Knowing what is planned, what is happening and what needs attention

Where this came from

A conversation with someone running a construction and electrical business. Not a project, not a proposal — a description of a problem, which is where most of this work starts.

He was not asking for automation and was not complaining about paperwork. What he described wanting was oversight: a reliable sense of what was planned, what was actually happening on site, what had been completed, and what needed his attention — without having to ring round to assemble it.

That distinction is worth sitting with. The value he was describing was not speed. It was confidence.

The problem, as he described it

Work is planned in one place, happens in another, and gets reported in a third. Progress lives in phone calls, photos in someone’s camera roll, variations in a text message, and certificates in an email attachment. Every individual record exists. None of them are in the same place, and none of them are in the same place as the job.

So the owner ends up as the integration layer. Anything he does not personally ask about is something he does not know. That works, right up until there is more work than one person can hold in their head — which is exactly the point at which the business is doing well.

What a system like this would cover

The unit is the job, and everything attaches to it:

• Jobs and stages — what is booked, what is underway, what is finished, and who is on it.
• Actions and updates recorded against the job, so progress is a property of the work rather than something reconstructed from a chat thread.
• Documents, photos and sign-offs kept with the job they belong to, which matters most on the day someone asks about work done eight months ago.
• Materials — what is needed, what is ordered, what is on site.
• Risks and blockers raised explicitly, so “waiting on something” is a visible state rather than silence.
• Handovers, with what is outstanding at each one made obvious rather than assumed.

On top of that, one view answering the only question the owner actually asked: what needs me, and what does not?

What it would not do

It would not make site decisions, and it would not replace the judgement of people who know the work. It would not, on its own, make anyone compliant with anything — it can hold certificates and sign-offs, but the obligations stay with the business. And it would not be worth building all at once; the first slice would be one crew and one job type, in use within weeks, before anything else was added.

Status: concept

Nothing here has been built. This is a write-up of a problem I think is worth solving, published so the shape of it is clear — not a claim of delivered work, and not a description of a real client’s system.

If you run something similar and this reads like your week, that is a good starting point for a conversation.