A quick note first: this client is anonymised. I've withheld the company's name and changed identifying details. The work is real, but the screenshots I share of it show the software running on fictional demo data. Every name, address and figure in them is invented.
Work falling through the gaps
This is an operations platform I've built for an Australian remedial building company. Before it, the company's admin was spread across a shared inbox, off-the-shelf customer management software for trades, accounting software, spreadsheets and phone calls, and work fell through the gaps between them.
Some of those gaps were surprising. New enquiries had stopped reaching site-visit scheduling, and it turned out the import step only ran when someone remembered to press a button. Quotes died quietly in three places: site visits nobody closed out, visits nobody quoted from, and finished quotes nobody sent. Meanwhile, the GPS tracker company's daily reports cut off mid-morning while the vans kept driving all afternoon.
The company already had an in-house system. In June I rebuilt it from scratch around one idea: the software should do the admin and ask the office for decisions, not the other way round.
One list for every decision
The home screen is called Today. It's a single ranked list of everything that needs a person: drafted messages, site visits that need attention, quotes and jobs missing details, emails, and tasks the AI has prepared. Each item can be approved, approved with edits, dismissed or snoozed, and every decision is logged.
The first time it ran, Today surfaced 129 items. Ninety of them were a backlog of quotes and jobs missing details that, until then, nobody could see.
In the background, automatic jobs check the inbox, schedule site visits, read documents and review quotes. Each one has its own on/off switch under one master switch, and they all start switched off. The office decides what runs.
Lead triage where nothing sits idle
Every email gets sorted by type: new lead, complaint, warranty defect, reschedule, invoice or payment, supplier and so on. Each one becomes a task or lands on a triage list, so nothing sits unread. The spam filter refuses to bin anyone who matches a known customer.
New enquiries now flow into site-visit scheduling automatically, every time the inbox is checked, so there's no button left to forget. New leads join a pool, and a planner packs the quoting technicians' days: oldest lead first, split by area, within each technician's travel radius and work days.
Quotes drafted from the site visit
When a technician finishes a site visit, the platform drafts the quote from their photos and notes, using the company's approved price list. Every line is checked against that list and rebuilt from it, so the AI can't invent a price. The office edits the draft, handles variations and sends a branded PDF.
Crews, schedules and the fleet map
Field crews have their own phone app: today's primary, next and backup jobs, scope checklists with sign-off, photo and video upload, and one-tap logging of customer contact. The office sees the same job cards and can view any team's board.
And as of this week there's a Fleet tab. It pulls positions from the GPS trackers already fitted to the vans, keeps them in the company's own database, and shows where each van went on any day, along with when it was turned on and off. No more reports that stop at morning tea.
What I'm not claiming
I'm not putting numbers on hours saved or faster replies to customers. What I can say is that the office and field apps were checked against the live system with real company data in June, and the platform keeps growing.
Good software does the admin and asks you for decisions. Bad software makes you do the admin for it.
The full write-up is in the case study, and there's more on consulting and custom builds and AI assistants that draft the admin for you. If your business runs on an inbox, a spreadsheet and somebody's memory, I'd like to hear how your week works.