Tell us where to find you and we will reply within one business day.
Thanks. We will reply within one business day.
A guided walkthrough of FMIQ Nexus using representative sample data, following one job the whole way, from the moment it is reported to the invoice that matches the work.
Fewer things slip through, response comes faster, work on site stays safer, and every cost can be defended.
Whoever spots it, a cleaner, a teacher, a contractor, just says it or types it, hands free if their hands are full. FMIQ helps turn that into a structured record, but it never approves, closes or signs anything off. A person's name is on every sign-off.
Speak it or type it · the AI assists, people decide
A hazard is a photo, a place and a quick risk score in one short form. Tick one box and it becomes a make safe job, with its priority taken straight from how serious it is.
The risk score, set as it is raised
It comes through with its own reference, a severity and the job it raised, so nothing sits in an inbox waiting to be written up later.
The job linked back to the hazard
The work order carries the pump it belongs to, its site and its make safe clock, so whoever turns up is working from facts. The record builds up as the work is done.
The asset and the make safe clock on the job
Hot works, working at heights and confined spaces go through an approval chain that has to be completed before work starts. A contractor whose cover has lapsed is stopped before the job is ever handed over.
The approval chain, signed in order
Every live isolation shows who put it on, when, and which lock it is, so the board always reflects what is actually happening on site.
Live locks, each with a name and a time
Every asset holds its own story: what sits above and below it, what has been done to it, and what it has cost. That is what turns next year's budget into an argument you can back up.
Condition, cost and history, on the asset
From the job you raise a purchase order to the right contractor, and the money stays in view the whole way, from the first estimate to what actually gets paid.
Estimate, PO and payment, on one line








A serious job rarely needs one trade. The request becomes a prioritised work order, and that one job brings in the vendors it needs. Each vendor sits on its own purchase order, all linked back to the same work order and priced from a quote you accepted. It all stays one job, from the service request to the work order to the purchase orders. The timing, purchase orders and invoicing shown follow how a site is set up, not a fixed automatic path.
Illustrative sample workflowEvery job you just followed rolls up into one live view. It scales from a single site to your whole portfolio, and each team shapes it around what they own, so it is the picture you need, not a wall of charts.
The same picture goes up on the wall in the ops room and into the field crew's pocket. The field app works offline first, and native iOS and Android apps are publishing to pilot users now.
A job does not end when the spanner goes down. The cost, the asset's remaining life and the report all fall out of the same record, so the loop closes.
Every job's cost is captured, estimate through to paid invoice, and rolls up by asset, by site and by contractor.
Every asset carries its condition, its age against useful life and its running cost, so you plan replacement and capital works before things fail.
Every visit closes with a service report and its evidence attached, ready to hand to a client or an auditor.
Reporting, OH&S and cost run natively in the platform, filtered to a site, a client or a period. Where you already run your own reporting, safety or finance systems, they can connect through APIs, subject to integration, so the platform fits your stack rather than replacing it.
That was one job on sample data, start to finish. Bring us a real site of yours and we will run the same on your own estate: a live fault turning into a connected record, from the first report through to a cost you can defend.