Software for energy and water companies, built for the van

Yes, and it is usually the biggest part of the job, so we treat it as the main risk rather than a footnote. Before we quote, we list every place your numbers live: control room software, meter readings, maps, customer records, finance and the contractors' own tools. For each one we work out how it hands data over and who controls access. Each goes into the plan on its own line. Where something has no clean way of handing data over, we say so up front and price the way round it. And we only read from operational software. We never write into anything that controls equipment.

It starts with one call. After that you get what we will build, when, and for how much, in writing, before anyone starts. You are not paying by the hour and hoping for the best. The build is a project with a price on it. Once it is live, keeping it running is a separate monthly fee that covers watching the software, fixing faults within an agreed time, improvements, security patches and a monthly report. We do not publish a rate card, because the price depends almost entirely on how many feeds we must read from and how many types of equipment are in scope.

We keep it running. That is the second half of what we do. For a monthly fee we watch the software, so a problem is spotted by us rather than reported by a crew standing in a field. We fix faults within a time we agree with you, keep improving it and keep it patched. Every month you get a report showing what broke, what we fixed, what changed and how the whole thing is performing. If you ever want to bring it in-house, we hand over the code and the access properly. The default, though, is that we stay.

Yes, and it is built in from the start, not patched on later. The tablet holds the job list, the equipment details and the forms on the device itself, so an engineer can finish an inspection in a basement or halfway up a hill and the record is captured properly. Photos, signatures and forms queue up and send when the tablet gets a signal again. If two people worked on the same piece of equipment, neither wipes out the other's notes. We test this in real conditions before launch, not on the office wifi, because a field tool that needs a signal is a paper form with extra steps.

We build the report to whatever layout you are obliged to produce, under whichever rules you work to. The trick is to capture the underlying evidence once and properly: what was inspected, what the test said, what condition the equipment is in, what was done about it, when, and who did it. The return is then generated from that. When the required layout changes, and it will, that is a change to one output rather than a month of retyping. We ask to see your current returns and the guidance you follow during the first call, so what we store lines up with what you have to prove.

No published energy job, and we will not pretend otherwise. What we have done sits right next door. We built maintenance software connected to a machinery maker's ERP that flags a likely failure 10 to 14 days ahead, saved $1.2M a year and cut unplanned downtime by 72%. We built attendance software for a three-plant manufacturer, a support assistant for a fintech with a first reply in under two minutes, and one platform for a garage group across nine branches. The shared shape is work out on site, crews spread about, records that must stand up to inspection, and joining up software nobody wants to replace.