It is widely acknowledged that there is significant potential to improve the performance of construction and capital projects.
There is no shortage of advice on what to do, with recommendations being made by bodies ranging from the World Economic Forum, to global management consultancies, legal firms, governments, and professional bodies. But I think they are all missing the core issues.
Some of the common themes that emerge from these recommendations include:
This paper adds to the already crowded field of people making recommendations about what to do.
But it doesn’t repeat the same messages, because I feel something is missing. And I disagree that acting on the five items above will by themselves make much difference.
The Missing Links is a series of three detailed articles that discuss the root causes of poor performance on capital and construction projects, explains why local interventions are insufficient to have any major impact, and recommends what can be done to improve project results for both client and supply chain.
The articles will appear here as posts on my blog.
If you would prefer PDF versions emailed to you, please go to this page and enter your email address.
My main concern is that it can’t be as simple as ‘implement this list of X things”.
Most of the recommendations currently being made are not new. I used a range of digital tools on projects 20 years ago. On one project we had a 3D design model, we held virtual meetings where we shared screens between people on different sides of the Atlantic Ocean, and we used photogrammetry to digitally survey the existing plant. The project suppliers were engaged using a collaborative form of contract that aligned the commercial interests of the client and their supply chain (a 3-party project alliance).
Even further back in history, the Empire State Building in New York made extensive use of off-site fabrication and standardisation, along with many other methods that today we call “modern methods of construction”. The general contractor, Starrett Bros. & Eken, was selected only two weeks after Shreve Lamb and Harmon were appointed as architects. The selection process took a few weeks, and there was no fully detailed design package against which to bid.
And look at the performance statistics of that project, compared to a more recent tall tower project in New York – No.1 World Trade Centre.
These two examples demonstrate that most of the main recommendations for how to make the industry better, have been available for many years, and most project teams will have someone who has used them, or at least has heard about them. The ideas are already proven, but for some reason, they are still not common practice.
Surely profit-maximising companies would want to do things that improved their profitability, wouldn’t they?
The wouldn’t deliberately chose to be inefficient and wasteful, so there must be some powerful reasons why they can’t exploit existing and new innovations. It can’t just be lack of awareness.
This paper explores what is really holding the industry back today, if it isn’t simply the low levels of use of various value-enhancing methods and technologies.
This paper is the first in a three-part series.
So there you are, full disclosure. If you are not interested in challenging some widely held beliefs and “Deep Truths” , you may not like these articles!
A graph similar to this one is commonly used to highlight the problem of the construction sector.
The lines represent labour productivity. This one is from “Improving Infrastructure Performance”, published by the UK Infrastructure and Projects Authority, in December 2017.
The message is clear: Construction has a productivity problem, and manufacturing is doing much better. If we could ‘close the gap’ we would release billions. McKinsey, the management consultancy, claims that closing this gap will release some £15 billion of value across UK construction projects. And globally they say this gap is worth $1.6 trillion.
But what if we look at some different numbers? The following graph uses data from the same source as the one above – the UK’s Office of National Statistics (ONS).
This gives are very different picture.
From 1997-2009, as productivity in manufacturing rose dramatically, profitability fell, with a widening gap relative to the general economy. As productivity growth slowed down from 2009, profitability began to climb, and now manufacturing is eventually back in line with the general economy.
So it seems to be manufacturing that has had an issue over the past 20 years, not construction!
To understand these conflicting conclusions, we need to understand what is meant by ‘productivity’, and what exactly it is measuring. As a generic measure, productivity is fine – it measures the ratio of outputs to inputs. The problem arises because the first graph looks at only one of the system’s inputs – labour, whereas the second graph uses a more holistic measure – overall company profitability.
And if you want to see how well a system is doing you have to use a holistic performance measure.
It is wrong to assume that improving labour productivity in isolation will improve the overall project system’s performance. It might, then again, it might not, as the following example demonstrates.
Here are the simplified accounts of a construction company. I’ve defined productivity as profit/total people cost (= 5÷(20+5) = 5/25 = 20%).
To improve productivity this company decides to outsource work done by their own labour, and to use more prefabricated items. After these changes:
Productivity is up to 25% as labour has been turned into purchased materials and subcontracts, but profit has gone down, due to the increase in staff required to administer and manage the additional contracts.
I would take profit over productivity any day!
Trying to improve the overall company by acting on just one part of the system carries huge risk of unintended negative consequences. It lies at the root of some spectacular mistakes that have cost $ billions.
Each of these cases used approaches that on the surface seemed logical. They did not deliver the expected bottom-line benefits because the system-wide impacts were not properly considered. Even worse, people who raised concerns were seen as “luddites” and “change resistors”.
And the idea behind Boeing’s 787 strategy, and which has proven so costly, has lots in common with the construction tactic to increase offsite preassembly.
Construction is at risk of making the same mistakes if it thinks that investing in fixed assets rather than labour, or keeping staff utilisation high, are simple ways to quickly and sustainably improve overall project performance.
Although labour productivity by itself is not a system-wide performance measure, which means there are risks in addressing it in isolation, that doesn’t mean that it can’t be improved.
There is definitely a lot of waste in the processes used on construction projects today. But the answer isn’t necessarily to automate a few jobs, use machines rather than people, and outsource more.
Before we know what to change, we need a much better understanding of what is causing productivity to be wasted today. I think there are much bigger sources of inefficiency and waste than humans doing work that could be automated.
But before we look in detail at what drives inefficiency on construction projects, I would like to look at another industry myth.
Another common assumption is that the low levels of profitability of major construction companies explains the lack of investment in skills and innovation.
That profits are low in construction, is a widely held belief. For example, this is taken from the website of the UK’s Construction Leadership Council:
"… results in low profit margins across the industry. This means in turn that the construction industry is unable to sustain the investment in skills, technology and innovation needed to deliver the step change in productivity necessary…”
But I don’t buy it, for two main reasons.
Firstly, if construction was making low returns, surely that should increase innovation, as companies searched out new ways to make profit?
There are plenty of well-known innovative ideas that are not widely used, that could easily be used on the next project, or piloted on several projects, with very low investment, and a high probability of making a return on a single project.
My second reason is that the industry is, actually, quite profitable.
Everything I have read about low profit margins is based on the assumption that companies’ profits should be more than a few percent of their sales turnover. This measure of profit is known as “Return-on-Sales” (ROS = Profit ÷ Sales Turnover).
However, investors and governments do not use ROS as a key measure. They compare the profit to the amount of money required to set up and run the company, using “Return on Capital Employed” (ROCE = Profit ÷ Capital Employed).
ROCE is the measure of profitability used by the UK’s Office of National Statistics, and is the basis of the graph shown on the previous page.
When I looked at the performance of a sample of construction companies based on their ROCE, I found a relatively healthy industry.
I took 32 sets of results from 17 companies in the UK and Australia over the past few years, and added the ROCE data to the graph that I used earlier (the red dots). Here is the result:
As you can see I had to adjust the graph axes to display the ‘outliers’. 60% of these data points were higher than the UK all-industry average. 75% bettered the UK manufacturing industry average.
And most of the construction ‘underperformers’ had extenuating circumstances like one or two disastrous projects, or they had inflated levels of capital employed after acquiring businesses and incurring debt. The underlying businesses were in good shape,
Here is the data for three main contractors/project managers, showing their average values over the latest 2 years, along with a few comparators from other sectors.
Note how project manager Mace, is more profitable than Apple!
Sainsbury’s and Walmart’s results are typical for the supermarket sector. Compared to construction, they earn a similar ROS, and a lower ROCE.
But despite this ‘low’ profitability, I am not aware that the supermarket industry has a problem with skills, innovation or exploiting digital technology. This suggests that low profitability by itself can’t be the reason for construction’s perceived low investment in skills and innovation.
Construction's hidden profitability. Claims of 'low margins' can be misleading - at least one UK project manager is more profitable than Apple.
As an aside, retail also has low productivity levels.
Like all sectors, supermarkets have their problems and challenges, but the industry leaders are focusing on getting better at distribution and retail, rather than trying to become manufacturers!
And that is what I think the construction industry should do, and it is where the biggest improvement potential lies. Get better at managing projects.
This is also one of the least talked about aspects of the construction industry. The issue of project procurement is well recognised, but I sense a general reticence to challenge the prevailing methods. And I have seen almost no discussion of the underlying approaches to planning and project control.
I find this strange. Projects running over budget and late is a significant and longstanding problem for the industry. There is also an issue about value and affordability, both of which can be improved by delivering projects at lower cost and in shorter durations.
But few observers seem to be critically analysing how time and cost is managed today.
But I will talk about it: I think the way we do projects today is the problem. In particular, how we procure the team, and how we plan and control the project.
And it can’t be fixed by adding more stuff, like for example BIM or Lean Construction.
I heard a presentation recently from a senior industry figure, saying something like "If only we could do the basics right, our projects would perform so much better”.
This is at the same time, right, and wrong.
The fundamental processes of project management, and the procurement that establishes the project team, are vital to the success of capital and construction projects, and yes, we need to get them right.
But getting the basics right, is not much use if the processes that we learned early in our project management careers, are not the best way to deliver fast and low-cost projects.
I will illustrate what I mean, using an example from the sport of high jumping. When we start jumping at school, we use the ’scissors’ technique. It’s easy, intuitive, and anyone can have a go because we land on our feet.
However, if someone wants to progress to the top of the sport, they must learn a totally different technique. The last World Record that used the scissors technique was broken in 1924. Since then three different methods have been dominant, with the Fosbury Flop technique being used for every World Record since 1980.
The Fosbury Flop is very different from the more intuitive scissors. You take off with your back to the bar, you look to the sky, and you land on your neck!
The scissors technique isn’t ‘wrong’, and there are many situations where it is better than the Fosbury Flop. If you goal is to get over a low fence in a concreted car park, then the scissors is great. But if the goal is to jump as high as possible, you need to learn the Fosbury Flop.
To perform at the top level, you need to use the best technique.
And it is exactly the same with managing construction projects.
If we want to achieve the lowest cost, best performance, and fastest project possible, we need to use different methods from those that help us to get through our personal To-Do lists. It isn’t that the methods we use today are ‘wrong’, it is just that they are not good enough to take us where we want to go.
This idea is encapsulated by this quotation from Toyota CEO, Katsuaki Watanabe.
“Brilliant process management is our strategy. We get brilliant results from average people managing brilliant processes. We observe that our competitors often get average (or worse) results from brilliant people managing broken processes.”
Today, most construction projects are more like Toyota’s competitors: They need brilliant people to get average (or worse) results, because we make them use broken processes.
In Part 2 of this series of articles, I will use my project management maturity model to describe the characteristics of today’s dominant methodology for managing construction projects, and the different methods that will help us to achieve the next level of performance.
But first I want to discuss my final myth.
To get the right kind of collaboration between clients and their supply chain, they don’t have to work together over an extended period of time.
Not because this is not a good idea. It is, and long-term relationships can work very well, as for example has been shown in the UK water industry on the @one Alliance between Anglian Water and six supply partners.
But you don’t have to have a long-term relationship to make a step change improvement on a construction project.
You can get great results on a single project. Sure, if the relationship continues beyond this, you can get further benefits from continuous improvement and the learning-curve effect. But most of the improvement potential can be achieved on a single project.
Even a small capital project is going to take over a year. This is plenty of time to develop good working relationships and to develop and deliver a creative project. If the time is used wisely.
Long-term agreements are most suited to organisations who have an ongoing capital programme, especially those with a large number of smaller projects. Or, maybe as an expedient if you are forced to follow a bureaucratic procurement process! They can demand more management effort to set up and run, and carry more risk, than working with each project independently.
If the core problem isn’t labour productivity, the lack of profits in the supply chain, a short-term focus, the wrong ‘manufacturing’ approach, or a lack of awareness about the tools to use, what is it?
These items are certainly symptoms that there are problems in the industry, but they are not the root causes.
What the industry needs is a generic improvement plan, that will deliver improvement whatever the characteristics of the project and programme, whatever the size of the project, and that can scale from a single project to longer-term frameworks.
In order to design and test such a plan, we need a model that explains why today’s results are poor, and why after we make some changes, performance will improve significantly and sustainably. We need to explore causation more than correlation.
Then, once we have identified the root causes, we then need to understand what core beliefs and assumptions make it so difficult to overcome the root causes in isolation. These assumptions are where we start to look for a sustainable solution.
I have been working on a causal model to explain the issues seen across industry today. I have linked the various problems we see today, to some root causes. Although not complete, I think the analysis is getting close enough to share – hence this series of articles. And my conclusion might be surprising to many.
The flawed assumption, or ‘Deep Truth’, that seems to underpin most of the problems experienced today on projects, is that uncertainty is bad, and that you get the best project outcomes by managing each constituent part of a project using firm commitments.
This drives most of the issues we see today, and until this changes, the industry is unlikely to change.
For example, needing a firm commitment drives us to prefer fixed-price contracts, and to add targets and Guaranteed Maximum Prices (GMPs) to reimbursable contracts.
Fixed price and GMP contracts in turn drive conservative bidding as premiums against uncertainty are built in. Because of the high risks involved, bidding and selection takes quite a lot of time and effort, reducing productivity .
Main contractors try and mitigate their contractual risk by offloading much of the responsibility to companies further down the supply chain. This adds additional administrative resources, without delivering any more value, further reducing productivity.
Fixed prices also drive the late involvement of many of the contractors and suppliers on the project. Since they are the experts in their domain, this often means that the design and planning work done earlier on the project was sub-optimal (reducing productivity).
Late involvement also makes the planning of work and resources difficult for companies. This makes it risky to invest in skills and innovation, yet again reducing productivity.
A visible client pipeline of projects is not much help, because what an individual contractor needs to know is which projects they will be working on. To know there is a pipeline of £20 billion worth of roads projects in region X between 2018 and 2023, is kind of interesting, but an individual company would be taking a major risk investing in people, skills, and technology before they secured any work.
And today, once a contractor is selected on a project, the pressure is on to complete – it’s “Go! Go! Go!”. Leaving no time for creativity, considering options, and to change to a better design.
And the same philosophy is passed down the project supply chain, with each tier adding to the delay, cost, and inefficiency.
These are just a few examples from the analysis that links one or two fundamental root causes to the dozens of issues that project teams see every day.
We all know that projects, tasks, plans and budgets all carry a high level of uncertainty. But most of today’s methods tend to ignore it, and look for impossible certainty of performance on each individual task and work package. This in turn drives a range of consequential behaviours that explain much of what we see on projects today, including low productivity, long durations, difficulties in using project-wide systems and technologies, high costs, client dissatisfaction, and high levels of stress.
To deliver highly reliable, and better value projects, we need to design a system that accepts the inherent uncertainty in projects, and uses methods to procure, plan and control the project that embrace the uncertainty. It needs to provide clear measures of progress, priorities and risks, so that day-day decisions are made quickly and easily at the lowest possible level.
The elements of such a system exist, in fact some have already been identified in some of the existing improvement recommendations, though they are not given the importance I believe they warrant.
In looking at how to better manage this uncertainty, I will be turning to the relatively new science of systems theory. Projects and project organisations are examples of what are called “Complex Adaptive Systems”. In the next article I will look at how systems theory helps is to identify the kind of changes we need to make.
The good news is that complex systems can be controlled and managed by focusing on a small number of key performance levers.
The bad news, as I mentioned above, is that many of the practices we hold dear, don’t help, and some make things worse.
Ian is a consultant who helps clients to improve the performance of their capital projects and programmes. Before becoming an independent consultant, his experience included 15 years working as a project manager on capex projects, 10 years in procurement, including being chief procurement officer for a large construction company, and 10 years management consulting with niche consultancies in supply chain and procurement.