38thMove

38thMove · Workflow

Construction progress & reality capture.

Using cameras, drones and AI to record what has actually been built, compare it against the plan, and catch problems while they are still cheap to fix. This is one of the better-evidenced workflows in the library: named firms are using it in live operations, and the gains they report are large enough to be worth testing.

Talk it through → See all 14 Workflows →

Start here

The three questions worth asking.

Question one

Yes

Can AI actually do this work?

Eight public cases sit behind this workflow, across contractors, developers and industrial clients in the UK, Europe, the United States and Japan.

Question two

5 of 8

Has anyone really done it?

Five of the eight cases are in operational or scaled use rather than trials, including NCC, Mace, Vinci, Hensel Phelps and Intel.

Question three

Six figures

What benefit do they report?

The largest single claim is $342,000 of labour on one airport project. Others report saved hours or reduced rework cost. They measure different things, so each is named and attributed below.

The evidence

What firms actually report.

These are the headline figures from the public sources, shown with who said them. They are not audited results, and they are not a forecast for your projects. They show that the work is in live use across several named firms and producing results worth measuring.

Hensel Phelps · Track3D

$342k

Labour saved on one airport project.

Reported alongside 2,964 hours and three major reworks avoided at San Francisco International. The figures come from the technology supplier, not an audit.

NCC · Buildots

70%

Less time on manual progress reporting.

Reported with a claim that 2.3 times more tasks were completed on time. Supplier-published.

Mace · Buildots

4,200

Work hours saved or identified.

Attributed to automated progress reports. Supplier-published, and the project is not named.

Vinci Construction UK · OpenSpace

6,000

Documentation hours saved a year.

A supplier estimate, across 45 live Vinci projects. The breadth of use is the more reliable signal here.

Kajima · drone and AI

75%

Cut from one materials-tracking cycle.

One task on one named project, reduced from about two hours to 30 minutes. Confirmed by the contractor rather than a supplier.

Intel · Buildots

4.3%

Reduction in rework cost per plant.

The only figure here tied directly to construction cost rather than time. Supplier-published.

Seven of the eight cases rely on figures published by the technology supplier. That does not make them wrong, but it does mean no independent party has checked the calculation, the baseline or the cost of implementation.

The honest part

There is no single number, and anyone offering you one is guessing.

The figures above were produced on different projects, with different baselines, measuring different things. A saving on an airport fit-out tells you very little about a housing scheme. Almost none of them disclose what the technology cost to put in, so a reported saving is rarely a net benefit.

So the evidence should be used for what it supports: that this work can be done with AI, that serious firms are using it in live operations, and that the reported improvements are large enough to justify testing. What it cannot do is tell you what you will get. Only your own project can do that.

What to do about it

How to find your own number.

The evidence gives you enough confidence to start. A short, well-designed test gives you the number you can actually take to a board.

  1. Pick one project and one thing you want to improve, such as the hours spent producing progress reports.
  2. Measure that one thing properly before you change anything. Without a baseline you will never prove the gain.
  3. Run a small, bounded test with a clear owner and an agreed end date.
  4. Measure the same thing again, and include the time your people spend checking and correcting the output.
  5. Scale only if the result clears the threshold you set at the start, and count the cost of the tool in that judgement.

None of that is complicated. Doing it properly is a discipline rather than an instinct, though, and it is where most organisations quietly lose the benefit.

Talk it through →

Where I come in

You run it. I help you decide whether it is worth running.

This page is the public read of the evidence. What I do is advise the senior people making these calls. I will not run your test or sit in your delivery team — you have people for that, and they will do it better than an outsider. What I bring is an independent view before you commit, and someone to argue with while you are doing it.

The conversations usually cover:

  • Whether this is the right place to spend your attention, or whether something else deserves it more.
  • What the evidence genuinely supports, and the claims a supplier will make that it does not.
  • What to measure, and the baselines that will hold up when your board challenges the result.
  • What you are committing to with a supplier, and where the lock-in sits.
  • Why this kind of project usually stalls after the pilot, and what to watch for.
Talk it through → Explore the evidence map →