Every sector
What does AI actually cost a construction firm, and is it worth it?
The cost structure, the arithmetic and a break-even test you can run on any quote you are handed. Every figure sourced, every assumption stated.
Read articleBlog
There is margin hiding
in your business.
Practical views on where AI can improve work across the built environment, and what businesses need to consider before putting it into use.
The State of the UK Built Environment Survey 2026 closes on 27 November 2026. Take part here.
Frameworks
The longer arguments: on where AI pays back.
Every sector
AI partners for UK construction: an honest comparison.
Construction software with AI features, major consultancies and AI providers, compared for established UK firms. Our own entry is marked.
Read articleEvery sector
How to choose an AI consultant for a construction business.
Seven practical tests for choosing an AI consultant who understands construction workflows, data risk, adoption and commercial payback.
Read articleContractors
Three construction workflows worth reviewing for AI
These are three areas worth considering, not a prescribed starting order for every business. The right priority depends on the commercial effect, the information available and what the team can put into use.
Read articleMain contractors
Subcontractor compliance is a paperwork tax. Here's how AI cuts it.
Document collection, chasing and date checks create repeated administration. Reviewing whether the evidence meets the project's requirements is a different responsibility.
Read articleEvery sector
Why an AI pilot can stall after the first result
An initial result does not establish that a workflow is ready for normal use. Adoption depends on the work around it: training, information quality, ownership, exception handling and support.
Read articleEvery sector
Shadow AI: the data risk inside your business that nobody's talking about.
Your team is already using AI. The question nobody's asking: what are they feeding it, and where is that data going?
Read articleProperty & FM
The certificates, work orders and chasing that never stop.
A system can help organise records, track dates and flag missing information across a portfolio. Whether the evidence establishes compliance still needs to be assessed by the responsible people.
Read articleDevelopers
Every week a deal waits is margin you won't get back.
The commercial effect depends on the decision being improved. Faster preparation of planning, appraisal or handover information may help the team act earlier.
Read articleFounder-run firms
AI in construction: what the best-run companies get right.
The companies using AI well share a pattern, and it isn't budget or software. It's that leadership understood the workflow first.
Read articleEvery sector
The skill gap is the real AI barrier, and how to close it.
Some companies need additional internal capability. Others need specialist support for a defined piece of work, or a combination of the two.
Read articleEvery sector
The two questions every company should ask before buying AI.
Two questions, asked properly, save more wasted AI spend than any vendor demo. They aren't technical. They're business questions.
Read articleEvery sector
AI and your data: a one-page policy any firm can use.
A short staff guide can make approved tools, permitted information and checking responsibilities clear, within the business's existing data, security and contractual controls.
Read articleEvery sector
The first 90 days of putting AI to work
Early progress and a wider programme are not competing choices. The first 90 days should establish useful priorities, develop the team's capability and move the agreed work towards practical use.
Read articleQuantity surveyors
Cost plans without the weekend of spreadsheets.
Approved historical records can support the preparation and comparison of cost information. Rates, assumptions, scope and the resulting advice still need the commercial team's review.
Read articleM&E contractors
M&E runs on paperwork. AI gives the hours back.
Review the repeated preparation around estimating, reporting and project records. The value depends on the workload and the checking still required.
Read articleArchitects
Stop paying senior staff to format drawings.
The potential value depends on the volume of work, how information is prepared and the review required, without removing professional judgement.
Read articleEvery sector
What to ask an AI vendor before you sign.
Seven plain questions that separate an AI tool that earns its place in your business from one that just demos well.
Read articleEvery sector
What repeated administration is costing your business
Repeated administration uses capacity that could be spent elsewhere. Establishing its value starts with the work, the people involved and what the business could do differently if that capacity were released.
Read articleEvery sector
Most AI tools die in the van, not the boardroom.
The decision to buy AI gets made in an office. Whether it survives gets made on site. A short field note on why adoption, not technology, decides whether AI sticks.
Read articleEvery sector
Your nine-day quote has six days of waiting in it.
Ask how long a quote takes and you get a number. Most of it is waiting, not working. A short field note on why speeding up the wrong part of a quote changes nothing.
Read articleEvery sector
“We tried ChatGPT once” is not an AI strategy.
Individual experiments can be useful. A company-wide approach also needs agreed priorities, permitted uses, ownership and a way to assess what is working.
Read articleNothing in this category yet. Try another or browse .
Subscribe
Get sharper about AI every week.
We send regular notes from inside the work: what is moving in AI for UK construction and the built environment, the workflows worth fixing, and worked examples you can act on.