Another construction management system will make your life worse

John is an Operations Director running 20 live jobs.

Monday morning. Management meeting. Someone asks what’s happening with Riverside.

The PM says the job is at 16% margin. Finance says 11%. Last Friday’s report says 14%.

So six fairly expensive people spend the next 20 minutes working out which number is real.

Eventually they find it: Finance has costs the PM hasn’t entered yet. Fine. 11% it is. Now they can finally discuss why the job is losing money.

Neither system is necessarily wrong. They just don’t have the same information.

It’s Sunday evening. Manchester was supposed to finish Monday, and that crew was going to Leeds on Tuesday. Then John’s phone buzzes. Manchester has slipped to Wednesday.

John calls Steve. Steve thinks Mark has two people available. Mark says they’re on Birmingham. Maybe Dave’s team? No, Bristol until Thursday.

Eventually John rearranges everything. Then Leeds moves to Wednesday.

Excellent. Start again.

The programme knows Manchester has slipped. The resource plan knows who is supposed to be in Leeds. Someone knows where each crew actually is.

But there is no single place that knows all three. So John becomes the integration.

Nobody decides: “I think what this £100m business needs is a spreadsheet with 47 tabs that only Sarah understands.” It happens gradually.

The project system doesn’t have the finance data, so someone makes a spreadsheet. Then management wants labour. Committed costs. Variations. A view across all projects. Five years later there’s a file called:

MASTER PROJECT TRACKER - FINAL v7 USE THIS ONE.xlsx

Everyone laughs about it; nobody dares delete it. It’s the closest thing the company has to a complete view of the business.

That’s the important bit. You don’t necessarily need another ERP, a long migration project or months of retraining people. Finance can keep its system. Projects can keep theirs. Site teams can keep their apps. The missing piece is often simply the connection between them.

A few years ago, building those connections could be a serious IT project. Today, APIs, cloud infrastructure, and modern integration tools have made it considerably easier. And you don’t need to connect the entire company at once.

Take the monthly management report. Why does it take four days?

Finance exports data. Projects exports theirs. Someone combines it. PMs send updates. Someone finds an old forecast. The report gets rebuilt. Now ask a much simpler question:

Which parts of this can happen automatically?

Take the relevant data sources, connect them, and build the smallest useful version. At Tellum, for a narrow use case, we can often get to a working prototype with real company data in a matter of days. Not a finished enterprise platform. Just enough to put it in front of John and ask: “Would this actually help you on Monday morning?”

Maybe John can see actual cost, committed cost, and forecast for every live project without asking Finance. Maybe Ops can see what happens to three other jobs when a crew moves. Maybe Finance doesn’t have to wait for twelve spreadsheets to close the month.

No grand “digital transformation”. Something that took three hours takes five minutes.

Something you found out three weeks later, you find out today. That’s enough.

And sometimes the spreadsheet really is fine. If someone spends twenty minutes updating it every Friday and everyone trusts it, leave it alone.

But if senior people spend hours moving information between systems, meetings turn into arguments about whose number is right, or projects start losing money weeks before management sees it, it’s worth looking at why.