Why Doesn’t Business Improvement Work in Construction?

Short answer

Business improvement in construction fails at implementation, not strategy. It is bought as software, measured on activity rather than outcomes, and never reaches the culture on site. In one business I worked in, the time for a problem to reach the leadership team was 60 days. Measurement, escalation and a weekly discipline got it to 3 days in 18 months.

Last reviewed: November 2026

There is no shortage of ambition in construction. Directors talk about going digital, continuous improvement and cultural change. They attend conferences, buy software and run workshops. Then they go back to their desks, and the business carries on exactly as it was.

I’ve watched this happen more times than I can count, not because the people involved are incompetent, but because the approach is wrong. Business improvement fails at implementation. And it fails there because the people designing the strategy have rarely stood on a site at 7am explaining to a trade partner why the programme has changed.

“The gap between boardroom strategy and site reality is where your margin disappears.”

Why doesn’t new software improve a construction business?

It starts with a demo. A director sees a planning tool, a supply chain platform or a project controls dashboard. The vendor promises visibility, reduced risk and better decisions. The board approves it.

Six months later, three people in the office use it. The site teams have gone back to WhatsApp and spreadsheets. The data is incomplete, the dashboards mislead, and the vendor is selling the upgrade.

It’s not just construction. In BCG’s survey of 825 executives, only 30% of digital change programmes met or exceeded their target value and led to lasting change (BCG, 2020). You’ll often hear that “70% of change programmes fail”; that figure has never been backed by solid evidence (Hughes, Journal of Change Management, 2011), so I don’t use it.

This isn’t a technology problem. It’s a people problem dressed up as one. I’ve trialled dozens of construction technology products over the past decade. The ones that worked all had one thing in common: they made someone’s job easier on day one. Not after a three-month onboarding. Day one. The ones that failed were selected by people who would never have to use them.

“If you wouldn’t ask your best building manager to spend an extra hour a day on it, don’t buy it.”

Selection is the easy part. Any competent person can score software against a requirement. Very few can predict whether a site team under pressure will actually use it. More on that in the 2026 productivity playbook.

Why do construction KPIs mislead?

Most construction businesses have KPIs: programme, cost, quality, safety. They’re reviewed monthly, reported to the board and used mainly to manage the conversation with the customer rather than to improve delivery.

That isn’t measurement. That’s paperwork. Real measurement tells you something you didn’t already know, early enough to do something about it.

When I joined my previous company, the average time from a problem appearing on a project to the senior leadership team knowing about it was 60 days. By then a recoverable delay has become a contractual dispute, a quality issue has been built over and a supply chain relationship has broken down.

Within 18 months we had that cycle down to 3 days. Not through any dramatic intervention, but through measurement, escalation protocols and a weekly discipline that had everyone from site to board looking at the same objective data. Extension of time claims that used to take months started being agreed as a matter of course, because the data was clear and the customer knew straight away.

Nothing changed but what we looked at.

What is the cultural blind spot?

You can have perfect data, good software and a rigorous escalation process. If the culture punishes the people who surface problems, none of it works. The data gets managed rather than read. Escalations get managed up rather than solved. The improvement programme produces impressive board reports and no change on the ground.

Culture isn’t a values statement on the wall. It’s what happens when a building manager finds a three-week delay on a Friday afternoon. Does she call her operations manager, or manage it quietly and hope it resolves before the next monthly report?

People only bring bad news quickly if they trust that telling you early is better than not telling you at all. That trust is built or destroyed by how the most senior person in the room reacts when bad news arrives.

“If your people are protecting you from the information, that is a leadership problem. Not a people problem.”

Cultures can change, and not over years: over months, when the leadership team commits to different behaviours and holds them under pressure. It’s a series of small moments where someone tells the truth and the response is “thank you, how can I help?” rather than “why didn’t you tell me sooner?” More in the 5 culture problems hiding in plain sight.

Why can’t most improvement programmes fix it?

Because they are usually designed from one of two directions, and both miss the site.

Process-led programmes are good at governance, frameworks and board-level strategy. They produce a compelling document. Then the organisation slowly reverts, because the design didn’t account for how work actually gets done on site.

Technology-led programmes select the right platform, run the implementation and train the users. What they can’t see is why the planning manager’s relationship with the site teams is quietly undermining adoption, or why the head of supply chain doesn’t trust the data because of something that happened on a job eighteen months ago.

The missing ingredient is the same in both: operational experience in construction, combined with the discipline to turn it into change that lasts.

What does business improvement that works look like?

The work that has lasted for me has followed the same pattern, whether it was a project recovery, a supply chain reset or a cultural one.

  1. Measure first. Not what people believe is happening or what the board report says, but what the data shows. Go to the source: site teams, trade partners, project managers.
  2. Pick two or three things. Not forty. Construction is complex enough without a complex improvement plan.
  3. Design with the people who’ll use it. If they didn’t help build it, they won’t run it.
  4. Measure continuously, every week. Measure adoption before results: the ability arrives long before the habit.
“Investigate your hunches with data. And don’t look for data that supports your position. Look for data that doesn’t. Then reconsider your position.”

The hardest part is rarely the analysis. It’s the conversations with the people in the middle, the operations managers and heads of department, who have kept the business running with too little support and are understandably sceptical of another initiative from above. Those conversations need credibility, and credibility comes from having done the job. The engine underneath is a daily feedback loop: the Live Loop.

Case study: what happened when we measured the M&E supply chain?

When I joined a previous business, it was using 45 M&E partners across 92 projects. About a fifth of them failed. M&E was one of the largest shares of project cost, and nobody had put the numbers on a page.

Once we did, the answer was obvious: a much smaller pool of partners, performance frameworks built on objective data tied to project outcomes, and procurement on best value rather than lowest cost. We also gave partners their own performance data every week: honest and specific.

Within twelve months the business was profitable. M&E insolvency fell to zero while I was there.

None of it was complicated. The difficulty was in the room: procurement teams attached to existing relationships, partners not used to being held to data, and the shift from adversarial to collaborative. More on how that works in how Tier 2 contractors win with supply chain frameworks.

Why it fails, and what works

Software
Why it failsChosen by people who won’t use it; measured on go-live
What worksMakes a site job easier on day one; measured on use
Measurement
Why it failsMonthly KPIs that manage the customer conversation
What worksWeekly data that surfaces problems while they’re cheap
Culture
Why it failsBad news punished, so it stops arriving
What worksLeaders who thank the messenger and help

Software, measurement and culture: the failure pattern and the fix for each.

What does this mean for your business?

If you’ve invested in technology that hasn’t delivered, if your improvement work produces activity rather than outcomes, or if the gap between what the board believes and what’s happening on site feels wider than it should, you’re not unusual. It’s the default state of most construction businesses above a certain size. The question is whether you’re prepared to look clearly enough to act.

“Make a plan. Get buy-in. Issue it formally. Check in continuously. Measure everything. That’s it. That’s the whole strategy.”

That’s the work we do, alongside your teams, until the improvement is embedded and measurable: business improvement for contractors. If you want to know what it’s worth first, see what culture is worth.

The bottom line: business improvement in construction doesn’t fail for lack of strategy. It fails between the boardroom and the site, and that’s where it has to be fixed.

FAQs

Why do business improvement programmes fail in construction?

They are bought as software or workshops, measured on activity rather than outcomes, and never change how problems are surfaced on site.

What is business improvement in construction?

Changing how a contractor plans, buys and delivers so that results improve and stay improved: measured, owned by the people doing the work, and reviewed every week.

How long does business improvement take to show results?

Early signals come within weeks, once the right data is in front of the right people. Lasting change takes months; in one business the time for problems to reach leadership fell from 60 days to 3 in 18 months.

Do construction companies need a business improvement consultant?

Not always. They need someone who can measure honestly, pick the few things that matter and stay until the habit sticks. That can be internal, if the person has the credibility and the time.

Sources

Andy Pritchard MCIOB, founder of Constructing Culture

Andy Pritchard MCIOB

CIOB Gold Medal, Construction Manager of the Year 2019. Founder of Constructing Culture and author of Going for Gold. About Andy →