TRENA Talks EP 06: How To Get Deliver Large Projects In Renewables

Participants:

Host:

Raoul Kubitschek(孔榮), Chairman, Taiwan Renewable Energy Alliance

Co-Host: Y.D. Chang (張雅惇), vice chairwoman, Taiwan Renewable Energy Alliance

Guest:

Gareth Dooley, Senior Expert, Oxford Global Projects https://www.linkedin.com/in/garethdooley/

Raoul:

Welcome everyone to our sixth episode with TRENA Talks. I am Raoul Kubitschek, Chairman of the Taiwan Renewable Energy Alliance, I am joined by my Co-Host and Deputy Chair YD today and Gareth Dooley, our guest and senior expert of Oxford Global Projects.

Yesterday, at the National Climate Change Committee meeting, the Ministry of Economic Affairs provided an update on Taiwan's progress and its mid- to long-term energy development plan through 2039. The plan is an ambitious and comprehensive package that outlines the amount of new power generation capacity the government aims to add over the coming years. Between 2016 and 2026, Taiwan successfully installed 4.9 GW of offshore wind capacity—a remarkable achievement within just a decade, reaching this milestone is still an impressive accomplishment and provides a strong foundation for the next phase of offshore wind development.

In addition to the 4.9 GW already installed, Taiwan has approximately 3 GW of offshore wind capacity currently under development. While the original pipeline was around 3.1–3.2 GW, several projects have since been withdrawn or cancelled for various reasons, leaving the total closer to 3 GW. The government's target is still to reach 10 GW of offshore wind capacity by 2030. However, when you consider the current project pipeline and the time length, achieving that goal will be extremely challenging. Perhaps today's discussion will provide a good transition to what it takes to deliver successful complex projects—and the key factors that can help turn ambitious targets into reality.

YD:

Hello and welcome - I am looking forward to today’s podcast - looking forward to discussing how to deliver projects more efficiently and bringing them to success. Policy discussions are one thing, but turning auctions into action—and ultimately into electricity delivery—is even more critical.

Raoul:

Gareth, Why are you here?

Gareth:

I have been working on projects for more than 20 years. Starting with a background in geophysics, I spent 12 years in the energy sector. 8 years ago, I was moving into offshore projects and global geophysical research. Through working on large and complex projects around the world, he gained hands-on experience in fast-paced project environments and learned what it takes to deliver challenging initiatives. Later, while working in Japan on projects, I expanded my perspective beyond the technical side and developed a stronger interest in the business side. This led me to further develop skills in MBA there, shifting technical expertise to practical and broader view of how complex projects can succeed.

After time in Japan, I moved into the renewable energy sector in Denmark, where he joined Ørsted in a commercial role. This was a significant transition and provided several years of valuable experience in the offshore wind industry. He later moved to Ørsted Taiwan, taking on the role of Deputy Program Director and then continued the journey with SRE, taking on commercial leadership responsibilities and gaining further experience, including involvement in operations and maintenance. At the same time, I pursued a Master’s degree in Major Program Management at the University of Oxford. Later, I was invited by the CEO of Oxford Global Projects to work with the organisation, which became an exciting new chapter in his career. Over the past several years, I have developed a strong passion for helping complex projects. Through my experience across industries—including mining, renewable energy, tunnels, and infrastructure—I have seen many similarities in the challenges faced by complex projects.

Raoul:

Through years of direct involvement in executing projects to gaining a broader perspective on how to deliver complex initiatives, and learning how to help others execute complicated projects. Along the journey, we have witnessed a wide range of challenges across different projects.

Gareth:

This is one side of it, but a lot of what we do is very data-driven, so taking that next step is important. I would say, from an academic perspective combined with a practical mindset, that is the best way to address it. How do we make better decisions? By using data. Because data helps reduce a lot of bias. Much of this thinking comes from the work of Daniel Kahneman, who wrote the book “Thinking, Fast and Slow". His research has influenced a lot of how we think about decision-making and how we approach getting big things done.

Breaking complex things down seems obvious, but putting it into practice is often much harder. I have seen that in real projects: we know we should use data, and we know we should make evidence-based decisions, but ultimately many of us still rely on gut feeling. Experience is not a bad thing, however, the challenge is that just because you have delivered a project in one place and in one way, it does not necessarily mean you can do the same approach in other places. We need to understand the limitations and bring another driven view into the process, which is happening more and more. That’s a positive direction.

YD:

The renewable developments in Taiwan, especially offshore wind development has reached a high-level of maturity, with several projects successfully implemented. From a project management perspective, what role does good or not-so-good project management play in the project outcomes? How do we define a “complex project”? Is complexity mainly determined by the financial value of the project, or are there other factors that we should consider?

Gareth:

There is a definition that a project valued at over one billion dollars is considered a megaproject, provided it has a significant impact on society and is sufficiently complex. Whether one billion dollars is still the right threshold today is debatable, given inflation and the increasing size of projects being built globally across different sectors. Nowadays, projects worth 10 or 12 billion dollars are becoming much more common. There is some research on this topic. It shows that once projects reach a value of roughly 700 million to 1.2 billion dollars, there is a noticeable change in success rates. That threshold appears to correlate with a significant increase in project complexity. In general, the larger the investment, the greater the risk of project failure, and there is data to support that.

When you look at projects globally, we have a database of more than 22,000 projects. If you look at how many projects perform better on budget, it is only about 40%. If you look at schedule performance, only 39% perform better on schedule. Looking at benefits, the figure is around 37%. However, when you consider all three measures together—projects that perform better on budget, schedule, and benefits—the success rate is only 0.4%. Out of 22,000 projects, that is an extremely small number. That is where the Iron Law comes from—that is the whole concept behind it. While some projects may perform well on one or two measures, very few succeed on budget, schedule, and benefits all at the same time.

Globally, what we need to do is break that pattern. Whether we’re talking about renewable energy, offshore wind, or solar projects, this trend needs to change. How do we stop this from happening? In fact, it seems to be happening more frequently. I think that’s a function of the number of projects we’re doing. We’ll probably come back to this point later, but our understanding of the statistical data has also been diluted because there are so many projects.

Raoul:

A project of aroundUS$750million to US$1 billion is typically considered a megaproject. But from my work experience, complex projects can start at a much smaller scale. It could even be something as simple as moving an office.

Gareth:

Even when you are doing something as simple as a home improvement project, we all know what happens. We paint a fancy picture in our minds and think it will take about two weeks, but in reality it ends up taking two months, or we stop halfway through and have to bring someone else in to finish the work. So, we are often very good at overestimating our own ability and underestimating how long it takes to get things done. That same tendency just carries on.

Raoul:

Looking at Taiwan’s offshore wind market, one of the most significant recent developments has been the unfortunate situation involving Shinfox Energy. From the outside, we can see that Taipower has been working hard to manage the situation and minimize the impact. But this is not unique to Taiwan. Similar challenges have occurred in other offshore wind markets, including Germany, where companies have faced severe financial difficulties and even bankruptcy. When entering a new industry with new technologies, new teams and new business models, there is often a strong ambition to make it work. But the complexity of these projects can be far greater than initially expected.

Gareth:

If it's a new project, a new market, or a new technology, you don't just jump into it. You have to build up your capabilities gradually—you can't just rush into it. Most projects fail because they're planned with too much ambition and optimism, and not enough pragmatism and common sense. Looking over the fence and see what has really happened. Is it possible? Has anyone else in the world done it? And if we're going to do it, we have to be prepared to pay the price for that learning curve. That means investing in a lot of front-end loading—doing the necessary planning, preparation, and risk assessment before execution begins.

And if we do decide to do it, we have to ask ourselves whether we're prepared to pay that premium—for the right people, specialist consultants, and new customers because we have all these learnings to make. We also need to be willing to take some of the risks, and risk identification is the key to everything, yet you'd be surprised how often it doesn't happen. Ultimately, if you get it right at the beginning, the rest of the project becomes much easier. Investing money upfront is far cheaper than paying for mistakes later. A dollar spent during the planning and front-end loading stage can save even hundreds of dollars once you're in execution, where delays, rework, and other issues become significantly more expensive.

Most of the time, it doesn't pay to penny-pinch at the very beginning, yet that's exactly what many projects do. They're often uncertain whether the project will receive approval, secure permits, or pass the various stage gates, so they hesitate to invest money upfront. But that's the wrong approach. Running a project is very different from running an operation. I remember someone from an EPC company in Taichung once said to me, "Gareth, you need to remember this: EPC has all of the money and none of the time, while operations have none of the money and all of the time. That's the payoff. You do have to pay the premium upfront and be mindful of all these issues.

Raoul:

In offshore wind, it depends on the project owner or developer and their business strategy. Some companies are more focused on developing the project through to completion, while some developers also have large in-house engineering teams, might be more financially driven, push and then sell. Because there is so much uncertainty during the early stages—such as whether the project will win the bid or obtain the necessary permits—developers may be hesitant to invest upfront. For the project manager, they may be responsible for managing the budget but are not always the final decision-maker.

From your experience, upfront loading is one issue. How do you find the right balance? How do you start a project while making sure the other important things are also set up?

Gareth:

Working on a project requires organizational design and proper governance. Again, if the governance and assurance are not set up in the right way, you could introduce a lot of noise further down the line. People need to have control of the project. A simple example is to look at the Madrid Metro and make it as simple as possible. The Madrid Metro is not the most attractive metro system in the world. If you compare it with London and Paris, it doesn't have ornate stations. Everything is very plain and utilitarian, but it has been successful and transports millions of people every day across Madrid. They had a very clear principle when they built it: make the Metro as simple as possible. 

Offshore wind used to be modular. Over the years, however, many aspects of the industry have changed. In terms of turbines, foundations have evolved, and these advancements have introduced additional costs. At the same time, complex engineering requirements, many new technologies have also been introduced simultaneously, while more and more projects have reached the execution and delivery stage. This has increased the overall complexity of projects, and this is not unique to Taiwan—it is a global trend. With so many changes happening at once, there is often a great deal of noise that can put the project back to its fundamentals and ask, "What is the outcome we want to achieve?", Once that is clear, the focus should be on making the delivery as simple as possible. That is why, when you look at the scale of complexity and project success, simplifying where possible is often one of the most important factors in achieving successful outcomes.

When you look at solar projects, they rarely go over budget or behind schedule because of their simplicity and modularity. At the opposite end of the spectrum are nuclear projects, where budgets can escalate dramatically due to their different modularity. That is why so much investment is being directed toward Small Modular Reactors (SMRs) today—the goal is to make nuclear projects more modular. The same principle applies to offshore wind. There has been a strong push toward from V164 mega-turbine up to the next platform. However, every increase in turbine size introduces additional complexity. Larger turbines require bigger installation vessels, upgraded port infrastructure, and more extensive engineering support. All these factors make people assume that bigger turbines will automatically reduce costs. Instead, they can add significant extra costs because these challenges were not thoroughly sorted out beforehand.

I think there is still a lot to be done upfront, and part of that is adopting a right-to-left thinking approach. We need to first ask: where do we want the project to end up, and what do we need to deliver to achieve that outcome? Then we work backward and consider how we get there. Rather than starting from where we are today and asking, "How do we get here?" We should first define the end point, this is the goal, and then plan the pathway backwards. When you take this perspective, you begin to identify many issues that may otherwise be overlooked. It allows you to plan from the beginning—not just in terms of spending money, but by anticipating potential problems, understanding what could go wrong, and considering the project from different perspectives.

Raoul:

When we moved from R2 to R3, we took a very different approach to our localization strategy. As turbines became larger, the supply chain requirements changed, and the land and infrastructure needs also evolved. You cannot simply rely on the same suppliers unless they are willing and able to upgrade their capabilities, facilities, and skills. That is definitely one of the challenges that can throw you out of the curve on this kind of localization.

Gareth:

That is where policy and government can play an important role. The biggest thing they can do is to remove risks from the system. Nobody wants to invest time, money, and effort if they are uncertain whether they will obtain permits or meet certain criteria. If governments have an open dialogue and ask, “Where are the risks in this industry?” and “What risks can we take on?” They can help create a more stable environment. This approach has been applied in industries such as nuclear, where governments have taken on certain risks, as well as in major infrastructure and transportation projects.

Offshore wind, however, has often been treated differently. Subsidies have been removed in many countries, localization requirements have increased, regulations have changed from one generation to another, and access conditions have also evolved. Perhaps some of these challenges are partly the result of issues that the offshore wind industry has brought upon itself.

Raoul:

When it all started around 2016 to 2018 in Taiwan, all these promises were made by the industry. They said that if Taiwan could provide a strong project pipeline, developers would come over, create jobs, and set up manufacturing facilities. But it hadn't just been told to Taiwan; Japan and South Korea were told exactly the same thing. Regulators reacted that it was a good thing—more jobs, more localization, more technical know-how. All these islands—Japan has its rules, Korea has its rules—were very much looking into it. However, the Philippines curiously said, “Yes, you have to promise these things—development of workforce and infrastructure.” But they did not come into the auction and said, “We want the turbines, the vessel manufacturing, and all of these.”

Gareth:

There is one thing to plan for it, and another thing to mandate it. That's where this is connected. If you look at other industries or investments coming into the country, there aren't the same requirements. TSMC, for example—look at how much money they invest and how many jobs they create. Do they have to localize? Is there a localization requirement for their lithography machines? No. Because the best place to get them is in the Netherlands or the UK. They bring in the technology and develop it that way, and what happens? The industry grows because the market demands it, and it doesn't have the same restrictions.

I'm not saying that some localization isn't good, and things have changed. However, all countries need to be aware that the more risk you put into a project, the more cost your project is going to have.

Raoul:

So, as a Project Managing Director, somehow something feels not quite right. But it’s all very complex. How would you measure your project? What kinds of things would you look at? Is it having a happy client? Is it delivering truly on time? Or is it having a happy project planning team? What’s a good way to measure your project?

Gareth:

There are a lot of discussions and research on this across different industries. There are a few things that are key, and you almost can't hide them. One is the burn rate—the amount of money you're spending—and how much float is left on the critical path. If you're not spending enough money, you have to ask: What hasn't been done? What hasn't been paid? What if the contract hasn't been signed? On the other side, if you're spending too much, are there a lot of issues with delays? There's one truth that runs through every project, and that's money. Then you look at the schedule: Are we on schedule? And you also need to look at the project organization. Is the organization starting to turn over? Are people becoming overstressed? Those are all signs that tell you how the project is really performing.

If you start seeing a lot of people, maybe Shinfox is one of the indicators as well. If you think about those three indicators together—and there are a lot more—those indicators can help you identify problems early on. I think you would probably see high turnover, a high burn rate of spending, and the schedule starting to slip, slip, and slip. It's about stepping in sooner rather than later and saying, “Let’s pause. What’s going on here? What is the problem, and how do we address it?”This is something more fundamental.

YD:

The pause—how do we decide when to take that pause? I always find that fascinating. It’s very difficult, whether it’s a government decision, a developer decision, or a commercial decision, to say, “Let’s stop and review what’s going on,” and then move on.

Raoul:

It’s an extremely hard call. It takes a lot of thought, because if you look at the project only from the perspective of the company executing it, they might turn on the red light. But when do you sit down with your client? When does the client come to you, or when do you tell them politely, “We need to talk”? There is one side, and there is the other, there are different considerations and expectations. When do you open up the discussion with both parties about what’s really going on?

Gareth:

There’s something I always say, especially when you’re in a leadership position: I only like to be surprised twice a year—on Christmas and on my birthday. Everything else, let’s just talk about it. It can be a discussion where you say, “I see this is a bit of a problem. It’s not a big problem, but I just want to let you know that it might cause a delay or might cause something later on, and the whole liquidated damages thing is not an issue as well”. I’ve been working with a great operator in Europe, a TSO. They were building some offshore substations, and they were approaching the liquidated damages cap. Then the discussion becomes: What would you do afterward? Do you step in and bring the contract manager in to negotiate the contract for a number of years? That was my first job in offshore wind.

The last thing you want to do is bring that contract into it, build that relationship, and have that discussion. It creates psychological safety, and the client gets much better products and deliverables, and you get a much better manageable project. We’re assuming the client is reasonable and is willing to have those discussions. There are a whole lot of things that are definitely important to have a frank discussion about. There may be something happening down the line, but as long as you’re aware of it, nobody is surprised. Nobody likes surprises.

Raoul:

Perhaps just to close this off, and I’m sorry to bring up Taipower again. Taipower is a government-owned entity, so they have to follow government procurement rules. Even if they are WT, they just open it up for everyone. Linking this to a particular project, somebody could come and bid below it. But why did they use that contracting strategy? It is extremely risky to put everything into one EPC. And we always say, “This is how they need to do it.” There might be another way, but this is the best way for them to get into the project, get it done, and give it to one company, with that company doing the subcontracts and controlling everything.

In offshore wind, as we’ve seen for years, wind turbines, towers, foundations, cables and marine coordination—all of that. There are a lot of packages going on. Even for a big project, is the contract strategy important or not?

Gareth:

This is key. What you see more and more often is a more integrated approach, where it’s no longer the common structure where you have the developer at the top and everybody else fitting underneath. Again, going back to what we discussed about liquidated damages, everybody is in the same boat. If one of those packages is having a problem, everybody else is going to suffer. It doesn’t matter whether it’s the turbine package, the onshore substation, or even marine coordination, because all of a sudden, nobody can get any work done—the same issues with vessel coordination. That has to be reflected in the contracting strategy these days, and that’s why we’re seeing more and more integrated delivery teams.

You see Terminal 5 in the UK, and a few others as well. What they have done successfully is that they make everybody benefit when the project succeeds, but likewise, everybody feels a bit of the pain when it doesn’t succeed. Being that open and transparent as an organization means everybody knows how they affect everybody else. Whether that has not translated to offshore wind is because certain suppliers have their own specific way. Should we say, “big, aim-high attitude”! Then again, with the right contracting structure, if you deliver and get things done on time, there’s a bigger slice of the pie—we can all benefit.

But it does work, and there are lots of practical examples where it has been really successful. The days of developers coming in and strong-arming are over. We all know that, for better or for worse, depending on which side of the desk you’re sitting on. That needs to be built up more. There was a lot of collaboration earlier on, and then it kind of went the other way, swinging back in terms of the supply chains. They remember—they have a long memory.

I was speaking to somebody in Europe—I’m using your business example, and I don’t want to get into any trouble in Taiwan. They were working on some of the German projects, and they felt they were being pushed very hard, with a lot of pressure put on them. The other developer wouldn’t do that at the time, but years later, this is still ingrained; it’s still there. You can imagine somebody working with that developer again and saying, “I remember what they did in the past. I’m going to be extra careful, put extra LDs in the contract, and get my best contract manager on this. We need to move straight to the contract. That whole concept and process needs to be changed.

Raoul:

Given the significant progress we can see from the outside, particularly in construction and major projects, there is also a lot happening behind the scenes. We’ve seen successful examples such as CIP and their projects in Changfang and Xidao, Zhongneng, and Fengmiao Phase 1, which have really gained momentum. These projects come and go, but the question is: how do we learn from each other and carry those lessons forward? What is the best way to learn? What do we learn? What do we not learn? And how do we do that?

Gareth:

That is a very interesting question. That was actually the subject of my Master’s research at Oxford, particularly in offshore wind, where I had to speak to lots of senior people across many different organizations. Back to the academic perspective, the scholar looks at it rather than just intuition: how do your projects learn from the first time projects in a new country, using new technologies? And how do you bring that learning back into the organization from the project? It was very clear that there are lots of good initiatives within companies to capture lessons learned. The challenge is that by the time the project ends, everybody is ready to do the lessons learned?

The key people have already left and moved on to different projects, because they’re often the first ones to leave. Towards the end of a project, we tend to take the most senior and experienced people and move them on to other projects, and the team can continue to run the project. So, you don’t have that there. The other thing is very clear: the issue of “capture.” Did anybody read it? Did anybody actually do anything with it? Companies are structured to do that—some do, and some don’t. What was very apparent is that a lot of that expertise and knowledge resides in individuals, and that’s the challenge. The challenge is retaining those people with their knowledge and experience.

One of the things we see is that organizations have projects in different regions and different countries, and as offshore wind has grown, a lot of expertise has become concentrated within one or two companies. They’ve started to disseminate that knowledge internally across different projects, but also share it with other companies. The other thing to bear in mind is that these are purely transferable skills. They’re not necessarily specific to offshore wind—they can be applied in other industries and in other countries as well. Going back to the earlier discussion, developers are sometimes challenged by salary expectations. But actually, the amount of money involved in retaining these people is relatively small compared with the impact and value they bring to the industry. This is the missing part as well.

I’m not saying we should start bidding more for people—nobody wants that, because that would raise the general cost. You have to recognize that we need to retain these people within the region or the project, because of the value they bring. Especially in project management here, it is very difficult to keep a pattern and people. However, this is the key role, and AI is going to come in and provide support. It can change things quite a lot. But you still need that key experience and expertise to set things up behind it, at least at the moment.

YD:

The key person is someone who can really identify the risks, even just by observing during meetings, particularly the risks to the schedule.

Raoul:

Even with retention, just look at what happened in Taiwan. The market broke away, and people moved into other industries. Even when the market comes back, we have people interviewing now and asking them to consider coming back with the same compensation they had before, because they have the know-how.

YD:

When we visited Vice Premier Cheng, she was very impressed with what we had done over the past ten years. But the thing is, we are losing these people, and if there is no practical pipeline to support and retain them, it’s a pity. These young men are learning almost everything, and once they have the opportunity to practice it one or two times, they will become excellent people in the industry. In any case, whether you need the energy or not, it is very important that we keep practical pipelines to retain these people.

Raoul:

Perhaps this is a closing question. Would it be beneficial for companies to sit together, not only to applaud the amazing projects we’ve done, but also to ask how we can get some of the experienced people within the supply chain to bring that experience into the industry?

Gareth:

That is one of the key things in terms of suppliers, what they can bring to the table as well. Not every developer wants to listen to suppliers. Suppliers may be working across many different developers, and they can bring those discussions in. There’s a forum right there, of course, to bring those lessons learned. A lot more is happening on the owner side, sort of looking up, more on the investor side. To my knowledge, it’s not in Taiwan yet. There was some discussion a while ago, like an owners’ club, basically having discussions about O&M: How do we deal with this issue? Are there any problems? Can we share the resources?

Because we are not competitive anymore, we should all be working very closely together. But they still have a very competitive mindset, and they should be open to big wins. When you move into the development phase, there’s still a lack of trust between parties. The conversation was like this: the project director might say that he needs the vessel, and they worry that if they give it to him, they won’t get it back. There’s a little bit of mistrust, and there certainly could be a lot more done.

If you look at ORLEN Gas, that is sort of the way. There are lots of companies sharing information; they do lots of projects and learn together. It’s a different business model, and there are a lot of risks doing that. Maybe in the next phase or stage of offshore wind, we should look at how this thing is going, and let’s look at this as a program together to see how we can share and build resources around it.”Of course, suppliers don’t necessarily like that either, because they don’t like their customers talking and comparing prices. But that is definitely something that needs to be focused on and done, like a gray area. It can bring a lot of value.

YD:

Talking about the program from the government’s perspective, there are challenges with Round 3. A lot of projects have been awarded, but only a few are moving forward, and some have already been terminated. With Round 3.3 coming up, with the bidding deadline at the end of September this year, how can government stakeholders improve the auction design to better acknowledge the complexity of projects? And it’s not just Taiwan—the Philippines has also paused its offshore wind auction.

Gareth:

If you look at how the market has evolved over time, different countries have been revisiting their frameworks this year. It is time now to take a pause and say, “What’s changed?”It is time now to take a pause and say, “What’s changed?”I know how it’s changed. We’ve had discussions about technology, and also why? it may be a good reason to bring offshore wind back. Understanding, sitting back and saying, “We need to depend on this as a program.” The government needs to sit back and say, “How does the program work? How do we do the project management of it?” Not that they have projects to manage, but they have to understand and build the structure around the pipeline, and be more commercially sensitive to say, “What do we want out of it?” We want green electrons. We want the electrons to stop at the stage? It’s not necessarily about building up certain areas of expertise or a supply chain in the country. It’s about how we do this as quickly and as cheaply as possible, and how do we make that happen?

Again, going back to this right-to-left thinking: Taiwan aims to reach 10.9 GW of offshore wind capacity by 2030, 18.3–19.9 GW by 2035, and 24.7–27.9 GW by 2039. So 2039 is what we need—how do we get there? We need to work backward, not just say, “Here we are, how do we get to this capacity? How do we initiate all of these to make it happen?” This is the framework that these projects need to work with. How do we enable this to happen? Where are the pinpoints for them? Where are the risks? Can we take away the risks to make it easier for them to develop and deliver the projects? That’s the mindset we need to change.

Raoul:

You can see Round 3.3 that they’ve tried to reduce the risks. Some people really love it, partly compared to Japan. The point is a bit split. I could go into a spin on ESG and such localization, and that’s become quite complex. You can also see the discussion around 3.1 and 3.2, but even earlier, the government seemed to have no good mechanism to terminate projects. They basically had to wait until the other party would kind of give up or return. They seemed to have no way of saying, “We know from everybody else that you’re not even close, and you haven’t talked to your supplier, so we have to stop you. We need that energy. How can we help you? What is your problem? Do you need access to capital? Do you need the permits?”

Gareth:

Exactly. Internally, they go through three years. That whole stage gets processed, everybody does it. It’s something that can be echoed from a policy perspective. I know from experience, going in and talking to the government, giving them updates and letting them know how the project is going. There are no steps, and also no carrots. There’s no mechanism to say, “If you reach this stage, you get a certain amount of incentives.” Everybody is racing to the finish line, because we need to get this done, need to get the contract signed before this happens. You need to continue with the pipeline. I don’t think it’s too late, but I need something to push, to get it going again. That is where the government can step in as well.

Raoul:

For the government, if you look at Taiwan, and even if you look at the Philippines, now you have to coordinate with all kinds of ministries. One ministry is in charge of one business unit, and they have to coordinate with others, whose KPIs are not aligned, or who have other issues. That becomes an important exercise for them.

Gareth:

I’m not saying it’s easy from the government side, because they have all those factions to overcome. There are super-smart people there, but they are not unified. They are not aligned on the objectives, and their goals are not aligned in the same way. Policies to align all of that could help a lot. That’s similar everywhere, not just in renewables. In all areas, that’s the challenge for the government. To make that huge shift, which is what it is, is very difficult to turn and move. But you need to start doing it in small increments and moving it forward. I’m not saying the government doesn’t have project management experience. They need to take that perspective from the developer and say, “If I were a developer, what would I need to move this project on?”

Raoul:

Thank you. AI is everywhere. I just came across “consultancy AI slop” yesterday. This side is really being used by people who understand that AI has so much potential. If you look at this particular area, what do you see? You talked previously about how AI can provide better support, probably organize some of the agencies in a better way and help with control. But what is AI doing on your side, in your industry?

Gareth:

A lot of good ideas and thoughts are coming out, advising on AI startups in Germany, and what they plan to do is using AI to augment TMO, and a lot of the issues around information flow. So much good information is generated and put out there, and it kind of sets the news out. There’s too much for anybody to process. That’s a whole part of the insurance company, the information coming through, down to the key things that need to be looked at, which is normally periodic. AI allows you to do that on a daily basis, whether for the project director or one sitting at the board of the company, to be able to come in, have a look, and say, “Where is the project now? Are we talking about the red flags project coming up?

We’ve done some work—actually, OGP has done some work with the Hong Kong government—to create a dashboard that basically flags up issues with projects. Because the portfolio of projects is so big, hundreds and hundreds of them, it’s very difficult to manage all of that. We set up a central machine-learning system to flag up any issues that may be coming. It doesn’t tell you the answer; what it does is, “You have to look at this.” And that’s where AI can help a lot.

Back to noise and effort, everybody likes to talk about Lean(Lean Six Sigma)—it was very sexy a few years ago, and it's still there, you can’t take that away. The other thing Lean focuses on is variation, and variation is automatically caused by risks. They create noise, and things go wrong. Normally, this happens when there’s a governance issue. Then we have to deep-dive into it and throw people at the problem.

A project is struggling to deliver or trying to fix an issue, and suddenly questions start coming down from on high. You start pulling people off their tasks just to answer those questions. That friction can be mitigated by AI, because you don’t necessarily need to ask, “Why didn’t this work? What is going on, and what are you going to do to fix this?” With AI, you can get real-time insights, and take noise and pressure out of the system. So there’s a lot to gain if it’s done right—that’s the main thing. There’s a lot happening, and it’s very exciting.

Raoul:

Finally, from your perspective and experience working with OGP, what truly makes an effective project manager? When a project starts to go wrong, how do you turn it around? What are the key factors that help a project manager continuously improve over time?

Gareth:

There is a human expectation of being a project manager that is often overlooked. There is intention to details, being able to see the big picture, dive into the details, huge part of project managers getting things done with people, and being able to motivate people, and have the difficult conversation, to be approachable and understand what’s going on with projects, being able to communicate that is huge part of it, that’s sometime overlooked. You often find project managers are the ones that are more objective, look at the data rather than having a strong feeling, or deep intuition which is important because the experience part can be married with data, and being able to communicate that to the team and outwards is the key there.

Back to surprises, being able to manage those surprises, being able to say to the team, “Come to me, we fix it, and solve it!” be able to go up to senior leadership beyond and say “Listen, just let you know, we have this issue, I am on it, I manage it. Do worry about it. And if it does get worse, I will come and have a discussion with you.” That is the piece often missed. The PMI gives you the framework, but is how you work with that framework, and bringing it to all your stakeholders is the key.

Raoul:

In my experience, some people are naturally good project managers. They may not have formal training, but they have hands-on experience, communication, and practical knowledge. However, you mentioned that project managers need formal training, such as PMI or PMP. So, what is the point that project managers should have all these types of training? What is your take on that one?

Gareth:

Training is the key. One benefit is that it gives you the vocabulary. When everyone talks about a work breakdown structure and critical path, we all know what we’re talking about. That is crucial and key. Another benefit is that there are a lot of tools out there that can really benefit and help projects. For example, QRA, risk analysis—that bottom-up side of risks—and if you look at references and forecasts on the other side, those are key as well. Early on, it takes bias out of decision-making. Are you aware of what can help you make better, more qualified decisions? And, wrapping all of that up, it’s about understanding how to communicate these things effectively.

Back to risks, I’ve worked on significant projects—$5 billion-plus projects—that didn’t have a risk function or a risk register, and yet they were coming up to sign contracts. Now, if early on, all the people involved—it wasn’t a government project, but if it had been exposed to these practices—they would have had that checklist: “What do we need to pay attention to?” before the contract, and then brought that to the table. And the project would be a lot more robust, less likely to go over budget and schedule. It’s important. Even if you don’t get the certificate at the end, just being there and understanding it is the key.

I’m passionate about offshore wind in Taiwan. Not only is it a good thing to do, but it needs to be done. I just hope that all the leavers that have been put in place help shovel things along. There’s a lot right there, and there needs to be a bit more data-driven decision-making, which is important. Whatever comes, you should deeply ingrain what Oxford Global Projects does. That’s why we end up talking to governments all over the world—we deeply influence policies. The reason is because not enough has been done. It doesn’t matter who does it, as long as it gets done. Let’s make life easier for everybody.

Conclusion:

Gareth’s interview highlights a fundamental challenge for Taiwan’s offshore wind industry: how to turn the experience of one project into learning for the next. Project knowledge should not stay with individuals or disappear when people leave the industry. It needs to become embedded in organizations, supply chains and government systems.

Training, risk management, integrated delivery and AI can all help capture and share that knowledge. But ultimately, it comes down to people working together, learning from both successes and failures, and creating a stronger project-management culture. For Taiwan, the opportunity is to build on the experience of the past decade, retain the people who have gained that experience, and make every project a stepping stone to the next. At its core, it is about learning from projects and experience—and turning that learning into better projects for the future.