illegible.
None of these are technical problems. They're hand-off problems. The information exists, it just doesn't travel from the field to the office and back reliably.
The full flow, from request to paid
Good field service software follows a job through its whole life, not just the scheduling part.
- Request: from a call, email, portal or contract schedule, captured once with the right details.
- Planning: assigned to the right technician by skills, location, availability and parts.
- Dispatch: the technician gets the job on the phone with address, history, notes and required materials.
- Execution: checklists, photos, signatures, time and materials recorded on site, even offline.
- Stock: parts taken from the van or warehouse deducted automatically.
- Invoice: drafted from the actual work and materials, pushed to accounting.
- Follow-up: the customer is notified, the report is shared, the next service is scheduled.
What the office needs to see
The office needs a live board: which jobs are open, who is where, what's late, what's waiting for parts, what's done but not invoiced. Most importantly, it needs exceptions surfaced, not a wall of data. The job running over budget, the technician with no job after lunch, the customer who has called twice.
Managers also need margin per job and per customer, which is only possible when time, materials and invoicing live in the same place.
What the technician needs
Technicians need the opposite: less. The next job, how to get there, what the customer reported, what was done last time, and a fast way to close the job with photos and a signature. If closing a job takes longer than filling a paper sheet, they'll keep using paper.
Offline support is not optional. Basements, rural sites and plant rooms don't have signal, and the app has to keep working and sync later.
Generic FSM apps vs a custom system
There are many field service management apps, and for a small trades business with standard jobs they're a reasonable start. The limits show when your work doesn't fit their model: contracts with service levels, multi-step jobs involving several crews, property owners and tenants who both need updates, warehouse stock shared across vans, or billing rules that differ per client.
One of the systems we've shipped is exactly that kind of case: property and renovation, with repairs, crews, warehouse, tenants and owners connected, so a leaking tap reported by a tenant becomes a job, a plumber assignment, parts taken from stock, a tenant notification and a cost line on the owner's report, without anyone retyping it.
Adding an AI layer
Once jobs, people and stock live in one system, you can ask it questions in plain language. Which jobs are done but not invoiced? Who is free tomorrow afternoon near the north depot? What are we running out of? An AI receptionist can answer the customer asking when the technician arrives, because it can see the job. An AI clerk can draft the invoices from yesterday's completed jobs for someone to approve.
The rule is the same as everywhere: AI acts on real data, with the same permissions as the person asking, and every action is logged.
How to measure whether it's working
- Days from job completion to invoice.
- Share of completed jobs invoiced within the week.
- Jobs per technician per day.
- Calls to the office asking for status.
- Materials used vs materials charged.
Rolling it out without stopping work
Field businesses can't pause operations for a software project, so the rollout has to happen while jobs keep running. The approach that works is to start with one crew or one region, run the new system alongside the old paper for a couple of weeks, and fix what the technicians complain about before widening.
Migrate customers, sites, equipment history and open jobs before go-live, so technicians see the full history on day one. Train with real jobs, not a demo. And make the office stop accepting paper sheets on a fixed date, otherwise the old way lingers for months.
Give technicians a direct line to report problems with the app in the first weeks, and fix the small annoyances quickly. A button in the wrong place costs seconds per job, and seconds per job across a whole crew is what decides whether the system gets used or quietly ignored.
Getting a system that fits your jobs
We build operations systems around how your jobs actually run. Core (€18,000) covers one company with up to five roles, full data migration, and phone and desktop apps that work offline in the field. Maxxed (€48,000) adds the AI layer, integrations with accounting, calendars and banking, a client or tenant portal, and automations. After kickoff, you get a first working version running on a sample of your real jobs within 24 hours. If your jobs still come back on paper, apply and show us a week of job sheets.