In short
A special Energy Gang episode from Wood Mackenzie promoting the book Connected, arguing the energy industry faces rising volatility and complexity, so strategic and operational planning must shift to scenario-based decision-making using integrated data and AI.
Guests and backgrounds
Jason Liu, Wood Mackenzie CEO; previously ran major software/data companies for ~30 years, including energy software firm Allegro; first outsider CEO and brings AI/tech background. Sunaina Odjalan, formerly Senior Director for Corporate Strategy and Climate Change at Hess; has electrical engineering degrees; did six years strategy consulting at Rice/Boston; moved to Bernstein as an equity analyst for energy/energy transition.
Key claims
Traditional planning paradigms are “broke” because predictability is worse even over 6–9 months. Energy should be framed as “energy evolution” (a dial, not a switch) needing hydrocarbons plus renewables. AI/data can improve clarity, but synthetic data and siloed datasets cause bad decisions.
Notable examples
Wood Mackenzie analysis estimating ~$80B value loss from suboptimal European majors’ renewables moves; congestion/interconnection queues for AI-driven grid load; Hess’s CCS/adjacent decarbonization work (e.g., Bakken carbon capture); US developers adding natural gas to solar/wind/storage “integrated energy plans.”
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Chapters
Tap a time to open that second in VOGuests Introduction and Backgrounds
0:45 to 2:34
Introduction of Jason Liu and Sunaina Odjalan, exploring their backgrounds in the energy sector.
“And it's also a pleasure to welcome to the show for the first time another new guest, Sunaina Odjalan.”
Jason Liu's Path to CEO
2:34 to 4:28
Jason Liu discusses his journey in the energy sector and his perspective as CEO of Wood Mackenzie.
“And as you say, very much traditionally at Wood McKenzie, we've appointed CEOs from inside the company.”
Sunaina Odjalan's Career Insights
4:28 to 6:56
Sunaina shares her career journey and the transition to energy equity research.
“You know, energy transition, and I'm sure we'll talk about this as well, is at such a critical point.”
Challenges in the Energy Industry
6:56 to 12:04
Discussion on the current challenges in the energy industry, including volatility and strategic planning.
“Yeah, I think that's really interesting.”
The Energy Transition Debate
12:04 to 14:00
Exploration of the concept of energy transition and its implications for the future of energy.
“And I want to pick up on something you mentioned, Salina, earlier.”
The Evolution of Energy Perspectives
14:00 to 16:53
Discusses the evolution of energy perspectives and the complexities of energy demand.
“But also, we need to see more renewable type generation going on as well.”
Integrating Energy Components
16:53 to 18:09
Explores the integration of various energy sources and their interdependencies.
“And the other thing I would just add about what's made energy transition a little bit more, I think less of a relevant term is the integration of all the energy components.”
Impact of Regional Differences in Energy
18:09 to 20:58
Examines how regional differences affect energy security and transition strategies.
“As you say, those differences are very significant and persistent.”
Investment Strategies in Energy Transition
20:58 to 22:42
Discusses decision-making and investment strategies in the context of energy transition.
“And Greg Hill, who was our president and John Hess, came to my team and basically said, you know, figure out net zero.”
The Importance of Data in Energy Decisions
22:42 to 24:01
Highlights the role of data in making informed energy decisions and the pitfalls of data reliance.
“And we looked at our knitting and we said, can we actually move into some of these spaces?”
Show all 17 chapters
Challenges in Energy Data Utilization
24:01 to 28:00
Explores challenges faced in utilizing data for strategic planning in the energy sector.
“Can you talk a little bit about what you mean by that and what are some of the big issues in terms of energy data as you see them?”
The Role of Data in Energy Decision Making
28:00 to 31:00
Explore how data and AI can enhance decision-making in the energy sector.
“We were linking it to associated gas, if there is associated gas.”
AI's Impact on Energy Modeling
31:00 to 33:50
Learn about the transformative effects of AI and hyper-modeling in energy forecasting.
“And obviously, it's a credit to her and her team.”
Strategic AI Initiatives at Wood Mackenzie
33:50 to 36:15
Discover Wood Mackenzie's AI initiatives aimed at improving energy insights.
“And there's an infinite number of models you can use, tree models, reoccurring neural network models, you know, TFT models.”
Advice for Energy Industry Decision Makers
36:25 to 38:40
Understand the key advice for energy leaders in a changing market.
“But, you know, outside of reading the book, the way we typically think about things is kind of what we call the winning trifecta, which first and foremost is what Sinan and I just talked about, which is people.”
Embracing Uncertainty in Energy Planning
38:40 to 42:02
Explore how energy companies can thrive by embracing uncertainty and data-driven decision making.
“I think I'll just double click on something that Jason said.”
Opportunities in the Energy Sector
42:02 to 43:31
Explore the significant investment potential and growth opportunities in the energy sector.
“Anything else we haven't talked about that you wanted to raise?”
Transcript
Automatic transcript. May contain errors.0:02Hello and welcome to The Energy Gang, a discussion show from Wood Mackenzie about the fast-changing world of energy. I'm Ed Crooks and welcome to a special episode where we're going to be talking about understanding the connected world of energy. The reason for this is that we at Wood Mackenzie have just published a new book which is called Connected and it's written by our chief executive Jason Liu and our chief analyst Simon Flowers. You can actually download it from our website woodmack.com so please do do that. Go ahead and check it out. What it's about is about the key challenges facing the energy industry today and what we see as some of the solutions.
0:36And to talk about that, it's a great pleasure to welcome Jason Liu, who, as I say, is our chief executive here at Wood Mackenzie. Hello, Jason. Welcome to the show. Hey, Ed. It's great to be here. It's a pleasure. Yeah, thanks very much for joining us. And it's also a pleasure to welcome to the show for the first time another new guest, Sunaina Odjalan. Sunaina was until recently the senior director for corporate strategy and climate change at HESS, the oil and gas company. And you've recently started a new job, haven't you, at the investment company Bernstein. Hello, Solana. Welcome to the show.
1:03Thanks, Ed. Thanks for having me. Hey, Jason. It's really good to be here. Yes, I'm excited to say that I'm going to be an equity analyst for Bernstein in the energy and energy transition sector. Fantastic. Yeah, well, great that you're able to join us and great to take part of this conversation. I think that's a fantastic perspective to share this discussion we're going to be having about the future of the energy industry. One of the things we always like to do when we get people on the show for the first time is talk to them a bit about their careers, how they first got interested in energy, and how they got to the roles they now hold.
1:34Maybe, Jason, to start with you on this, as I say, what attracted you to energy, and what was the pathway that took you to being chief exec at Wood McKenzie? Oh, great, Ed. You know, I just would probably start off and say that I think energy's been a bit of the family business. My father was a civil engineer and actually a wind expert, and so in my youth, we would actually spend vacations touring wind farms. You know, I think we you know, as younger kids would have preferred being on the beach or amusement parks, but we'd be touring wind farms. So I guess that's where it all started. I will say over the last 30 years, I've had an opportunity and a pleasure to run some of the larger software and data companies in the world.
2:12And, you know, I did have one stint where I ran one of the largest energy software companies in the world, Allegro. And now I've been at Wood McKenzie now for about 15 months as CEO. And the only thing I would add is, I think at Wood McKenzie, we have this fantastic 100-plus year history, but I'm the first outsider from Wood McKenzie, but also the first one with a tech, you know, AI background. And I think that's part of the outsider's perspective that I think I can help bring to the energy world is that kind of how does energy meet tech and AI? Yeah, absolutely. That is very interesting. And as you say, very much traditionally at Wood McKenzie, we've appointed CEOs from inside the company.
2:49you I think the first one in the whole history of the business to come from outside as you say really interesting to bring that tech perspective and that's definitely something we're going to be talking about on this show before we get to that though Sunaina and tell us about your career so you spent a long time at Hest didn't you but what's your your journey been you know I was reflecting on this Ed I actually entered energy by accident it was early 2000s and I found myself with not one, but two electrical engineering degrees at a time when tech was crashing. Slumberger was hiring. It was not a company that was on my radar, but oh well.
3:27And in hindsight, it was fantastic because it provided such a great foundation for understanding how complex and how large the energy system can be. I got into strategy through business school at Rice and did six years in strategy consulting in Boston before joining Hess. And at Hess, I've done various roles in onshore and offshore before joining corporate strategy and doing our climate change strategy as well. So I've moved around a lot, joined energy by accident, but stayed by choice. And then getting into energy equity research in particular, as you've been saying, that's kind of a new thing for you.
4:06So this is a new field for you to explore. You know, it's more of like a shifting perspective for me. At Hess, I had the chance to shape energy and climate from the inside. And now it feels like a natural next step because at Bernstein, I'll be analyzing the sector, a sector that I love, from the outside through more of the market lens and more of the companies. You know, energy transition, and I'm sure we'll talk about this as well, is at such a critical point. And I'm excited to bring my background in engineering strategy, as well as climate and sustainability to help investors and companies really cut through some of the noise that's been there in identifying the opportunities as well as the risks that lie.
4:48Right. Absolutely. So let's start off then by talking about the shape of the energy landscape. Maybe, Jason, get your thoughts on this then. So, as you know, you've been about 15 months leaving Wood Mackenzie. When you think about the energy industry as it stands today, and the challenges that it faces, what do you think are the biggest issues? Obviously, I wish my partner, Simon Flowers, and my co-author of the book had an opportunity to also chime in. But I think the way we laid out the context of this question in the book was a substantial kind of deep dive into why has the energy landscape changed so dramatically, particularly in the last five to 10 years.
5:27There's an incredible amount of additional complexity now, whether it be now the rise of energy demand, energy transition or climate issues, coupled with now regulatory involvement and other pretty substantial changes around national security and others. And that has created a lot of additional complexity. And what also is kind kind of added to the mix is an enhanced amount of volatility. And so in the last year plus, myself and Simon have gone out and met with several hundred of our largest energy customers. And we met with the CEOs of these large energy providers. And we definitely started recognizing some patterns and almost like a zeitgeist that was happening in the industry, which was there was just a lack of predictability going on in the industry.
6:11And we heard that from multiple CEOs that felt like they could no longer predict the future. And we're not just talking about 5, 10, 15, 20-year outlooks. We're talking about predicting over a six-month, a nine-month period of time. And we ultimately came to the conclusion that the historical ways in which strategic planning and operational planning had to be fundamentally rethought. In fact, they arguably had been broke because of this volatility and complexity. And so the book is all really lays out the reasons why we ended up in a situation where the old paradigms don't work. And then our proposed solutions on what the new paradigm should be for planning and decision making, both strategic and operational.
6:56Yeah, I think that's really interesting. And I do think, you know, people sort of seem to always say, oh, everything's very uncertain and things are more uncertain than they've ever been and so on. But I do think there are real grounds for thinking that the uncertainty today is higher than it's ever been. as you say, a lot of different kind of interrelationships, interconnections that didn't exist before that exist now. The whole issue of climate, as you say, which is something that people didn't really think about in the energy business 30 or 40 years ago, is now very present still, despite everything.
7:25And then through into that AI and what's happening, you know, this kind of brand new world-changing technology, which is kind of evolving at such a fast pace. As I say, I think, you know, it's not just an empty thing to say, to say uncertainty is greater than ever, I do think there's something really in that. Absolutely. You know, I think, Ed, just to add, you know, I think on a couple of the vectors that you mentioned, you know, first of all, with AI demand, I've been in countless customer meetings just recently. And one of the obvious questions is, what is AI demand going to be? You know, is it going to be the hockey stick increase, or is it going to be more tempered?
8:00And, you know, with Wim McKenzie, we're obviously doing everything we can to generate that type of output, whether it be our sensor data. But also, you know, when you look at things like congestion queues and interconnection queues. There's a lot of ways that you can try to estimate that information. And I think those are some of the techniques that we would say helps bring a little bit more clarity to some really important questions because the impact is substantial when you're talking about that type of increase in load on our grid. Yeah, as you say, any then insight you can bring into that uncertainty is then kind of particularly valuable.
8:33Now, what do you think then when you think about key challenges in energy right now, what are the big ones that stand out for you? Yeah, you know, looking at it from the strategy perspective, so inside a corporate perspective, one of the things we used to deal with is, it's exactly what you said, Jason, where basically the energy sector is so different because you're dealing with these four forces, essentially, geopolitics, policy, tech, and then the regular consumer supply and demand, right? And all of these four forces are interacting constantly. And so you could have a breakthrough in tech, but if policy shifts, then the economics might not work, right?
9:13Or if consumer demand shifts, kind of like how we're seeing in EVs right now, you have to start again from, you know, forecasting oil demand for the long term, right? So that's one of the inherent challenges that companies are working on. And so really one of the solutions is to think about how do we think through exit ramps so that we can pivot quickly when we find the information changing. I think an inherent other, more so for oil and gas companies, challenge is that it's a depletion business. So every barrel that we have produced was our best barrel, right? And we have to replace that barrel somehow.
9:53So fields decline naturally about 5 % or 7 % a year. And if we're not constantly reinvesting, your production and your cash flow shrinks. And so do you go find more resources? Do you buy the resources? How do you actually allocate that capital? And then two other challenges that I'll highlight in the energy space is the time horizons, not for every part of the energy space, but largely for infrastructure as well as offshore fields or pipelines, things that you were talking about, Jason. The time horizons are long. So you're making a decision thinking about a certain future, and that future might not exist when you're coming to FID, right?
10:35And then finally, in the commodity space, we're essentially price takers. we're not price setters. And that's the last part of that challenge. So it's an interesting dilemma. To maybe add what you're saying, just some real life customer stories. One is, we actually did some analysis recently on some of the European majors and some of their, obviously, recent, over the last three to four years, decisions to obviously move into the power renewable space away from some of their core competency in oil and gas. And unfortunately, there was obviously some less optimized decisions that were made. And we actually calculated the exact impact.
11:11And it was roughly about$80 billion in lost value with kind of suboptimal decisions made in that, you know, in some of the investments they made. But also, I just was out meeting with one of the largest developers in the US and the CEO. And, you know, one of the biggest issues that they're dealing with is obviously with the changes in the big, beautiful bill, you know, what does that mean to them real time from an investment and planning perspective? and even to a point now where they're considering options as gas, you know, both trying to understand the price of gas because it has an impact on power, but also, you know, gas power plants obviously have a big potential impact on AI generation.
11:52And so they need to understand that piece as well to make better decisions. So the world has obviously gotten a lot more complex and the volatility is much more substantial than, you know, 5, 10, you know, 20 years ago. Yeah, that's really interesting. And I want to pick up on something you mentioned, Salina, earlier. You talked about the energy transition. You just mentioned the energy transition. I wanted to kind of just interrogate, I think, as they would say in academia, that use of words. Because, as you say, Jason, if you look at some of the decision making that was taken by European oil and gas companies over the past kind of five to 10 years, there's clear value destruction there, as you say, that figure,$80 billion.
12:30dollars. And I think part of the reason for that was that kind of conceptual framework about whether there's a transition happening or not, in the sense that there was kind of an inevitable transition and a progress away from fossil fuels towards low carbon energy, particularly towards renewables. So you could say, oh, well, the words we use, whatever we call it, don't particularly matter. But on the other hand, as I say, I do think it kind of creates a mental framework that then actually helps guide a lot of decisions. So, Jason, how do you feel about that language? And do you like to talk about the energy transition?
13:02Is that something you think is a useful kind of mental framework for what's going on? I think what the way we tend to look at, and this has been pretty consistent with WoodMac, you know, for the last, you know, five to 10 years, is I do think WoodMac has consistently said that we need every ounce of energy we can get. You know, there's going to be constant need. There's an insatiable appetite of new things that need new energy. You've got still 7 billion people that are below, you know, kind of modern usage of energy. You know, roughly 2 billion people that are still using firewood right now as their main source of energy.
13:32And then an additional 3 to 5 billion above that that are still not able to use full optimal amount of energy or electricity to power their daily lives. And so this demand is obviously coming online in the coming decades and will really serve to drive further increases in energy demand. And so, you know, we've increasingly started thinking about the word energy evolution, that it's not a light switch. It's more of a dial, if you will, and that there will be continued usage of oil and gas products for many, many decades to come. But also, we need to see more renewable type generation going on as well.
14:11Yeah, no, I agree with that. I think over the last five years, maybe five to 10 years, the thinking has evolved, right? And so energy evolution, energy addition, whatever you want to call it, energy growth is tied to GDP growth. And so if you break that out into how some of the myths have been busted over the last five years, it's kind of like what you said, I think people have aligned on the fact that it's not a switch, and you can't just switch off hydrocarbons, right? But if you take every commodity, so oil demand has slowed down or is slowing down. Roughly one to two million barrels a day growth over the last decade, it's probably looking at like 0.5 million barrels a day going forward, right?
14:53So it's growing but slowing. Gas demand, obviously, with everything you've mentioned on AI, but also onshoring of manufacturing, electrification, and more just cooling needs with climate change. That's gas demand. There's a ton of gas bowls out there as well. And then on the new energy side, I think it's interesting to think about what are some of the near-term technologies that are going to work? And then what are some of the technologies that are on the horizon that are probably more like a 2035-ish sector? So for instance, solar and wind are obviously growing. Battery storage is great and is needed if we want to use more solar and wind.
15:34And we can talk about all of these as well. I think some of the capital that went into the clean technology side of things was probably chasing hype, was probably learning as well. So green hydrogen, for instance, not sure where that's going to go. But then nuclear is definitely going to be needed, right? And that will be more of a 2035. So I think there's more of an understanding of the fact that energy is really complex. We're going to need all sources. Some are more near term and economic right now with and without subsidies. And some are more of the long term, mid 2035-ish. I think in many ways, you know, I think we tend to think of the world as energy addition or energy evolution.
16:18And one of the reasons is kind of the bouncing of those four really key, you know, pillars. I think, unfortunately, there was kind of an oversteering on one of the pillars, which was climate. Climate is super important. But when you look at energy demand, it's not only AI, it's also the 7 billion folks in this world that are living, you know, in levels of poverty and trying to, you know, bring them up to a higher quality of life. But even when you look at regulatory, I think we all understand now that national security has become a big, big component of that regulatory piece. And so now I think what's really complicated the matter is trying to balance those four areas.
16:53And the other thing I would just add about what's made energy transition a little bit more, I think less of a relevant term is the integration of all the energy components. They're all integrated. It's not, you know, it's not siloed. And I think the energy transition view was everything's kind of siloed. And now I think everyone sees that the impact and the interdependence of each of the energy generation capabilities or oil and gas is all interrelated. And so we've got to think about it in a much more kind of holistic, integrated view rather than in separate piece parts. And I'll just add to that.
17:25I think there's regional differences as well, right, Jason? So, for instance, I think, you know, the over-indexing of climate probably was more of a European side of things. China is definitely focused on energy security, right? And so it's looking at a lot of the critical minerals work, a lot of the magnets that they're building or manufacturing capacity that they're building is more from an energy security perspective. the U.S. has a really interesting opportunity being resource-rich and also having companies that are responsible producers and so wanting to do the right thing on the climate angle as well.
18:03So I'll just add that there are regional differences there. Yeah, no, that's very true. That's a really important point. As you say, those differences are very significant and persistent. Just going back to something you were just saying, Jason, very much resonated with a lot of the conversations I've been having. You said you were talking to people in the US about, for instance, integrating solar and gas. And that seems to be a real kind of buzz thing now. Just I've lost count literally of the number of people that I've been having exactly that conversation with in the US just over the past couple of weeks in terms of, apparently, the new buzz phrase.
18:40I think it's the integrated energy plan, but this is a thing that a lot of people now interested in, as you say, these kind of developers who, up until very recently, probably, you know, up until kind of nine months or so ago, would have been very strictly kind of solar and wind and storage focused, are now saying that they need to offer natural gas as well as part of the package, given, you know, what the market wants, given what the policy environment is now. You know, this is something which absolutely makes sense, as I say, in terms of kind of where the market is moving, not always easy to deliver, particularly not if you're not super experienced in that field and you're kind of learning about it for the first time.
19:18And yeah, that's just one very interesting little kind of micro example, as you say, of not so much sort of, you know, transition and kind of leaving one thing behind and moving to something else, but actually kind of trying to integrate the different technologies together. Yeah, absolutely. I think to your point, I've been in now several meetings recently in the power renewable space, and now everyone's trying to ramp up on being gas experts. So it's just really interesting to see in these conversations. Yeah, very true. So, Naina, I wanted to go back then to you on this and just to think about, you know, what that means in terms of decision making.
19:53And maybe just to tell us a little bit about your experience at Hess then. So this is an oil and gas exploration and production company, had an interesting record, if people know the history, but founded by Leon Hess way back, was originally in the fuel distribution business, diversified, then became a very, very successful upstream player. and in particular had this incredible success in Ghana and these massive fields, which are very important now to future oil supply. But there was always very strictly an oil and gas company. When you thought about the energy transition and where it was going, how did that affect your decision making?
20:37Did you ever, for instance, think about investing in renewables? Was there something or did you think about what it meant for the decisions you had to make in terms of the types of oil and gas assets you invested in, what did it mean to you? You know, Ed, when we started, it was about 2019, right? And there were a bunch of net zero commitments coming out. There was a wave of net zero commitments coming out. And Greg Hill, who was our president and John Hess, came to my team and basically said, you know, figure out net zero. What does this mean? So we took our time to understand this space because is we're really good at our netting, which is oil and gas, more oil than gas.
21:15And we're really focused on our portfolio and understanding the role of the assets in the portfolio, etc. And so when we started looking at net zero, a couple of things came to play, right? And we distributed sort of the solutions in three big pillars. How can we decarbonize our own operations? Are there technologies or additions in the adjacencies of our current assets that make sense for us to invest in? So new technologies like maybe CCS or even hydrogen, if that made sense. And then we knew that at some point there was going to also be a financial instruments angle to it. So offsetting what we can to decarbonize ourselves.
22:00And so when we started looking at each of those pillars, our decarbonization angle was the one that we could very systematically go after because it's just basically building out a marginal abatement cost curve with all of your solutions and then going after it from an operational perspective. On the CCS side, as well as some of the newer technology side, like renewables as well, we looked at a bunch of different things and we realized very quickly, and this is one of the learnings that we had, is to make a switch, and this probably goes to what you were saying, Jason, about the European majors as well, to make a switch of a business model, you need an investment thesis that makes sense, right?
22:42And we looked at our knitting and we said, can we actually move into some of these spaces? Are we, you know, what gives us the unique advantage to move into some of these spaces? What gives, you know, how are we actually going to create value down the line? How are we going to bring costs down the line? And then what portion of our portfolio can we actually invest in that space? And we very quickly realized that the most sensible solutions in order to decarbonize were solutions that were in adjacencies to our current assets because we were really good in the Bakken. And so we started exploring what potential carbon capture in the Bakken would mean.
23:23We've got Asia, which is high CO2 assets. And we looked at, you know, what does carbon capture mean? Again, carbon capture is similar in terms of skill generation, right, between oil and gas and CCS. CCS. And so, again, it's the adjacencies of the assets that we were looking at. Right. Really interesting. And actually another great example of kind of interconnectedness between sectors that you're in upstream oil and gas, but you realize you need to know about carbon capture as well. And that's something else you're going to be wanting to explore. So I wanted to talk a bit about data, Jason, because I know this is a big theme of yours, and you talk about bad data leads to bad decisions.
24:01Can you talk a little bit about what you mean by that and what are some of the big issues in terms of energy data as you see them? Yeah, I mean, certainly we can, you know, spend an hour on this, but I don't want to, you know, put your listeners to sleep. But I think what I would say is that when you look at the topic of data, we're in a point in time now that we have access to an unbelievable amount of data. And we kind of see kind of three kind of traditional mistakes made with data. But the first is just the lack of using the amount of data. So when you looked at historical ways in which decisions were made, humans just tend to think very linearly.
24:34It's cause and effect, right? And so the amount of data they collect is a very small amount of data and oftentimes misses correlations or interdependencies between data sources because the human eye just doesn't see that, you know. And so when you start looking at these machine-based programs, they just crunch massive amounts of data and they start recognizing patterns or interdependencies that humans, you know, don't see. And so when you look at the amount of data that's out there, you know, Wood McKinsey is arguably the largest proprietary aggregator of data in the market right now. We have the most proprietary data of anyone out there.
Read the full transcript
25:09And we see this as an arms race. And so we continue to do M &A to buy more and more data sources. And we see the data sources out there. And what you can see now is beyond kind of traditional ways of gathering data. You have everything from drones to, you know, in our case, we have the largest sensor network worldwide where we track energy movements, power lines. We now have sensors now on hyperscaler data centers tracking energy usage. So this is really an incredible amount of data that's out there, whether it be drone, satellite, and sensor data that's being put into, you know, instrumentation with IoT, etc.
25:46So that is clearly one kind of thing is the more data, the better. But the second thing is sometimes we see an over-reliance on synthetic data. And that's probably a little bit worth more exploring is that now every vendor out there or company says that they can use machine learning to estimate data, which I call synthetic data. It's really not actual data. And you find out it's only as good as its source. And you can't just trust that a vendor who says they have modeled this data is actually real data. I mean, I gave the example of sensor data. When you look at congestion, there are obviously ways that you can potentially approximate congestion via modeling.
26:27And it's much worse than actually having the true sensor data. Another one that we see at Wood McKinsey is when you look at valuations, Wood McKinsey has values on every energy asset in the world, whether they be solar, wind, oil, gas. But one of the biggest impacts of valuation on energy assets is really the tax and regulatory treatments from those local authorities. And that's not something you can estimate via, you know, a model. It has to be collected through human intelligence or human relationships. So I'd say that's the second kind of big mistake. And the third one really is the lack of integration.
27:02You know, as we talked about the book Connected, the data is connected and understanding how the data you collect connects, whether it be, you know, power data has to also understand what the impact of that is around EV or oil or gas. And that's another kind of, I think, fallacy or mistake that's being collected to data still is being collected very siloed. That's very interesting. So Nina, how do you think about this? So you've been a user of data, you're a user, you were a user in one way running strategy for HESS. You're going to be using data presumably in different ways, but not completely different ways in your new role at Bernstein.
27:42What do you want from data and what are the kind of the big problems that you encounter when trying to use data? From a strategy perspective, from within these energy companies, strategic planning in energy is actually all about data. And the scale and complexity is so interesting to understand it because we weren't just forecasting oil demand. We were linking it to associated gas, if there is associated gas. What does that do to gas prices in the U.S.? What does that mean for, you know, power markets if we're looking at things like BPPAs or, you know, to decarbonize? Just on the oil side of things, there's also EV adoption, which means critical minerals and, you know, even forecasting physical climate risks, right?
28:34And so for us, data can absolutely help in so many different ways. I think AI can help in two ways for strategic planning. It's what Jason said. It can process vast interconnected amounts of data sets really much faster than any human can because we don't have the mental ability to keep so many nodes of uncertainty if you imagine a decision tree. You can also respond to policy changes really quickly. So now if we use scenario analysis, for instance, if you have future scenarios that you're looking at, you can update that scenario analysis really quickly. You're not waiting for this, you know, insane Excel model that takes six minutes to open and that two people in a room can sit and do, right?
29:21You can have basically a reservoir engineer in Guyana, a data scientist in Houston, and a strategist in London all looking at the same scenario analysis. And that's really useful. I think that's the second piece, which is it breaks down some of the siloed decision making. I think the biggest piece that we saw with the benefit of using extended data sets is bias recognition, right? Because if you're using a small data set, you're essentially propagating that bias into your decision making. And so bias recognition was an interesting way to think about why data sets can help. Not using too optimistic type curves can really help with not over, you know, overpromising production data, right?
30:06And what it can do with unstructured data is really helpful as well. So whether it's monitoring flaring data from various sources, there's a bunch you can do there. I think the only thing I'll say is the key point is AI can be very useful and data can be extremely useful. It doesn't replace judgment, though, right? So it augments it. And so it helps to elevate everyone in the organization to do more with the data. And in this, I think you guys saw the article in the journal today, you know, another super major is letting people go. So if we're going to be in this new world of using less people to do more and keep the same level of production and bring costs down, companies that are going to be differentiated are the ones that are really making use of the data set correctly.
30:55Sunaina and I had a chance to meet almost a year ago when I first started. And I really think Sunaina had one of the most advanced teams that we saw of any of the large majors that we met with. And obviously, it's a credit to her and her team. And I think, obviously, when you start looking at smaller producers out there or even, you know, a small developer in the P &R world, they don't have access to that team of folks. And so, obviously, it becomes, you know, how do they actually leverage the power of AI to make better decisions and better planning? And maybe just to piggyback further off of what Tsunada shared, you know, when we look at AI, an entire chapter we wrote in a book around this, but we kind of identified kind of four major changes of AI.
31:36And Tsunada kind of identified one of them, which is the enhanced compute power allows you to run scenarios. You know, you're talking instead of dozens of scenarios, hundreds of scenarios, millions of scenarios, and allows you to run them in real time. So in essence, instead of taking weeks or months to make changes and outlooks, you can now do it in days and if not hours and hopefully someday real time. I think we found this very important, particularly when we were meeting with the clients around the tariff changes that were going on. They were changing, you know, obviously very rapidly. And how did that affect the planning process?
32:10But that's definitely one of them. The other one that we felt very strongly about is having an integrated outlook. You know, it's not just one model. You know, when McKinsey has 1 ,100 models, you are going to need different type of models for oil and gas or the local markets and power, even within ERCOT or, you know, JPM. They had different behavior characteristics that may require different models. The question is then how do you integrate this model into an ensemble, into an orchestra? You know, if you have all these different, you know, models, which are instruments, how do you have that orchestrator that pulls it all together?
32:44And that is obviously where I think AI starts playing with large neural networks. The other two areas, just to quickly mention, is also prompt-based AI is a game changer. We have an entire chapter based on this. I think it was affectionately called the tyranny of the propeller heads, is that often too often that the modeling is actually done by one person in a corner office or in a corner room or corner cubicle. and now with prompt-based AI, you can actually have executives engage with that model real-time asking very simple questions like, what is the EV demand going to be if the price of oil drops to this?
33:21You don't have to ask a modeler to then run the scenarios and three days later it appears. It's all done real-time. And the last thing I want to maybe spend a second on, which is a pretty nerdy topic, is hyper-modeling. So I won't go too much in depth and our CTO could give a course on this, But the concept is, historically, modeling was done linearly, you know, via Excel spreadsheets. What increasingly you find is that the world is much more probabilistic, you know. And as the world is problemistic, it opens up now all sorts of different models. And there's an infinite number of models you can use, tree models, reoccurring neural network models, you know, TFT models.
33:59What you find is that you're probably going to end up using different models for different use cases or scenarios. But what hypermodeling allows you to do is now you can use AI to test these models real time and run scenarios where you can actually use these AI to actually find the best model for you that best, you know, predicts the outcome the best. And then you can actually use an AI suggested model to then be one of those, you know, instruments in the orchestra, which you then pull together into a full sabo. And that's a game changer. Historically, you would have to test each model individually, manually.
34:35And now you can run these compute, massive compute over, you know, millions of scenarios and then find out which model best represents that particular scenario and then pull it into a final orchestra. So as I said, it's really fascinating what's going on and it's happening real time and really game changing. Yeah, that is really cool. I have to admit that was a new word to me. Hypermodeling is not a concept I've come across before. So thanks for that. That has been educational. And as you say, very, very exciting potential. So where are we with that then? Is there something we are using now, something we will be using in the near future?
35:09Where's it got to? Ed, absolutely. AI is a substantial strategic investment for Wood McKenzie. When I first started, we kicked off this initiative called Synoptic, which is really an umbrella term for all of our AI initiatives. We have multiple dozen now initiatives where we're embedding AI in our products to provide better insights and better predictive capabilities. Also, we've been adding some incredible capable people, our CTO, Bernardo, incredible, you know, PhD in computer science and also chief architects and, you know, literally best and brightest in machine learning. And all of this is really to help drive better tools and insights, which will make our human researchers even more capable in providing better insights to our clients.
35:53I think we're just about out of time. We should be wrapping it up soon. But just before we do then, so Jason, look, if people have been listening to this and people have been listening to you and find it convincing, what should they do? And what is your kind of advice to people then? If you are a decision maker in the energy industry, let's say, how should you be responding to this changing world? Yeah, you know, first of all, I would just encourage them to go to www.woodmac.com and download the book, you know, and it's a light read. It's about 90 minutes, and I think you're going to really enjoy it.
36:24It's a lot of really big, heady ideas that I think are a bit provocative and will definitely, I think, get you to think a bit differently. But, you know, outside of reading the book, the way we typically think about things is kind of what we call the winning trifecta, which first and foremost is what Sinan and I just talked about, which is people. Like, you have to have the best people. This is not, you know, simple stuff. And finding the foremost experts is key. I think the second one, obviously, we talked a lot about data and using big data to help make better decision-making. And the third one is leveraging tools like AI to helping that decision making.
36:59And the last kind of, you know, part of the trifecta, which really kind of the big thing that captures everything is that it's got to be integrated. We would maintain, and I give this analogy whenever I meet with a customer, is the way you engage with your doctor and you try to get medical advice is you could try to go to specialists on everything. Or you could use a family doctor or use a concierge doctor. Someone that can provide you a holistic view on things, can recommend specialists when necessary, but provide you that holistic integrated perspective. And we would recommend that's how you approach the energy world.
37:33Right now, the world is too siloed. There's too much of a reliance on specialists. And I would advocate this is not just at a corporate level. I'd also argue this is also at a government level, kind of an over-reliance on a lot of individual industry folks that have certain, you know, kind of policies they're trying to advocate for. So we think an integrated holistic partner that can help you interpret and bring a holistic view is directionally a huge part of the advice that we'd recommend going forward. Yeah, that's a fantastic point. And as you say, it's very possible to see, I think, that that kind of truly comprehensive understanding to kind of see all the different parts, see how they fit together, is sorely lacking in a lot of places.
38:15And definitely kind of people could benefit from seeing that picture more clearly. Salina, final thought to you, just in terms of that aspect, or just in general, in terms of what you think energy companies, what energy decision makers are going to need to do to be successful in this changing world. What do you think? What's it going to take to kind of deliver the results that people are going to want to see? Sure. I think I'll just double click on something that Jason said. And I think one of the things, you know, we talked about scenario planning and scenario analysis a little bit. I'll just probably, you know, talk about that a little bit more.
38:54And I think one of the things that's going to really help companies is understanding the role of the assets, right? Because the benefit of scenario planning is not about predicting the future. It's looking at a plausible set of futures, right? And if we do that at the asset level, and you understand the role of each asset in the portfolio, right? So scenario analysis on the asset level, and then you take it up and you do the scenario analysis on the portfolio level to see how the entire portfolio behaves collectively, then the real value comes from sort of asking, you know, across scenarios, what adjustments are you going to make on the portfolio level to achieve the best possible outcome, right?
39:40Because I think one of the things that energy companies, you know, are challenged with, and Ed, you and I talked about this offline a little bit, is this short-termism, right? You're trying to basically, you know, end up trying to meet every quarter, targets every quarter, but you're also trying to predict, you know, what life is going to be 10, 15 years from now, so short, medium term. And so the understanding the role of the assets and then how they play towards the portfolio, that's really your strategy. And then the quarters just become a piece of communication, right? You communicate what your strategy is, and any missteps that happen, you communicate why those missteps happened, and you explain the role of each of the assets as part of the communication on the quarter, right?
40:29And I think that exercise, along with all of the data and the hypermodeling, we never did hypermodeling, it has, but all of the hypermodeling that Jason talked about, it really allows you to say, are there any clustered risks at the portfolio level? Are there any options that exist at a combined level? So it really, the power of scenario analysis just really gets enhanced with the data and AI tools that we now have. The companies that thrive are probably the ones that are going to embrace some of this uncertainty. It's what Jason said, everyone talks about uncertainty, but it's how do you actually model out plausible futures so that when things change, you can quickly adapt, right?
41:13It's hydrocarbons for resilience, but also low carbon technologies for growth and then using AI for quickly moving towards a new future. And I think winners are going to be folks who manage across all three of those pieces. Jason, did you want to say anything else? Final for them. Oh, you know, I think Sunita, you know, captured a lot of the sentiment, you know, and kind of your point, you know, Sunita, about ultimately, you know, kind of becoming successful in an uncertain world. Part of it is getting better at predicting, but part of it is also getting more agile, right? So in this world of a lot of volatility and change, being able to be very agile and not just decision making, but operational is going to be key to future success.
41:58All right. Thanks. And final, any final thoughts? Anything else we haven't talked about that you wanted to raise? Yeah. I mean, I think we obviously have talked about the need for better predictability and better scenario planning and better agility. But I do think there's a positive, there's a glass half full opportunity, which is, you know, there will be well over$75 trillion invested in this energy evolution over the next several decades. And that presents massive opportunity for investment growth, increased IRR for those that actually get it right. So I think if we embrace many of the topics that were discussed in this podcast and in the book, I think it really creates a massive opportunity for wealth creation for companies, individuals and nations.
42:42Yeah, as you say, an enormous amount to do. And I'll add to that by just saying, I think this is actually the best time for this sector, because some of the capital that went in was pretty inefficient over the last five to 10 years. And it came from, you know, various private sources. So philanthropy, family offices, etc. And now it's starting to move to public. And so, and I think there's a better understanding of the economics as well without the subsidies. So it's a great time. And I agree on that completely. Absolutely, yes. I remember when I first started getting into energy, whenever it was sort of 20 years or so ago, I found it an incredibly exciting industry then.
43:24I think it's even more exciting today, as you say, for a lot of the reasons you've been talking about. So unfortunately, we do have to leave it there, but it's been great talking to you. Many thanks, Jason. Well, thank you very much, Ed. Thank you, Sunaina. Yeah, many thanks, Sunaina. Yeah, thanks. Thanks, Jason. Congrats on your book. Yep, absolutely. Congrats on the new role as well. Thank you. All right. Thank you. Hope to talk to both of you again soon. Thanks very much to our producer, Dan Cottrell. And above all, many thanks to all of you for listening. We really do value your feedback, so please do keep that coming.
43:55And we'll be back soon with all the latest news and views on the future of energy. Until then, goodbye.
From the publisher
Host Ed Crooks talks to Jason Liu, Chief Executive of Wood Mackenzie and co-author (with Chief Analyst Simon Flowers) of a new book, Connected, about the fast-changing world of energy. They are also joined by Sunaina Ocalan, formerly Senior Director for Corporate Strategy & Climate at the oil and gas company Hess, now Senior Analyst and Co-Head for Americas Energy & Transition at Bernstein Research. Together, they explore how energy leaders can plan, invest and operate operate in a world where different sectors, technologies and geographies are interconnected in more powerful and complex ways than ever before.
They talk about the language of “the energy transition”, and whether it can lead to misconceptions. Global demand for hydrocarbons is still growing, and they will continue to play a critical role in our energy system for decades to come, even as new supply from renewables and other low-carbon sources surges higher. A wider appreciation of that reality is driving a shift from siloed thinking about individual sectors to integrated solutions. For example, companies are increasingly looking at pairing solar and storage with gas generation to meet demand from data centers for reliable low-carbon power.
Sunaina takes us inside the the thinking of energy leaders as they assess strategies and investment decisions. She sets out a practical approach to scenario analysis, with “exit ramps” so companies can pivot as facts change. The aim isn’t to predict one future, but to be ready for a range of possible outcomes. That means balancing the advantages and disadvantages of a wide range of technologies, and taking a strategic view through short-term fluctuations as far as possible.
Effective decision-making is impossible without reliable data. Jason warns about three traps: using too little real data, leaning on synthetic/modelled data without ground truth, and poor integration across different sectors. Data collection technology is advancing rapidly, and with sensors, satellites and market intelligence, decision-makers can increasingly see what’s really happening with precision and granular detail, often in real time.
Then there’s AI. Like other industries, the world of energy is being transformed by the tools that have become available over the past few years. Scenario runs have been cut from months to minutes, with hundreds of models combined to give a comprehensive coherent picture. AI tools can even assess the best models to use on particular data sets: a capability Jason calls hyper-modelling. And still there is a vital role for human intelligence and judgement, to find and interpret the information that the AI tools miss.
The challenges in the energy sector today are vast. It is a cliche to say that uncertainty is higher than ever, but today it genuinely seems true. The pace of innovation in AI is changing the world in ways that have never been seen before. But the opportunity is vast, too. The energy industry will need $75 trillion or more in investment over the next 25 years, to meet ever-growing demand while reducing the impact on the environment. The businesses that succeed in making the most of this opportunity will be the ones that get three things right: the right data, the right AI capabilities, and the right people, all brought together to deliver actionable insights.
Let us know what you think. We’re on X, at @theenergygang and Bluesky, at @theenergygang.bsky.social. Make sure you’re following the show so you don’t miss an episode.
See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.
