What do building errors cost in UK construction?

Long read, August 2026. Every figure sourced inline; sources listed in full at the end.

Avoidable error costs UK construction between £10 billion and £25 billion a year, which is 10 to 25 percent of project cost, according to research by the Get It Right Initiative (GIRI, A Call to Action, 2016). Even the bottom of that range is roughly double the total annual profit of the entire industry, and in 2026 a landmark training pilot showed that around half of it can be avoided, with three quarters of the avoidable cost traced to one cause: inadequate planning.

That is the short answer. The rest of this page is the long one: where the numbers come from, why the cost is invisible on most sites, what actually causes it, and what the contractors who have started fixing it did.

How much do building errors cost?

The most rigorous UK figures come from the Get It Right Initiative, the cross-industry body founded to tackle avoidable error, whose research drew on seventeen major construction organisations across clients, consultants, contractors and their supply chains (GIRI, A Call to Action, 2016, free at getitright.uk.com/reports):

• Around £5 billion a year in direct, recorded cost of avoidable error across the UK sector.
• £10 billion to £25 billion a year once indirect costs, latent defects and unrecorded process waste are included. That is 10 to 25 percent of project cost.

In 2026 GIRI put the top of that range, £25 billion, at 21 percent of the overall value of UK construction projects (reported in PBC Today, May 2026). The US numbers rhyme: the Construction Industry Institute’s field studies put direct field rework alone at around 5 percent of project cost on average, before design rework and indirect costs are added.

What counts as the cost of an error?

The direct cost, the item somebody actually books as rework, is the small part. The full bill has four layers:

• Direct rework: the labour and materials to take out wrong work and put in right work.
• Indirect cost: the site set-up, supervision and programme time consumed while it happens. A defect fixed in week 60 is not just a defect; it is a reason week 60 exists.
• Latent defects: errors built in, covered up and discovered after handover, when they cost the most to reach and the most to repair.
• Unrecorded process waste: the queries, re-sequencing, waiting and duplicated effort that error creates, which no cost report ever labels as error.

GIRI’s 10 to 25 percent range is the sum of all four. The 5 percent figure most people have heard is only the first layer.

How does the cost of error compare with profit?

The Construction Index Top 100 (2025) puts the average pre-tax margin of the UK’s biggest contractors at 2.4 percent. Forty three of the hundred made less than 2 percent and sixteen lost money. Set the two published ranges side by side on a single mid-size job:

Across the sector, even the conservative end of GIRI’s range means error consumes roughly twice the industry’s entire annual profit, before a single dispute is priced. GIRI’s own 2016 estimate put the annual spend on error as high as seven times annual profit. Margins are not thin because the work is cheap. They are thin because so much of the work is done twice.

Why don’t contractors see the cost?

Because almost none of it appears under the word “error”. It hides as:

• Extended preliminaries that stay on the job week after week while defects are closed out.
• Design queries answered on site, at speed, at the wrong point in the sequence.
• Snag lists still growing while the first fix is unfinished.
• Latent defects surfacing a year after handover, out of retention or goodwill.

Because it never appears as one line, it is nobody’s number. And a cost that is nobody’s number is nobody’s job to remove.

What causes most building errors?

Planning. Not workmanship, not weather, not bad luck. When GIRI and CITB tracked avoidable error across 25 live UK projects worth a combined £942.5m, inadequate planning was, in GIRI’s words, by far the biggest cause: £71.2m of the £92.6m of error headed off, three quarters of the total by value, traced back to the plan (Construction Management, “Inadequate planning causes three-quarters of errors by value”, April 2026).

That finding matches what the industry has been saying about itself for a decade. In a 2022 industry survey, 31 percent named poor original planning and unrealistic scheduling as the number one reason projects run late. It was the top answer in 2016 too.

Can the cost of error be reduced?

Yes, and 2026 produced the first hard evidence at scale. Four major contractors, Kier, BAM Nuttall, VolkerStevin and Taylor Woodrow, put 4,575 people through the GIRI Accredited Training Scheme, a CITB-funded programme, across 25 projects over 26 months. The results (Construction Management; New Civil Engineer, 15 April 2026; PBC Today, May 2026):

• £92.6 million in lost value avoided, close to 10 percent of the value of the schemes being built.
• For every £1 spent on training, an estimated £256 was avoided.
• The predicted cost of error roughly halved on the participating projects.

Training people to get it right first time returned two hundred and fifty to one. There is no other investment on a construction project that comes close.

How do you prevent rework instead of pricing it in?

At the end of a £12m school, we added up what snagging and damage had cost us. It came to more than £50,000. Some of that was vandalism, done by the people building the school.

The next job was twice the size. That gave us a simple decision: take the £50,000, double it and put it in the cost plan, or do something different.

We did something different, in two ways.

The mechanical fix. We looked at where the money had actually gone, and almost all of it was in the finishes. So we bought a lock for every door, around a hundred rooms, for roughly £2,000. Once a room was finished it was locked and it stayed locked. Those locks became hard stop points on the delivery programme, and we added “doors locked” to the room-by-room quality sign-off sheet, so it sat inside the quality sequence rather than being a good intention.

The psychological fix. We filmed the head teacher and the pupils talking about how excited they were to be getting a new school. Every trade watched it during induction week, and we asked them two questions: who are you building this for, and who do you do your best work for?

The result, on a job twice the size, was nothing. Nothing spent on vandalism. Nothing spent on rework. Nothing.

My one regret is that we could not run an A/B test, so I will never know whether the locks or the video did more of the work. That is exactly why I now plan both, every time. Lock the doors before the finishes go in, and make sure the people building it know who they are building it for.

How do you tell if your project is bleeding error cost?

Five signs, any of which means the 10 to 25 percent range is alive on your site:

• Rework is happening but nothing is booked to a rework code, so the cost is invisible by design.
• The snag list is growing faster than it is closing while trades are still in first fix.
• Prelim weeks are being added at the end of the job to close out quality, not to build.
• The same defect appears on more than one floor, which means the error was systemic and nobody stopped the line.
• Handover is planned as an event in the last month rather than a standard set on day one.

If more than one of these is live on your job, the free self-checks on the free tools page will put a score on it in a few minutes. And if the programme has already gone, start with project recovery.

What does good look like?

The fix is not inspection at the end. By the time an inspector finds it, the error has already been built, and the GIRI pilot shows where the cheap fix sits: in the plan, before the mistake exists. Good looks like quality planned the way programme is planned. Progress measured by what is actually finished and signed off, not what is claimed in a meeting. Digital quality sign-offs driving the programme, so it reports the truth whether or not the truth is comfortable. Work independently checked before it disappears behind the next trade. And nothing left hanging: every query, drawing change and defect with an owner and an end date.

That approach, treating the finish as something you plan from the first week rather than clean up in the last one, is the method in part two of Going for Gold: Constructing Project Managers, written from twenty-two years of building jobs. All royalties go to The Lighthouse Charity.

Do it Monday

Open a rework code on your job and book one week of honestly recorded error cost to it. Just one week. The number you see at Friday will tell you more about your margin than the CVR will.

Sources

• Get It Right Initiative: A Call to Action (2016) and Strategy for Change (2015), research across seventeen major construction organisations. Free at getitright.uk.com/reports.
• GIRI/CITB training pilot: GIRI Accredited Training Scheme pilot results, April 2026 (Construction Management, “Inadequate planning causes three-quarters of errors by value”, April 2026; New Civil Engineer, 15 April 2026; PBC Today, May 2026).
• Contractor margins: The Construction Index Top 100 (2025), average pre-tax margin 2.4 percent.
• US comparison: Construction Industry Institute field rework studies, direct field rework averaging about 5 percent of project cost.