Batteries included: Building and operating sustainable gigafactories

As economies move toward more sustainable transport options, more electric vehicles (EVs) are rolling off production lines than ever before. These vehicles need to be powered by lithium batteries, which are built in specialist facilities called gigafactories. With more than 30 planned in Europe alone, companies are working fast to develop the construction and operating playbook for these highly technical sites.

This episode of McKinsey Talks Operations, featuring Stefano Piscitelli, vice president of group construction at Northvolt, and Evan Horetsky and Mark Bakker, a partner and associate partner, respectively, at McKinsey, looks at how demand for batteries has changed; the opportunities to use digital twins in the planning, construction, and operation of gigafactories; and the role of a new approach to construction to ensure that both the factories and their products are sustainable. An edited transcript of their conversation follows.

Daphne Luchtenberg: McKinsey Talks Operations is a podcast where the world’s C-suite leaders and McKinsey experts cut through the noise and uncover how to create a new operational reality. Each episode considers the challenges companies and economies are facing and the opportunities leaders can seize for competitive advantage. We’ll look at how to connect the boardroom strategy to the front lines, ways to increase performance, and where and when to infuse operations with technology. And last, we’ll discuss why empowering the workforce with skills and capabilities is key to success.

There’s a green revolution happening in Europe and across the world. And one of the major shifts leading the way in that revolution is the move to electric vehicles. In 2021, some 6.5 million electric vehicles were sold, which was up almost 110 percent from the year before. And with all these new EVs rolling out into our roads, they need batteries to power them. That’s a lot of batteries, and therefore a lot of factories, known as gigafactories, to make them, with 30 planned in Europe alone.

This episode of McKinsey Talks Operations investigates the role of gigafactories in the global economy’s move toward reducing carbon emissions, and why construction of these sites is just as important as the products they produce in achieving net zero. Joining me today are Stefano Piscitelli, the vice president of group construction at Northvolt; Evan Horetsky, a partner in McKinsey’s Stockholm office; and Mark Bakker, an associate partner in our Amsterdam office. Great to have you all here. Welcome.

Evan Horetsky: Thanks, Daphne. Yes, it’s incredible to see the need for energy storage as the world turns over to a decarbonized industry, to a carbon-neutral industrial base. I mean, when solar and wind gets installed on the energy grid, or as electric vehicles launch en masse into cities, you need a lot of batteries. Specifically, we measure this production of batteries in gigawatt hours, or the amount of energy that is stored in each of those batteries being produced. And in 2014, this number was something like 35 gigawatt hours per year being produced in the world. That was pretty big. That was 20, 30 production lines of batteries. Now, the projection into 2025 and beyond is over 3,000 gigawatt hours per year, or basically 100 times that need, and it’s all because industrial growth and product growth will need these batteries to really drive it. The gigafactories themselves are front and center of this impressive industrialization needed.

Daphne Luchtenberg: Yes, fascinating, isn’t it? And Stefano, why does every battery manufacturer require such specialized sites? It’s not an easy challenge, is it?

Stefano Piscitelli: Daphne, thank you for having me. Well, absolutely. Gigafactories are simply the answer to this incredible and continuously increasing demand for batteries that Evan was mentioning. And why is it so? First of all, it’s their capacity to supply what we call gigascale.

Since I joined the company three years ago, we have continuously been revising our plan upward. And if we look only at our first gigafactory, originally we were supposed to provide 32 gigawatts, and now it’s at 60. This is also because the economy of scale is what actually enables us to meet our ambitious cost target. If we look at cells, they are a substantial part of the vehicle cost. And despite cell pushes and subsidies that drive the sector, for the full transformation what we really need is to ensure that batteries are also competitive on the market and building at scale fast, and to continuously reduce capex [capital expenditures] to actually allow us to get there.

Daphne Luchtenberg: Fantastic. If we are thinking about some of the challenges in getting the planning and building done and delivering a gigafactory, how would you say it differs from other major capital projects?

Stefano Piscitelli: We are building the plane as we fly here. The technology is continuously evolving in an effort to being always more competitive and more efficient—how many “giga” you can fit in a building, for instance, or what is the perfect balance between the footprint and output in terms of giga power. It is very agile and fast. And the construction industry is normally very traditional, very risk-averse. So, this is definitely a huge challenge, because we don’t need to transform only this sector; we also have to transform the construction business.

On top of this, we have seen that the huge challenge is the raw material, and how we access what drives the basic element in a battery. In this case, I can share a specific trait of Northvolt, which is the vertical integration. We don’t only build the factory but also construct the active-material plant and the recycling plant. This means that, on the same site, you can have very different technology and need very, very different capacity, skills, and competencies.

Daphne Luchtenberg: Wow, it’s really a complex challenge. Evan, what can the role of a digital twin be in helping organizations navigate this and also anticipate some of the potential risks?

Evan Horetsky: I want to build on a bit of what Stefano said. They’re building the plane while they’re trying to fly it, right? Why is that so? Well, battery manufacturing for lithium-ion batteries is kind of an old industry. For at least 40 to 50 years, these lithium-ion batteries have been made at scale. However, everyone is having this challenge of designing and iterating on new cells and new equipment on new gigafactories recently because, like I said before, this amount of scale had never before been seen.

Lithium-ion cell manufacturing is really, really, really hard. It was hard back in the early 2000s, and it’s hard now. And when you add this scale, when you add this need for speed, how do you get to 3,000 [gigawatt hours] when we fought so hard to get one factory up with 35 gigawatt hours per year? You run into kind of compounding challenges and compounding complexities on, first, the product to the production technology, which people call design for manufacturing. It’s the way in which the building might be wrapped around equipment, the way in which you estimate the utility or the electrical and piping needs. And then the way in which you turn on these lines and try to get materials and people and lights on in the shop itself, to start making cells.

To put it in perspective, Stefano said 60 gigawatt hours per year; that could be two million or more cells a day. We’re talking hundreds per second coming off of these lines, and it’s many lines in parallel. There’s a point in which I think it’s quite obvious that you need a very well-thought-out and very lean digital backbone to control all of this. And there’s been, for 20 to 40 years, the talk of Industry 4.0, digital twin, factory of the future kind of buzzwords, but what does this really mean?

This really means that for battery manufacturing, you need an idea of how you put into a database the exact characteristics of the equipment, and the changes to your equipment that you’re putting in this factory. Then you use this model or this compendium of information about your factory equipment—what’s really important, the things that make the cells—and you use that to connect to each of your other design and planning efforts: How many pipes do you need to connect for this equipment to turn on? How big should the beams be on the structure for this? It’s very important to get the first thing right, the equipment simulation, and early on, maybe two or three years before you need to start up the factory and then build off of that.

Almost all other players right now are building digital twins in silos. I think, also, because it’s such a sexy topic, that construction teams might have a building-information model, a digital twin of the building itself. It’s separate from the equipment digital twin, it’s separate from the material control tower or supply chain link that people are using to find the materials. So, instead, what we need to do moving forward is this interconnectedness.

Stefano Piscitelli: I perfectly agree with Evan that the temptation is to work in silos, but the real benefit is to connect them all in one unique blueprint that can connect cross-functionally all the different disciplines and the work, and at the end of the day generate not only the capex but also the operational cost of the factory.

Daphne Luchtenberg: And Mark, Stefano has been talking about building the plane while we’re in flight. But a playbook is emerging. Getting it wrong is such a high risk and has costs associated with it. Can you talk a bit more about the playbook that’s emerging?

Mark Bakker: Great to be here, Daphne. Building on this need for cost-efficiency and using things like digital twins to drive down costs and increase the speed of getting these plans ramped up: maybe, to start with, when we look at a lot of these plans globally, and we look at plans that have been constructed and those that are under construction, we see a very large variation in capex cost—that means the investment cost per gigawatt hour. We see ranges of three times the cost between one plant versus the other. We also see big ranges in the timeline of getting these plants actually into operation and ranges in the opex, or the operational cost, when they are ramped up.

Some of this can be attributed to location. Some locations simply are cheaper to build than others. But it doesn’t explain the full range. There are a lot of investors and owners of battery plants that are now trying to really optimize the way that they roll out these plants, build these playbooks, these blueprints, as the final set, and really leverage the learnings that they have internally to get faster production up and running at a cheaper cost plant by plant by plant. And a digital twin is, of course, one of the tools to do that and also to capture these learnings centrally in the organization.

Daphne Luchtenberg: And Evan, Stefano has talked about the complexity of these operations, and the need for very specialized skills and talent. Can you talk a little bit more about that?

Evan Horetsky: If you dive into the gigafactory, what do we mean when we talk about making cells, in terms of process? This is incredibly hard, and I think one of the main reasons is because, unlike automotive manufacturing, pharmaceuticals, or heavy industry, making a battery cell is not one type of factory. It’s like multiple, independent specialties and types of factories under one roof. First, you are making this foil—people call it a roll-to-roll process. It’s like you’re making paper, or maybe you’re making films, but now the films have a metal on them and the films themselves are made of a different metal, which makes things quite complicated. Then you have assembly, or more like a micro-controlling, high-speed pick-and-place. Or a micro assembly, like you might have put together a cell phone or a product that you own.

Then you have formation or characterization steps at the end of your battery factory. These are the steps that set up the cell to be long lasting. This is almost like a data center; it’s a bunch of cells with a bunch of electronics being charged and discharged and heated up and cooled down for sometimes two weeks at a time. It’s quite complex. You need very specialized skills, you need people who are electronic technicians, you need people who know these high-volume, automatic micro-control systems, you need people who know roll-to-roll conveyors, and you need people who know metallurgy and fluid dynamics of these metals in solvents. And people are on a hunt for this labor, because it’s quite limited, because this scale has only just started in the industry in the past ten years.

Daphne Luchtenberg: As we talked about the plan in Europe, I think we know of 30 sites that are planned there. But where is the need greatest, Evan?

Evan Horetsky: We’re going to need hundreds or well over one hundred of these factories worldwide. Greater China and Asia have been leading on the number and size of gigafactories, and they will continue to lead. With the Inflation Reduction Act in the US, there is a huge and urgent momentum for players throughout the world to move into American manufacturing of these gigafactories. I think that’s a great thing. Because the consumers are mainly in China, America, India. So, having a manufacturing base in the US is a great thing. And then Europe has a consortium of needs. Mostly in Germany, France, and Spain, you’ll need these batteries, but you really have this need spread out across the European regions. And the Nordics are bubbling up to be a bit of the Silicon Valley of sustainability, and they are really building up an ecosystem for these battery factories and other renewable and sustainable technologies. So we’ll see a large spike in [the demand for] talent and factories coming online in the Nordics as well.

Daphne Luchtenberg: Mark, is this understood by the investment community? Is there more to do there to ensure that the investors meet the opportunity?

Mark Bakker: I think that, for sure, there’s a lot of interest from investors. Is it sufficient? That’s the big question. Right now we see billions of private investments going into companies like Northvolt. The big question—I’m sure Evan can go deeper on it—is whether that will be sufficient to drive the huge need for these plans moving forward.

Evan Horetsky: There has been a series of private equity investors that have led rounds in Europe that have been quite impressive. Northvolt, I think, was one of the top private-investing rounds most recently, when they got over a billion dollars in funding to help launch these gigafactories. So, you see investors in Europe, a lot of these consortiums in the Nordics, they are investing heavily into sustainability tech and into gigafactories. The US, it seems, was maybe either a bit more skeptical for longer or was looking for the right incentives, like the Inflation Reduction Act, to really go in. But we’re seeing an amazing upswing in investment in North America as well now.

Daphne Luchtenberg: Let’s change tack a little bit. When we have these gigafactories and once they’re in operation, how green are they really and what are the opportunities to limit their environmental impact? Stefano, first to you.

Stefano Piscitelli: It requires a tremendous amount of energy to produce batteries. If you ended up taking that energy from coal or fossil fuel, then you’re not that green after all. The key revolution is to ensure that you use renewable [energy] to feed your gigafactory. This is one of the main points that we have been pushing quite hard. Where do we find energy that is renewable and in that amount? There are not many sites. We managed to secure one in the north of Sweden, but Sweden, especially the north, is pretty rich in terms of wind and hydro plants.

If you look at Germany, the mission has been a bit tougher. We had to go to the north and look at where the energy is coming from wind farms and go to those locations. This means that, somehow, you end up finding locations that are perhaps not the best for, let’s say, access to your customers, or to materials, or to contractors, or to the right skills and competences. But it’s definitely an essential factor if you want to be green. Then it’s the supply chain and what you really put into the battery. And if you don’t get the proper traceability of the raw material, then your carbon footprint at the end will not really be under control.

Daphne Luchtenberg: Mark, Evan, anything to add there?

Evan Horetsky: Yes. I think Stefano said earlier, if the cost of an electric vehicle is primarily in the battery, the carbon emissions of the electric vehicle is overwhelmingly in the battery. So, we need this battery to have a low carbon footprint, right? This includes logistics to get material to the factory and to get the cells out. This is using all renewable electricity; simplify and delete anything you can. I think the battery pack or the battery itself should be simpler. I think the factory should be simpler; you should fight to use less pipe, use less electricity. This is where smart engineers, smart talent coming up, can really make a difference on sustainability and on faster scale to battery production.

Stefano Piscitelli: Evan, going back to the digital twin, it’s another tool that we know can help to predict the utility consumption of these processes. So, getting a better grip and control on how much utility these processes will end up consuming will definitely help not only the capex but ultimately also the carbon footprint of the gigafactory itself.

Mark Bakker: I think when you talk about capex and carbon footprint, there’s a lot of attention on the final operating process of the plant, and how green is the battery itself. Another angle that will come to the forefront more and more is, how green are you actually building your plant itself? What materials are you using there? Are you optimizing the use of green materials in your factory itself, where you build your green batteries? I think that’s the next step that can reduce this carbon footprint even further.

Daphne Luchtenberg: Yes, fascinating. So, what is next for gigafactory development? It sounds like we’ve only just started.

Mark Bakker: There’s a lot of things happening. With the bigger and bigger maturity that we’re getting in how to build these gigafactories, I think you will see a whole landscape forming, with certain EPC [engineering, procurement, and construction] contractors, certain OEMs, and other suppliers really getting used to this landscape, getting specialized in it. There are new entrants, for instance, in the equipment markets coming in. And all of that means that the whole supply landscape becomes more mature and can deliver these plans more effectively.

This is also about the relationships in this market. What we see is that certain battery manufacturers are getting long-term relationships with suppliers and EPC contractors, plant by plant, to jointly learn and to jointly improve on this. So, I think that is sort of professionalization of an industry, which a couple of years ago was still trying to find out how to build these plants most effectively.

Stefano Piscitelli: Yes, no doubt. That’s the key. If we spend time discussing digital twins, design, and engineering, then next is to really go to work on a full value chain of all the partners that contribute to a gigafactory. We had to put tremendous effort into the company to deliver our first [gigafactory], and now we have to deliver two, three, four. There is no way we can do it without a proper network of trusted partners and suppliers. And where do we find them? That’s super hard. As I mentioned, the construction industry is, by tradition, very risk-averse and not used to the speed and agility that we need.

Daphne Luchtenberg: Yes, fascinating. It sounds like we also need to change the relationships. This is where colleagues have been talking about the implications of collaborative contracting, for example, so that everybody has skin in the game, as you work toward these new, quite ambitious goals.

Stefano Piscitelli: That’s 100 percent correct. I believe that—especially when you are then chased by lenders and investors, and they are not really in favor of risk—then you might be pushed to try to transfer skin in the game and risk to partners, suppliers, and contractors as much as possible. But what is really key, if you want to build the partnership, is to align with this, let’s say contractor, and understand: How do we share the risk? Which risk should you own, typically productivity and interface management, and which risks should I own, typically technology road map and an interface with processes. And if we find the right balance, this is really what generates that type of win–win setup that will help us not just build the first gigafactory but partner to build factories two, three, and four at reduced cost and time.

Mark Bakker: When you talk about three, four of these plants that you build together, you can also escape from this eternal problem in the construction industry: that there is a focus on one project; that as a contractor, we have to get our profitability out of this one project. And if, for some reason, the incentives are not aligned or the contract is pushing risk toward us, it becomes sort of a battlefield of claims and counterclaims and change orders that start to increase the cost. By really aligning incentives across multiple plans, we see that you can escape that sort of competitive positioning and work together and both profit from it by working more efficiently together.

Daphne Luchtenberg: That’s great, Mark. And that’s needed, because we need to get this all mobilized quickly. Let’s talk about the nature of ecosystems and how they are emerging to kind of help shepherd this along. Mark, what are we seeing there?

Mark Bakker: I think the end-to-end value chain is becoming more mature. Maybe two things to mention: one is the long-term relationships with the EPC contractors. The other, which I find very interesting—and I think Evan can go deeper on that one—is the equipment producers, the machine producers in Europe who have not been active in the battery industry are now seeing this as a great opportunity for getting into this industry and therefore creating more capacity and also more competition, which should also drive costs down.

Evan Horetsky: Absolutely. There are regional ecosystems developing where the entire value chain—renewable-energy supply, raw material, battery cell manufacturing, recycling—is coming together. This is very exciting. And then there are ecosystems building around some of the key bottlenecks. First, raw material, labor, and then equipment. I’m really excited about the equipment innovations I’m seeing. In some cases, a piece of equipment that was producing, let’s say one times the batteries a year ago, there are players from other industries, paper pulp, tobacco making, etcetera, that might be looking at equipment that’s making ten times those cells in just four years.

Daphne Luchtenberg: Stefano, how would you summarize where Northvolt is at the moment and your prospective future?

Stefano Piscitelli: That’s a very interesting question. I believe what we are doing is to build on the four pillars that we have discussed during this interesting conversation. So, we’re really pushing the engineering to find the balance between flexibility and capital efficiency. We are working very hard on strategic supply chain and partnership with contractors and keeping the manufacturer. We are definitely working on a playbook, or a digital twin, a blueprint, something that allows us not to reinvent the wheel. And last, but not least, we don’t do anything without people. We have to retain our talent and we have to continuously reach the pool of our competence and skill. This is what really allows us to keep driving an incredible journey, as we have done so far.

Daphne Luchtenberg: That’s a great place to finish this conversation. Stefano, Mark, and Evan, thank you so much for joining me in this initial conversation about gigafactories. We look forward to talking with you again in a few months, to see how this journey is progressing around the world.

You’ve been listening to McKinsey Talks Operations with me, Daphne Luchtenberg. If you like what you heard, subscribe to our show on Apple Podcasts, Spotify, or wherever you listen. We’ll be back with a brand-new episode in a couple of weeks.