The second Oxford Major Programme Management Conference was held on 24 February 2018 at Saïd Business School, University of Oxford.
Featuring an array of distinguished speakers and a varied agenda, the theme of the conference was: ' Technology and Major Programmes: Mastering scaling’.
Is there a role for big capital projects in a technology-driven world?
Sir David Higgins, Chairman, Gatwick Airport Ltd; Chairman, High Speed Two (HS2) Ltd
Technology has always held the promise of being able to change the world, said Sir David Higgins, asking us to cast our minds back to the early days of video-conferencing. Video conferences were going to transform business meetings and eliminate the need for costly and environmentally unfriendly air travel. And they did make some communication easier, despite the challenges of meeting participants’ talking over each other, pictures freezing, and people suddenly dropping offline. But travel still continued because, as he said, ‘The need for human interaction and to meet face to face to exchange ideas is still vitally important.’
This demand for continuing human connectivity was at the heart of his argument for the new High Speed Two (HS2) network in the UK.
Why develop HS2?
The current mixed-use railway system in the UK is still rooted in its Victorian past; it is one of the oldest, densest, most expensive, and most complicated in the world. There seems to be an obvious need to move into the modern world and build a dedicated consumer railway system.
However, it has been difficult to justify the cost in an age of austerity. Higgins was perpetually having to ‘go back and rebuild the business case,’ he said.
Part of doing that was challenging the myth that the productivity problems in the Midlands and the North were as a result of a lack of skills. He spent a year meeting people in universities, local businesses and councils, and concluded, ‘It is not an issue of skills. Newcastle has some of the most sophisticated and highly trained baristas working in the country. There is no shortage of skilled, highly trained technical students at universities.’ There is also no shortage of industries located in the area: technical businesses in the automotive and airline industries are clustered around a great Y shape stretching down the spine of the country.
It is, he concluded, ‘an issue of connectivity – and not just capacity but also resilience and reliability.’
What’s in it for the regions?
Five years ago, though, cities such as Liverpool and Crewe still did not see what was in HS2 for them. It seemed to have been originated in Westminster, driven by well-meaning and competent engineers, and apparently something of a vanity project, a determination to build something better and faster than anything else. ‘Understandably people were saying: £50 billion just to shave 20 minutes off my train journey? Just give me wifi and a cup of coffee,’ said Higgins.
So he had to go back to the drawing board and revisit the purpose of developing a high-speed network throughout the country. What he found was that it was about more than connectivity: it was about addressing the whole issue of the overheated south of the country – the concentration of economic activity in London with the accompanying hothouse effect on real estate – and the underdeveloped north.
He needed to convince the regional authorities that HS2 was not going to be simply ‘a highly effective drainpipe from north to south’, and that would give them the feeling of ownership that they needed to back the project.
The role of technology
Construction is still a very traditional industry. Higgins claimed that ‘I could go back on a construction site with the technology I used 40 years ago and still be an efficient site engineer.’ He said there was an ‘enormous opportunity’ to use benchmarking and data (‘that exists but is fragmented’) to improve prediction, cost-planning and decision-making.
Higgins concluded, ‘So am I an advocate for big capital infrastructure projects? Of course, but it’s never either/or, never broadband or high-speed rail. It has to be both. … But you have to analyse why you are doing something. What is the purpose of it? How does it fit in the overall economic world? You have to engage with communities to explain this to them so that they have ownership. And as an industry, we have to use modern technology to make decision-making in this project-rich nation better.’
Atif Ansar, Programme Director, MSc in Major Programme Management, Saïd Business School, University of Oxford
Atif Ansar picked up on Higgins’s point that the construction industry is ‘desperately antiquated’. He believes that it is in need of a ‘large-scale technological revolution’.
The good news is that getting big infrastructure investments right does create prosperity. Parts of Dubai went from desert to creative, exciting, dynamic city within 18 years. Singapore was a colonial backwater in the 1970s, with a nineteenth-century port. Now it is the world’s highest GDP-per-capita country. Get the sequence of investments right in major programmes and you can create enormous growth and change.
The bad news is that major programmes have a very poor record of delivery. In every asset class and every country, from new roads to the Olympics, Ansar painted a dreary picture of projects that have overrun on costs and time, and failed to deliver their promised benefits.
What is more, these problems are not confined to construction projects. ‘As much of a tech fan as I am, I have to say that one in five tech major programmes have the potential to bankrupt your company,’ said Ansar.
Major infrastructure projects are simply too slow, he continued. There is a ‘massive temporal mismatch’ between when people need a service and when they can actually get it. Arguably it is ‘almost pointless’ to start projects that take such a long time to deliver. In addition, the difficulties recovering costs are enough to sink both small contractors and large companies. Indeed, Ansar was willing to bet that ‘most construction companies today active in major programmes will not last this century because of insolvencies related to major projects.’
He wanted to move from ‘sad’ major programmes of finite gains and infinite pains to ‘happy’ programmes – on budget, on time, and to benefits every time. And for this to happen, he said, we need a technology revolution.
Ansar directed the audience to look at technology firms such as Amazon, Facebook, Google and Oracle, which are ‘masters of scaling’. He described scaling as moving in four ‘waves’: 1.0 Manual 2.0 Mechanical 3.0 Electrical 4.0 Digital.
Construction, naturally, he placed at between the first and the second wave, largely because its projects tended to be ‘big and bespoke’. This is not scalable because there is too much variety. More ‘repeatability’, more boxes, will enable projects to be both faster and cheaper.
Ansar’s recipe for scaling boiled down to: Modularise, Digitise, Automate.
Watch the opening video.
- Can an existing company transform the construction industry? We tend to assume that change will come from outside, in the form of a well-financed disruptor. But is there a modern construction company that will emulate IBM in the 1950s and hit the pause button, not accepting any more projects until the business has sorted itself out?
- Technology scaling is not easy either. What helps is having common platforms such as Linux. The problem is that the effectiveness of these common platforms is linked to their being given away: ‘Part of the challenge is to persuade VCs to give up greed.’ In construction, the equivalent of common platforms could be common specifications, for example bathroom ‘pods’ that are assembled off-site and delivered in their entirety to housing development projects.
- Why are we still talking about railways, even high-speed railways, in an era of drone taxis? It’s true that the organisation of UK railways is ‘bizarre’ – a hangover from the 19th century – and we are spending vast amount of money just on patching up the system. HS2 is at least 20 years late. But, ‘whatever time doesn’t break, persists’: railways have survived for so long that they will probably continue for centuries.
- It would be possible to deliver a project such as HS2 faster, if you did not spend time relocating badgers or reducing noise – but that is the cost of democracy. Democracy is not something that anyone in the audience would trade, but they could not help looking slightly wistfully at changes in Dubai and Singapore.
- It is worth remembering that what people say is not necessarily what they do or what they want. This is where insights from fields such as behavioural psychology come in. They can help us observe how things really are and predict what people do. Change is possible if we respect the way things really are and do not try to force them to fit our aesthetic.
Panel: Will your digital transformation programmes sink your organisation?
- Professor Bent Flyvbjerg, BT Chair in Major Programme Management, Saïd Business School, University of Oxford
- Shaun McNally CBE, Chief Executive, Legal Aid Agency
- Sherief Elabd, Director Industry Strategy, Oracle
- Marek Kujawa, Head of HR IT Talent and Core Processes, Roche
- Moderator: Alexander Budzier, Fellow in Management Practice, Information Systems, Saïd Business School, University of Oxford
Why are digital transformation programmes so challenging?
A non-scientific poll at the beginning of the session suggested not only that so-called digital transformation projects are ubiquitous, but that they are almost universally troubled.
This is because changing work habits and culture, and how they come together, is much more difficult that you might think – and it’s no easier in the private sector than in the public sector. Technological change is also organisational change. It’s not just a case of bringing people along with the programme: you need also to make sure that the organisation continues to function during the period of transformation and that people will have the right skills to thrive in the new environments. ‘It’s like trying to change the wheels on a bus as it’s hurtling down the outside lane of a motorway.’
What do people get wrong?
Many so-called digital transformation initiatives are really no more than IT projects, because they are not aligned with the strategic plans of the organisations. And ‘Gap-fillers are not the same as digital transformation’.
People don’t think enough about the vision or objective of the technological change. They get excited about technology for its own sake (‘We need robots! We’ve seen some and they look good’) and because they have an urge to be ‘at the forefront of digital transformation’. But they forget that it does not always contribute to strategy.
What should they do to get it right?
Don’t rush to say what the solution is. Time spent planning upfront is never wasted. Spend time at the beginning to do the behavioural insights work, and look at and simplify processes. And then bring everyone with you as you build and test the solution, being prepared to make changes along the way.
Think big. Don’t just aim to increase efficiency but think about how you are attempting to disrupt the business or market. If you don’t have a big vision, if you are not reimagining an outcome, don’t do a data transformation. Think of technology as just a tool – useless unless it does something fantastic.
Keep it short. In digital projects, time will kill you, which is one reason why so many people are adopting Agile. When you are analysing processes always think, how can I make it shorter? At the same time, processes must follow the expectations of the customer, which means that occasionally you will find yourself making compromises. A process may not be the simplest, but it suits the customer. In this situation the key is to make the interface simple and sort out processes in the background.
How do you ‘bring everyone with you’ as you build and test the solution?
The panel suggested that people should be enabled to try things out, to learn new technologies and then join with the company in the transformation.
This could be achieved by creating a ‘playground’ in the company for experimentation on a small scale, making time by outsourcing some operational tasks if need be. The best ideas could be picked up, scrutinised, and – if suitable – presented to decision-makers.
What are the risks of not embracing a digital transformation?
Digitisation is not a choice. Ultimately companies will have to digitise or they will become obsolete.
The problem is that it is also extremely risky. You have to balance the risks with the necessity of increasing efficiency; and also balance ambition and reimagining with being realistic about what and when you can deliver.
The biggest risk is ‘setting ambition too low and achieving it.’
What is the role of leadership?
In a world where technology is no longer a real differentiator, what makes a company successful is leadership that is reliable, predictable, and open to new ideas. Leaders need to be able to make democratic and collective decisions, and have the confidence to empower people in complex environments.
If you are not empowering the people who are executing your digital transformation, the programme will be disaster. On the other hand, one of the biggest mistakes in transformation projects is for the C-suite visibly to lose interest. It is a delicate balance: you have to empower but you cannot back out and leave everything to the project managers.
If you are a leader, you have to be able to work with people who think differently to you. Diversity is not just about getting people round the table – it’s knitting diverse experiences and opinions together and reaching consensus. This, of course, is much easier said than done.
Don’t just listen more than you tell: you also have to observe more. What are people doing and thinking that they are not talking about? Remember that any project needs to be a collective success story. If anyone is thinking of it as an individual success story you will fail.
Hire talent. If you hire talent that does not have precisely the right skills, then train them – for example at Oxford Saïd! Most revolutions happen because people believe it’s the right thing to do. Therefore you need to find individuals in the organisation who passionately believe in the need for transformation and are prepared to argue with everyone from the Chairman down to get their points across.
Watch the video.
John Holland-Kaye, CEO Heathrow Airport “Using technology to improve the experience of both passengers and airline partners”
Watch the video.
David Wilson, Chief Innovation Officer, Bechtel “Applying technology and innovation to supercharge the engineering, procurement, and construction (EPC) industry”
Watch the video.
Jan Walstrom, Senior Vice President, Jacobs “The power of collaboration between an entrepreneurial government and a committed industry”
Watch the video.
Panel: Will software kill tech hardware?
- Dan Madrigal, Data Center Lead, Real Estate Acquisitions and Infrastructure, Facebook
- John Stroud, Principal Vice President and Global Telecommunications Sector Lead, Infrastructure, Bechtel
- Burcin Kaplanoglu, Senior Director, Industry Strategy and Innovation, Oracle Construction and Engineering Global Business Unit
- Gia Schneider, Chief Executive Officer, Natel Entergy In.
- Sean Fishlock, Managing Director, Berkeley Research Group LLC
- Moderator: Leasil Burrow, Associate Fellow, Saïd Business School, University of Oxford
Is software king and hardware just along for the ride?
The panel were unanimous in stating that there is a symbiosis. From automated cars to drones, it is impossible to have the software without the hardware to go with it.
A better question might be: is software development outstripping the pace of hardware? Is the challenge being able to develop and deploy hardware to feed software’s appetite for big data?
For example, there has been a massive explosion in software designed for smartphones. This could not have happened without a very good hardware platform. But now there is a clear need to build out the rest of the telecoms platform, including fibre networks, broadband etc.
Software development is only getting faster. The key is to look ahead to see what software will be needing over the next two to five years and try to devise hardware to improve that. For example, 5G will be coming in two or three years. Phone networks are already designing for that need.
What will be the future need and how do we avoid ripping things up? Some industries start with hardware, but then achieve scale by applying software intelligently.
Where is the cloud?
Facebook is thought of as the ultimate software company, especially as it has added applications such as Whatsapp and Instagram to the Facebook platform.
However, its growth has triggered the opening of multiple data centres: between 2011 and 2015 Facebook opened a new data centre every year. Since then it has been opening a data centre every quarter. ‘If anyone wonders where “The Cloud” is, look at the data centres in Facebook.’ So software is driving a very large amount of hardware development.
Data centres are not located in vacuums. They are developed to make sure that the technical needs of the population are met, but not everyone welcomes the building of a new data centre in their neighbourhood. There is a need for companies to develop partnerships with jurisdictions and to create strong relationships within communities.
Will expansion be limitless?
Facilities that process data and networks that support mobiles form the technological infrastructure that are rapidly becoming as all-pervasive as utilities such as electricity. Developing this utility is the major programme for the next couple of decades: how do we supply it at the volume and velocity that is needed?
But is it all necessary? As software and data expand our current response seems to be continually to expand the hardware to store it, run it, and manipulate it. But not all data is valuable. How do we keep data applications lean?
What are the environmental implications?
It feels as if computing is cheap and that software has no discernible environmental impact. But when you build a data centre, it is a major construction project, using a lot of water, power, and other resources. And when it is running, it needs cooling and continuing power. But that consumption of resources is invisible compared with driving a car or travelling by air.
It is easy to see the short-term environmental benefits of technology but forget to look at the long term. For example, electric cars are more obviously environmentally friendly than petrol or diesel. But, as one panel member asked, what goes into making those batteries? How long will they last and what will happen to them at the end of their life?
What about data?
Typically, according to the panel, six months after a project has finished a dispute crystallises between two parties. They start to look back over the data and are astounded to find how useful it is. This information has existed all along but no one has thought to bring it together. All parts of the project – from planning to operations and climate teams – have operated separately, focused inwardly on day-to-day delivery.
Essentially there is either no data or too much data, but no one looks at it until after the fact when they have to sort out a problem. The promise of future programme management is all about capturing data as early as you can and then using it proactively.
‘Data has to be meaningful to you, and also has to be timely.’
Watch the video.
Panel: Disruptive technologies in the workplace: are artificial intelligence, blockchain, automation, and virtual reality mere buzzwords for major programmes?
- Henrik Schoedts, Project Chief Executive, RegionH
- Gerard Newman, Partner, PwC
- Daniel Stauthamer, Vice President, Director Human Resources Middle East, Africa, and Asia Pacific, Jacobs
- David Porter, Programme Director (MD), Endeavour Programme
- Hobin Kim, Chief Executive Officer, Seevider Inc.
- Mariarosaria Taddeo, Research Fellow, Oxford Internet Institute; Deputy Director, Digital Ethics Lab
- Moderator, Atif Ansar, Programme Director, MSc in Major Programme Management, Saïd Business School, University of Oxford
How can these new technologies help major programmes?
Artificial Intelligence (AI) features are already embedded in many project and back-office applications. They can drive workflows – approvals and payments, for example. They can be used to send alerts when certain conditions have been met, and also to analyse large quantities of data. As the previous panel discussed, shortage of data is not a problem: shortage of time to analyse and understand it, is.
However, it is and will remain important to preserve the human element. We do not want automated decision-making. If we are to have a technological revolution, let us make sure it is a socio-technological revolution.
What about data?
Despite the vast amounts of data being collected, organisations and projects do not seem to be good at using it. When attempting to collect data on large dam projects, for example, the Oxford research team took four years to wrest data from 0.6% of the global population of dams.
While banks share data habitually, because it is in their interests and helps them to avoid fraud, projects and programmes are not good at sharing. There is data about projects, but each collects different data, and each stakeholder gets data presented to them differently. We need to find ways to pool data without undermining commercial and political positions.
A further challenge is that organisations do not necessarily understand the information that they have. Data is being captured but people do not understand, analyse, synthesise and react to it.
What are the ethical considerations?
There is a perpetual worry that new technology will take away jobs. We used to worry about automation; now we worry about AI. But do we actually have data that supports this claim?
It is easy to focus on the negative, but the reality is that there are risks and benefits. What we need to do is mitigate the former and maximise the latter.
This does entail raising questions. For example, is AI contributing to de-skilling? Do we want to de-skill? Lose our ability to read an X-ray or analyse data? What happens if AI goes wrong? On the other hand, we can look at the benefits: AI frees us from jobs we don’t want to do, allowing us to maximise our humanity.
Ethical questions such as these in business are not new, but the new technologies are leading us to express them rather dramatically. In fact, most business people are very pragmatic – interested in making things just a little big cheaper, or a little bit more efficient. Ethics is an important component of decision-making because of the links between trust, reputation, and profit. Break the trust or reputation links and profit does not happen.
Can new technologies make major programmes more investable?
There is currently an under-investment of 25% in infrastructure. For the World Bank and European Investment Bank the answer is to use more private money through Public Private Partnerships (PPPs) of Private Finance Initiatives.
The cost of setting up a PPP project is around 10%. Compare that with setting up a similar project with Private Equity: the costs are between 2 and 3%. If we fill that gap it means more projects. And more jobs!
The reason that this is not happening is down to there being more stakeholders. Aligning the leaders and goals is more difficult in a PPP. New technology tools, such as blockchain, could ease that by giving easier access to data and transparency, enabling organisations to set up joint frameworks and discuss how to work together and approach problems as they arise.
Will we still have the same problems of cost and time overruns?
Existing data suggests that we are still not getting better at managing cost and time overruns in major projects. We can all remember the technical ‘silver bullets’ that have been introduced to programme management in the past. These resulted in incremental improvements but no major change.
AI works when there are clear inputs and outputs. Major programmes are highly complex, subject to uncontrollable external factors. So certainly it is worth a try but no one should expect it to be a universal panacea. We cannot take our eyes off the ball of other problems in major programmes.
Watch the video.
Conclusion: Technology in society: content and the future
James Clarke, Founder and Chief Executive Officer, Clarke Capital Partners
In delivering the final keynote of the day, James Clarke said that enough had been said about the ‘what’ and ‘how’ of major programmes: he wanted to talk about the ‘who’.
He pointed out that all the big businesses in the room had started out as an individual with an idea. But ideas, he said, are ‘a dime a dozen’. As an investor he sees hundreds of them, either pitched formally in the office or just passing across his desk.
The success of an idea depends on a particular kind of individual – ‘intelligent, full of enthusiasm, expertise, gumption and passion – and these are not a dime a dozen’.
Henry Edward Roberts, for example, developed the personal computer – and yet no one has heard of him. Philo T Farnsworth was a television pioneer. He was an inventor, but ‘he didn’t have the gumption or the skills to build a billion-dollar business behind it’. Only Thomas Edison combined the inventive and leadership skills necessary to become a ‘Firstling’ – an inventor of an entirely first-of-its-kind product or service, who then grows the company to at least a billion dollars. As the nineteenth-century inventor of the lightbulb – ‘the very symbol of ideation’ – and leader of the company that became GE, he was perhaps the ‘ultimate firstling’.
Firstlings, however, are the rarest of a rare breed. Slightly more common, but still rare, are ‘Afterlings’ – the people who can take others’ ideas and give them success.
‘Most ideas start with “what if?”,’ said Clarke. ‘Large dams, a tunnel under the sea, putting a man on the moon. All of these were just ideas, a dime a dozen – until they met the right who.’
Clarke used the story of the first sub-four-minute mile to illustrate the power of these people. The runner John Landy tried many times to break the four-minute-mile barrier, but finally admitted: ‘Frankly, I think the four-minute mile is beyond my capabilities. Two seconds may not sound much, but to me it’s like trying to break through a brick wall. Someone may achieve the four-minute mile the world is wanting so desperately, but I don’t think I can.’ That ‘I don’t think I can’ quote was to be Landy’s ‘legacy’, as in 1954 in Oxford, Roger Bannister broke the four-minute barrier with a run of 3:59.4. Less than two months later, however, John Landy ran 3:57.9 – four seconds faster than he’d ever managed before,
‘Bannister had broken down Landy’s wall,’ said Clarke. And with that mental barrier removed, belief could drive performance. This is what leaders do. They ‘allow us to prove that the impossible is possible. Leadership ensures and inspires a company effort.’
Professor Marc Ventresca, Associate Professor of Strategic Management, Saïd Business School, University of Oxford
Marc Ventresca agreed with the importance of looking at individuals, especially those doing ‘something distinctive and new. This is the cornerstone of what entrepreneurs do, the cornerstone of innovation.’
However, he pointed out, even notable firstlings such as Mark Zuckerberg work in already existing systems. ‘They rebuild them and innovate in them, recreate them and pull together things that weren’t linked originally’ – but they are not going to produce the sort of systemic step-change needed to create the transformations discussed during the day.
We need to pay attention to governance and supply chains, and the edges of existing systems, said Ventresca. ‘Who are going to be the boundary scanners? Who will walk across those incumbent systems and find new kinds of combinations that create efficiencies?’
There are two different kinds of efficiencies, he said: productive efficiency – how do you get more output from less input, do what we already do better? – and adaptive efficiency, which is the ability of economies and institutions to change over time and adapt to new circumstances.
As numerous speakers had already suggested, there is plenty of scope for major programmes to develop new kinds of productive efficiencies, especially from engaging with new technologies. But Ventresca called in addition for the community of leaders, policymakers, and investors to come together and ‘imagine adaptive efficiencies that look beyond what we know how to do today.’
Asking ‘what are the shared fateful independencies that we need to manage better?’ he closed the conference with a plea to ‘form a commitment to new forms of political and institutional organisation, new forms of democracy and governance, new kinds of skills that may not look anything like what we know of today.’
Watch the concluding video.