Greetings, and welcome to this presentation of ABB's third-quarter results. You will hear from our CEO, Morten Wierod, and also from our CFO, Christian Nilsson. They will talk through the results as per usual. After this, we will open up for a Q&A. I am Ann-Sofie Nordh, Head of Investor Relations. With that, let me hand over to you, Morten, to kick off the presentation.
Thank you, Ann-Sofie. In the third quarter, we continued to see a strong market, with orders once again reaching just above $12 billion. I would say that we performed well in a vibrant market. When it comes to our operational performance, we progressed more or less as planned and guided for, a good mid-teens revenue growth. The 20.8% operational EBITA margin is the strongest Q3 margin ever for ABB. The same is actually valid also for the EBIT margin of 19%. We continued to generate a standout return on capital employed, which was at 29.9%, and I expect our business to continue to improve and support ROCE. We have said that we will deliver more than 20% return on capital employed also when we start making more sizable acquisitions. Speaking of acquisitions, following the Rotork announcement in July, we today announced further capital allocation toward M&A.
We will take a 51% ownership in a joint venture formed with Mitsubishi Electric's low-voltage power distribution business, and we also have a clear midterm path to full ownership. This deal makes us the low-voltage market leader in Japan and one of the world's largest electrification markets, where we have had limited reach until now. We also view this as a door opener for growth across Southeast Asia to markets like Malaysia, Thailand, Vietnam, and Indonesia. On the product side, they bring circuit breakers, motor starters, contactors, and control devices, and beyond that, energy measurement and monitoring equipment. The product portfolios are similar, and we can apply a best of both worlds integration principle for platforms and technology. The Mitsubishi business coming into the JV has around 2,800 employees, and in fiscal year 2025, it generated around $640 million of revenues with a low double-digit EBITA margin.
In terms of performance improvement, it will be similar to what we successfully did with the GE Industrial Solutions acquisition. We will drive profitability by expanding sales through combined channels, and we will introduce ABB products and technology to Mitsubishi Electric's customer base. We aim for the JV to be running at an operational EBITA margin of at least 20%, five years after deal completion. The transaction values the new company at an EV of around $890 million. This translates to EV over sales of 1.4 and EV over EBITA of 11.1 based on 2025 actuals. Performance is initially low in the JV. Hence, multiples paid are at a meaningful discount to ABB, even before we accounting for synergies. When we include the expected synergies, the EV over EBITA multiple reduces into the mid to high single-digit range.
We see strong returns on this deal with internal rate of return well above our cost of capital, and we expect the deal to close during the second half of 2027. Another significant milestone this quarter was the launch of the Infinitus direct current portfolio. This new portfolio enables us to support AI data center customers across the full spectrum of DC architectures, from source to rack. It spans across the white space localized DC designs to native centralized high-density architectures. DC is not new to us. We started our R&D investments into solid-state technology already 15 years ago. At that time, it was aimed at marine applications. Now, there are multiple examples where direct current solutions drive the industry forward, like in rail, renewable energy, microgrids, and now also in future AI data centers.
We decided to invest early in DC because the technology solves real customer needs. First, it enables higher power levels. It facilitates data center rack densities to shift up to one megawatt, and facilities can scale from 200 MW- 1.5 GW. Second, DC is more energy efficient. Fewer AC/DC conversion stages mean lower energy losses. A 500 data center shifting to DC architecture can reduce conversion losses by more than 5%. This will save almost 200 GWh annually. That's the electricity consumption of 60,000 European households or some tens of millions of dollars in cost savings. For 100-megawatt facility, a 5% improvement in energy efficiency can free up additional five to six megawatts for token generation. This is worth close to $100 million in potential annual revenue for the operator. Our portfolio is built on the concept of five modular building blocks.
This allows our customers to select and combine what they need to construct a complete system. This means we can support diverse data center topologies from hybrid transitional architectures to optimized target design. The building blocks span the entire power path, from medium voltage grid connectivity or behind-the-meter generation, to DC conversion and distribution, to power protection systems and cooling optimization. The modularity is critical because the transition to native DC architecture will be staged rather than immediate. We talk about four distinct design phases. I've said earlier that we view the sidecar as an interim solution, and we rather expect the economics of centralized DC architectures to ultimately drive customer decisions. With the Infinitus portfolio, we are ready, including our solid-state transformers available to order in the second half of 2027. That said, there will be no universal architecture.
The market will evolve across a spectrum, and with the Infinitus portfolio, we can serve every segment. When you compare the localized DC with a centralized design, labeled two to four on this slide, it is visible how the weight of the electrical power infrastructure shifts from the white space to the gray space. We believe that in the target DC architecture, as much as 75% of the spend on electrical power infrastructure can be linked to the gray space. This compared to about 55% in a legacy AC architecture. The gray space already represents the majority of our data center exposure today, and the shift towards centralized high-density DC meaningfully expands our addressable scope. Another proof point of our DC capabilities is our partnership with Ferrari Hypersail. We will support Ferrari Hypersail with expertise across marine power, DC technology, and advanced electrical systems.
Our solutions will be installed on their 100-foot oceanic full-foil monohull yacht, powered by renewable energy. The DC architecture will function as a high-performance, self-sufficient microgrid with systems operating at 48 volt and 800 VDC. The vessel is designed to bring together speed, resilience, and safety with onboard energy generation and management in one of the most technically demanding environments. Yet another DC happening in September was the report we published in collaboration with BCG. The report makes the case that direct current technology is moving from the periphery to the center of industrial, commercial, and digital infrastructure. The future of power distribution is not an either/or choice between AC and DC, but a hybrid system that combines the strengths of both technologies. AC and DC each have important roles to play, and together, they can help build a more efficient, resilient, and future-ready electrical system.
Now, let's turn back to the third quarter results. I already mentioned the order intake of $12 billion. This reflects a comparable increase of 40% from last year, and we're in a good market situation for all three of our business areas. Orders were supported by strengths across our entire portfolio, meaning short-cycle products, service, as well as the project and system businesses. We saw orders increasing in the majority of our customer segments. Data center was the standout, delivering triple-digit growth with no signs of demand softening or CapEx hesitation from our customers. But there are some softer areas, and similar to previous quarters, we see low customer activity linked to process industries and residential buildings. But I'll let Christian go into segment details when he talks through each business area.
In total, book to bill reached 1.24, and the order backlog expanded to just about $32 billion, almost equivalent to 2025 full-year revenues. There is always a mix of short and long cycle, but naturally, it is the project and system business that predominates in the backlog. As you know, we have had very good order intake from segments like data centers, marine, and rail. These are businesses with long lead times, skewing mix in our backlog further toward longer delivery times. So mix in the backlog is one thing to keep in mind. The other is production capacity. We have an overall good position, and you often hear me talk about our strong delivery track record. This is important to us, and we are investing incrementally, as you have seen in our announcement. CapEx is tracking at the higher level of around $1 billion.
With demand having been so strong in the last 12 - 18 months, some product lines are now approaching high utilization, mostly so in electrification, until ongoing investments gradually becomes fully operational during 2027. All three business areas contributed to comparable revenue growth of 14%. Volume expansion was the primary driver, complemented by net pricing realization of just about 2%, a stronger contribution than in Q2. We saw double-digit improvements for both the short and long cycle businesses. Looking at the different geographies, comparable orders were up by double digits in all three regions. The Americas continues to be the strongest growth driver and increased by 70%. This was driven by the U.S., where orders were up by as much as 82%. This high number includes large bookings, but base orders were very strong and improved by about 50%.
Europe was up by 10%, supported by increases in most of the top 10 markets. Asia, Middle East, Africa improved by 29%, with China up 25%. China orders increased in all three business areas, with strong double-digit growth in both Electrification and Motion. We turn to earnings, which reflect both favorable market conditions and a strong operational execution. We improved operational EBITA by 24% to $2 billion, up in all three business areas. As I mentioned in the beginning, a group margin of 20.8%. Operational leverage on higher volumes was the main contributor to our improved result. The solid price execution and the forever ongoing efforts on continuous improvements compensated for inflationary cost impacts and higher spend on R&D and SG&A. We reduced expenses in corporate and other, and losses were lower in the E-mobility business.
All included, a job well done by the team, and we keep moving in the right direction. That said, while we improved business performance, the earnings per share of $0.65 actually reduced by 2% from last year. There are two main drivers to this, the first being a non-cash unrealized FX hedge linked to the ongoing Rotork acquisition, which weighed on earnings by about $115 million. Depending on FX movements, we may have some additional volatility in finance net until we close the deal in the first half of 2027. Secondly, we had a lower result in discontinued operations. This is, of course, linked to the ongoing separation process in the robotics business. We are working towards our ambition to complete this divestment by the end of the year. With that, I hand it over to you, Christian.
Thanks, Morten. Let's take a look at what happened in the different business areas, starting with Electrification. Demand remained persistently strong and drove order intake to a new record of close to $7.7 billion. This represents a standout increase of 70% from last year. It was also up from the already high level of second quarter. What drives this exceptional demand? The segment trends have not really changed from what we've seen earlier in the year. The data center segment stood out again with a triple-digit order growth with no signs of slowing customer activity. That said, it is a tall order to expect triple-digit growth in each quarter going forward. It may be as simple as impact of timing of order bookings, and also the base is getting rather sizable. Electrification is not all about data centers.
It was very encouraging to see order growth, excluding data center, reaching the high teens level. All our large customer segments in Electrification grew at a double-digit pace. To mention a few, we saw a strong increase in the utility business, with investments going into the grid refurbishment and expansion. Orders in the building segment was up due to strong commercial demand. It improved also in China, where we performed well in an otherwise bleak commercial building market. We have executed well in areas like high-end hospitality, as well as niche-backed public funding like healthcare and education. Turning now to revenues, which amounted to $5.4 billion. The very strong comparable growth of 21% was above all supported by higher volumes. The team did a good job on pricing, which contributed with around 3%. This is a stronger contribution compared to Q2, so a good realization trend.
It was encouraging to see that pricing now broadly offset commodity inflation. The strong revenue growth with added contribution from continuous improvement efforts converted to 27% improvements in operational EBITA to $1.4 billion. The margins was up by 110 basis points to 25.6%. Again, new records for Electrification. All included, a great performance by the team. Looking at the fourth quarter, we expect comparable revenue growth to be in the high teens to low 20% range. While we expect a repeat of the usual seasonal pattern of sequentially softer margins in Q4, we anticipate a slightly softer drop compared to the historical average of around 150 basis points. We turn to Motion, where order intake of $2.5 billion corresponded to a strong comparable growth of 12%. This improvement was driven by the short-cycle business, with large order decline from last year's high comparable.
Similar to previous quarter, we saw strength in the grid-related investments. HVAC for commercial buildings and data center cooling, as well as food and beverage, also contributed to the growth. A more stable trend was noted in oil and gas, and while there may be quarterly differences, I would say that process industry-related demand in pulp and paper and chemicals continues to be overall muted. Turning now to the revenue at $2.3 billion, which grew by 9%. The increase was mainly driven by strong demand in the short cycle business, while we also executed on the order backlog. Main driver for revenue growth was higher volumes, with added contribution from a solid price development. In Motion, we also had about 2% revenue contribution from M&A. As in recent quarters, profitability in Motion was under pressure despite the positive impact from higher volumes.
Year-on-year, the operational EBITA margin declined by 100 basis points to 19.1%. Our Gamesa Electric acquisition hampered margins by about 40 basis points, still operating at a small loss. Pricing was robust, also in Motion, and we were able to offset commodity inflation in absolute terms, although still with some margin dilution. For the fourth quarter, we expect Motion to achieve comparable revenue growth similar to the Q3 level. In contrast to the usual pattern, we expect operational EBITA margin to slightly improve sequentially. The stronger than usual quarter-on-quarter development is supported by improved operational performance in the project and system business, where we've had some inefficiencies earlier in the year. Next, we turn to Automation, where orders remained stable versus prior year at $2.1 billion.
The order number is sequentially lower, but this is more due to timing of order placement rather than a real change in underlying customer activity. We still see good demand linked to our offering for higher energy efficiency and energy resilience. Demand for marine and ports, electrification, and automation remains high. We see improving market sentiments linked to discrete automation in the machine building segment. Order growth rates were strong, although still on a fairly low comparable. Demand in the oil and gas segment is generally robust, barring weakness in the Middle East. Similar to what I mentioned for Motion, we continue to see bleak process industry markets for chemicals and pulp and paper. Orders in mining remain muted in the quarter, but our colleagues are mentioning that they sense some initial improving sentiments in the customer interactions about future mining projects.
Early days, so let's see and follow up on this one. Nuclear is a small segment for us, but an area where we see high customer activity. Revenues reached $2.1 billion, up 5% on a comparable basis. The improvement was driven across the spectra of backlog execution, product business, and service. Despite gross margin pressure from items like unrealized FX and commodity derivatives and some cost inflations, the operational EBITA margin improved to 14.8%, up 90 basis points from last year. The margin improvement was driven by solid execution of the order backlog with an increased profitability, and also a stringent cost control in SG&A and R&D, which reduced in relation to revenues. For automation's fourth quarter, we expect a low to mid-single digit comparable revenue growth. We have the ambition to keep the operational EBITA margin broadly stable from Q3.
This will be a stronger outcome than the usual pattern of a sequential drop in Q4. Looking at cash flow, it was another solid quarter with free cash flows of $1.6 billion, supportive to our strong free cash flow margin trend. The slight year-on-year increase of 2% was driven by higher operational earnings in continuing operations. We also kept disciplined in our trade net working capital management with no material cash impact. This was partially offset by somewhat higher CapEx spend and the lower cash flows in discontinued operations. All in all, it was a good cash quarter. So far this year, we have generated $3.7 billion in free cash flow. We expect good business-driven performance also in our fourth quarter, and we received the cash contribution of around $300 million linked to the real estate sale we just announced.
We are well on track towards improving free cash flows from last year's $4.6 billion. With that, I hand it back to you, Morten.
Thanks, Christian. Let's finish off with the outlook. For the fourth quarter, we expect comparable revenue growth to be broadly similar to the third quarter. When it comes to the operational EBITA margin, we have a seasonal pattern of a softer Q4. We expect a sequential decline also this year, but slightly less than what we normally see. Note that this guidance does not include the positive one-off linked to a real estate sale we just completed. This sale will trigger a gain of about 240 million Swiss francs booked in the fourth quarter. This results in a slight upgrade for the full-year 2026 guidance. We now call for comparable growth to be in the low teens range, and we expect the reported operational EBITA margin to improve by at least 270 basis points from 2025, including real estate sales. Now, Ann-Sofie, let's open up for questions.
Yes, we're ready. Let's do it. First, a quick reminder for those of you who have dialed in on your phones, please press star 14 to register to ask a question. Also, remember to mute the webcast when your line is opened. We also, as per usual, we kindly ask you to limit yourself to one question at a time. This is so we can allow for as many of you as possible to be heard. You can also put your questions through the online tool in the webcast, and I will then voice them over from here. With that, let's take the first question, and we open up the conference call line for Will at Kepler Cheuvreux. Are you with us, Will?
Yes, very good morning to you. Thank you. My question would relate to the strength of the ordered intake in electrification, specifically around data center. We've seen two quarters above 100% or above 130% in North America. What levels of risk control are you putting in place to ensure the durability of the orders and to ensure there's not elements of pre-ordering or orders that may fall subject to cancellation in the event of a change in the economic conditions? Thank you.
Thanks.
Oh, yeah. Thanks, Will. First of all, when we do the normal risk approach when it comes to customer assessment, before we take the order, most of them also come with a significant advance payment or clear cancellation policies, which is part of how we do with every large project at ABB. We have to remember also that these data center designs, even if they are standard products, they are not standard solutions. It is still engineering involved. It means that the order is for a specific site, which already have a permit, to have a grid connection, to have the funding in place to be built. When we get the order, then it's all about execution, and as we said, lead time can vary now from 12 to 24 months. That's more depending on really on the project when they want the delivery on site.
We are normally not the limiting factor when it comes to the lead time. That's kind of how we look at it. We do not see any pre-ordering, and I don't see any kind of build-up for something which is anyway built to a site. It's very difficult to handle this by pre-ordering as well.
Thank you very much.
Thanks, Will. We open up for the next caller, James from Rothschild.
Yes. Good morning, everybody. Hi, Morten. Well, encouraging to hear your comments on data centers and that you see no sign of slowing customer activity. But I wondered if I could ask your thoughts on the wider equity AI ecosystem market debate and concerns that are going on. Specifically, we've got this known path of AI supply ambition with huge amounts of gigawatts, and an ambition from hyperscalers to build those. I could see a path to orders increasing and a desire to continue those starts and complete the existing and the announced, but not yet started.
But on the demand side of the equation, where you look at the model players and the hyperscaler players, do you look much and think much about what kind of returns and demand that side of the equation is seeing, whether it's sufficient to support the kinds of industry CapEx levels that we're heading to?
Well, of course, we discuss and talk with our customers, like the hyperscalers, about their both financing models, but also their revenue models of the future. The confidence there, and I think, but you maybe you can hear that directly from them as well, the confidence there that these investments that is being made now will have very solid returns. That you see also with kind of the I often say that the hyperscaler or the data center industry do have deeper pockets than most kind of standard industrial customers when it comes to getting those grid connection. I think that's more the, for me, would be the discussion longer term that we see if you're running out of grid, who will take that capacity?
You have a choice between decarbonization of industries or building data centers, and that's more of the industrial debate that is happening in many countries. I think what we are doing there is to help building these data center much more energy efficient and that you can do more with less, and at the same time helping more energy and electricity supply to come online. I think that is the part of the equation for the coming years. We will see much more investment where hyperscalers and the whole data center industry will take a bigger part, also bringing equity or capital into the electricity power generation, because, in the end, there will be not enough free capacity to pick up, and therefore, you would also need to have more grid expansion.
We see that especially in the U.S., with a lot of LNG and other grid expansion activities happening. I think we'll see that also here in Europe in the coming years. That, again, it's a sweet spot for ABB and where we play. That's kind of how I see it. Therefore, I'm confident when it comes to the longer-term develop also of this market, but both the demand and the supply side needs to be part of that discussion.
Thank you.
Thanks, James. We open up for Magnus at Nordea. Magnus?
Hi, Morten, Christian, Magnus here from Nordea. I think you delivered a very solid 85% growth in the backlog in electrification in the quarter, and you are guiding for very explicitly a revenue growth into the fourth quarter. Looking beyond, what limits your backlog conversion into 2027? Is it more about your own capacity, or is it the customer's ability to take deliveries? How should we think about the backlog conversion and revenue growth into next year? Thank you.
Yeah, thanks. I can start. We come out in January, when we do our fourth quarter, we will come also with the full-year 2027 guidance. I leave that part till then. What I can say is how we are dealing with the capacity build-out and electrification, and it is a mix here of how we bring new capacity online, and you have seen with a higher CapEx spend, where a lot of that has been allocated, of course, to electrification with a factory and manufacturing build-out. You may see quite a few news stories, where ribbon cutting and groundbreaking over the last years. Those are investment now coming online, and that is why you see a higher revenue, in electrification business. It is a mix here of what is our own capacity, but it also has to do at the customer side.
It is not just, we are very seldom really the bottleneck when it comes to deliveries. Like a data center we talked about, when that is being built, it is both the construction work, but also a lot of other equipment like power transformers that goes in here, gas turbines. Those is more longer lead item than what we do as ABB. We are very seldom kind of the limiting factor there for the build-out on projects.
Okay.
Thank you very much.
Thank you, Magnus. We have a question here via the online tool, regarding CapEx. It is from Frederick at SEB. He says, "Given the growing backlog and high utilization rates, could you please update us on expected CapEx profile for 2027 to 2029?" That is for you, Christian.
I am going to take that. The CapEx plan for this year is around $1 billion, as we have communicated, in the past. We do not expect that to grow significantly as we move across the next few years either. This is really a good level for us. As a reminder, of course, this has grown quite a bit since some years back, but around $1 billion or so is still the target.
Very good. Then we open up for another question from the call. It is from Andre at UBS. Andre, please.
Yes, hi. Thanks very much for taking my question. I wanted to ask about the margin implications from the capacity ramp-up. So far, I think you've delivered very credible 30%, 35% operational gearing and electrification, with sales obviously ramping 20%. I think for next year, just judging by the speed of your orders growth in the last four quarters at 50%, it may well accelerate, and you've just talked about adding capacity in various ways. Should we think about operational kind of drop-through implications from that additional capacity ramp, or do you think you can maintain that kind of operational gearing that you've delivered so far this year?
I'll take that.
Yeah.
I'll take it. That gearing is going to be as similar as we move forward. That's a good performance from us so far, and there's nothing changing necessarily in the profile going forward for that.
It's very clear. Thank you.
Very good. Then we have a question from Max at Morgan Stanley. Your line should be open, Max.
Thank you. Maybe just to ask one sort of off the topic of data centers. Just in terms of the automation business, obviously step down in sequential orders, you talk about kind of timing of orders in the Middle East. I am just trying to understand, are those two comments connected? Was it one particular kind of larger order that maybe slipped from the quarter? Is there any kind of underlying weakness? Maybe just unpacking that sequential order deceleration that we have seen and maybe the moving parts between that. Thank you.
Yeah. Thanks, Max. As always, in the automation business, there is more volatility in the order intake, especially on the large projects. As I am sure you remember, we have had quite a few of those large projects in previous quarters also last year. So sometimes the comparable got a bit tough. But if you look at the automation business, when we are looking at for this quarter, we had good activities in the power generation on the energy side, I would also say on the machine building, on the B&R business. Where it is a bit slower is more kind of on the process, some of the process side, where we see still kind of waiting for. We have more optimistic when it comes to the future in mining, and there we see kind of more conversation, not on the number side, but more on the conversation with customers.
While when we look at the part of, let us say, traditional process industries, pulp and paper, the metals industry and so on, cement, those are more kind of a bit of a tougher market. That is kind of the overall outlook. But when we talk about the electrification trend, that benefits the automation side as well, that being when you are looking at capacity, like in LNG, it is going well, so the oil and gas side, but also into the machine building. Those are elements that are just driving. I think overall we are optimistic also when we are looking at the future intake here. But I say we also always need to look a bit kind of through some of these large order bookings when we are looking.
We look quite a bit on the base orders, what we call it also in ABB, when we are looking at the automation side, and there it is growth on the base orders, but again, a bit short on large orders just this quarter. Yep.
If I can just add.
Okay. Thank you.
I can just add, if I may, that the timing comment was not a direct link to Middle East. There you should see them as two separate things, because Middle East has been on the soft side since the beginning of Q1, and that has not changed. With that said, we have a next question. We open up the line for Ben at Oxcap. Please, Ben.
Oh, good morning, everyone. Thank you for taking the question. It was about the joint venture with Mitsubishi Electric. Very interesting deal. Could you just talk through the strategic rationale? I mean, of that 640 million of sales, how much is Japan versus other markets? I am assuming, and tell me if I am wrong, that this is about gaining access and penetration, mainly in Japan. The related question is, the scope of it is literally just the low voltage distribution products. When I look at your businesses, there is a huge amount of complementarity actually between the two in energy systems and factory automation. My question is: Is this really just about low voltage distribution, or could this be something broader in the medium term?
Oh, thanks, Ben, for the question. First of all, this is a joint venture that we talk about now between Mitsubishi Electric and ABB, that is limited so far to the low voltage product of the electrification market. The main reason why we want to go in here is that Japan is one of the largest electrification markets globally. But it has been so far, let us say, difficult for us to come as ABB, come into the market.
Mitsubishi is a clear market leader in this space. That is now with this joint venture, and we are taking a 51% controlling stake, will make us the local market leader overnight. You can think about it like the GE Industrial Solutions case all over again, where we are bringing in a global platform of products like circuit breakers, contactor, all the industrial devices that is needed in that electrification space. Then we are taking the best of the two, when we are looking at those two portfolio, and we will manufacture that majority in over this new JV operational unit in Japan. So that is, as you say, the majority of business, out of the $640 million, is done in Japan, as you say, with a clear market leadership.
We are looking at how we, as I say, when we are taking the best of the two, based on the GEIS acquisition kind of principles, we will also see how we can bring up the performance of the unit from today, a bit of a low double-digit EBITA margin and up to at least 20%. So, it will take a few years because that we know, as we have done it before, it will take a couple of years to get there, but we are very confident that we will get there, and been able to create a lot of value for the company. We also have a firm exit, kind of a firm deal for when we will take over 100% control of that new entity. If we can do more together in other areas, that is more discussions to be held.
At time, now the focus is to do this successfully, then we will build on that success onwards.
Understood. Thank you.
Thanks, Ben. Then we have another question here from the online channel, from Thomas Jager, who says: Free cash flow looks to be really good in 2026, significantly up from last year, approaching $5 billion. Do you agree, and can you talk about the drivers?
Well, I certainly can, and I do agree that year-to-date, the free cash flow at $3.7 billion is really strong. Good operational performance on that. If we walk from that level into, as the question alluded, towards the finalization of Q4, of course, there's different puts and takes. First, of course, the continuation of our strong operational performance will, of course, generate positive cash flow in Q4. Adding to that is the real estate that we have talked about, that $260 million also adding to that. Of course, subtracting of the opposite of that is our continuous investment, as we mentioned before, around CapEx, that we support our customers. Net working capital, probably a little bit of drag also on that as we are supporting our growth into Q4, then discontinued operations.
If you put it all together, I think the fair path is that we are moving towards a number indeed that is starting with a five, so the $5 billion or so. Yeah, I would agree that it is a good performance so far, and that we have a good path to $5 billion.
Then we take a question from the conference call, Delphine at Oddo, please.
Yes, good morning, all. Thanks for taking my question. It is another question not on the data center field. Your margin in motion was down 100 basis points or 60 basis points stripping out the Gamesa impact. Can you detail the main components of this - 60 basis point? If you believe Q4 will show better trend, notably in terms of price to cost equation?
You want to take that, Christian?
Yeah, indeed. For Motion, as we have had throughout the year, we called out some different areas of pressure points. Gamesa is indeed still there, and that will remain as we move forward with continuous improvements, a little bit still operating at a small loss. Then it is a little bit of low operational issues we have experienced in areas such as high power, that it is continuing to be there, but kind of keeps improving. We have said throughout the year that the margin of Motions should sequentially improve, and they have so far, and I think it is a good path for the team. Yeah, I think those are the main components for the situation of margins in Motion.
Thank you. Thanks, Delphine. Then we have a question here from the online tool from Mattias Holmberg, who says, "How is ABB using AI internally, and what value are you seeing in cost efficiency, time savings, or product and service improvements? Please quantify where you can.
Yeah. So how we use AI at ABB. Maybe I can give you a bit of an overview of how we are working on it. We have defined as three pillars, which is AI for people, AI for business and functions, and AI for technology or for product systems and solutions. If you talk about AI for people is kind of how we upgrade our whole population ABB. We have today, our own employees are doing all the AI certification, internal training programs that we have put in place. Also, our teams have created today, it is almost 10,000 agent that is also working for our team members. I think here, and of course all of these are driving the efficiency and how we are able to do that.
We have tried to, that everybody should use AI, and everybody should drive their own kind of self-efficiency, and to drive this as an improvement program. That is also how we measure it. We measure both the improvement, but we also measure the token cost internally, so that we have control of the expenses. Then it is not just about speeding up the internal processes that is being used, but it is also redesigning, and then we are into the AI, how it is used for business and functions. That is a redesign often of a process that we can make it kind of an AI original, and not just automate the manual process with AI tools. That is a significant program, it is part of our transformation and our, what we call data system modernization program at ABB, and how we link that together.
Thirdly is about AI for our customers and in the technology. There is investments into service agents that can be on customers on premise, on their system, to help them, kind of being that local service technician online that can help you immediately without, you do not need to call or you do not need to send anyone to site. These are some of the very kind of practical example how we do it. Then of course we, internally, it is very much about how we drive efficiency when it comes to speed and responsiveness, and the internal efficiency. When it comes to the outside is how you can also use that to drive revenue growth, together with, and making our customers more competitive.
What we have done to put in place also, say, both the control or the monitoring of the cost side, but also, and what is the benefits. I cannot give you a number. This has made ABB, like 18 or whatever is the percent, the number, faster. What we do is to measure it in function by function, and in the different parts of our business, that we are really getting the real return of the investments we are doing. So for us, it is more important that everybody have a chance and have an opportunity to use it, and then we have been more focused on this to get the awareness and the usage of AI up in the company, more than just strictly controlling the cost of using it. That is a process we talk about. We are probably moving more into now.
When we get more mature, you also need to be more critical to what are the applications that you are really using it for. So these are how we use it, those three main pillars is how we use it inside ABB.
Very good. Then we open up for Ben at Bank of America Merrill Lynch from the conference call. Please, Ben.
Yeah, morning. Thank you for the question. I just wanted to come back on the margins on electrification, because they are very strong. Could you give us a bit of an update in terms of where you are regionally for margins, and in particular, on the U.S., given that's historically where some of the weakness in electrification was? That would be great. Thank you.
Yeah, I may start.
Yeah.
We have the development on margin, as you said, Christian answered here as well, kind of the drop through when we have a significant volume growth do come out also as in improved drop through margin, and that's what we expect. There are still regional differences, where the United States is somewhat behind, but we are improving step by step as we get more capacity online. We're getting also more mature in our U.S. operation, but we're still not at the level where we'd say that the job is done. We have more work to do. And especially in the United States, but we also, I would say in general, we are looking also how can we drive this continuous improvement journey that comes always with an expectation, especially when you're growing, there should always be an opportunity to also drive efficiency, that you can get even more.
It's very hard to improve margins when you're going down, but it's always easier when you're growing. And I think that is what the team is also showing, an excellent performance in this year and also especially, with the numbers this quarter. But there is still more work to be done continuously and even go to an even higher level.
Yeah. If I just add on with that journey, I think it's definitely a journey and a path. There's more to come. I think we can think about the margins improvements to go up towards the kind of high teens as we are today. That means that we, of course, have still further opportunities to improve even better in the U.S. for electrification.
Very good. Linked to, let's see, electrification from George. Your electrification backlog is up high double digits year-on-year. How much of this is data centers, and can you help us with the phasing of these orders converting to revenue?
Talking about when they convert, what I can say is that normally, a normal lead time here would be 12 -1 8, 24 months in the data center space. That's kind of the normally requested lead time from customers on these projects. The orders we booked this year should go out in 2027 or in 2028. While in the automation business, it is a longer lead time there. We have projects that could go over even up to 10 years. But in electrification it's normally down to a couple of years. That's what we see on the backlog.
Maybe I add on the backlog generally for ABB, of course, closing this quarter at the record $32 billion, and some change, that will convert into next year at around 40%, as we stand right now across ABB. Of course, mix is always an impact of that also.
Very good. Then we open up the call with a question from Alex at Evercore. Alex, please.
Thanks, Ann-Sofie. Morning, Morten and Christian. I wondered if you could talk a bit about the acceleration in grid utility spend. It feels like that's really actually started to build a lot more momentum as you've gone through the year. I appreciate it's probably a bit of a halo effect from data center demand, but can you give us a bit more color on the utility behavior, what they're sort of planning for, the visibility that you have, and I guess any broader developments or color on the regional split would be really helpful. Thank you.
Yeah. Now we see that on the utility space, it's going well in all three of our reporting regions. One is of course, as you say, due to the data center higher demand there, that you also need to balance on the supply side. It's not only related to that, it's the whole electrification of society that is driving that, where you need more electricity available, for more cooling of buildings, where HVAC is a good growth area. It's also about the whole electrification of transport, and then comes the data center on top. All of these things kind of is a higher demand for electricity, and then this is on the supply side. Building out a stronger grid, and having better grid resilience and reliability is one of the key aspects here.
We see that China has invested for many years in this space, and continue to do so because their carbon neutrality, that is committed in 2060 means that there need to be a significant investment into this energy transition as they do. It's the same in India, have a similar commitment, but also their good growth in the market. Then we talk about U.S., and we all know about also there, the aging infrastructure. So often when you think about grids and utility business, it's about an aging infrastructure, it's about energy expansion, that we need more electricity, and then you have the energy transition on top. These three elements come together, means this massive need of investment on the utility side. That's what we see kind of in region by region, that this is clearly not linked with the GDP growth.
It is really about how you can get more electricity online, and that is what is reflected in the utility number when we looking at that, well above double-digit growth.
Very good.
Thanks very much.
Thanks, Alex. Then we take another question from the conference call from Eric at CIC, please.
Yes. Hi. Good morning. Thanks for taking my question. I was wondering, regarding the 800 VDC architectures, do you expect the emergence of new competitors in that space? Some new players trying to grab a share of this promising market and maybe new players with, I don't know, maybe new technology?
Yeah, I think always when you make a technology shift, there is always an opportunity to make market changes, and we want to be one of those disruptors in this field because we are very strong today on the gray side of the data center, not that strong on the white side. Now, and I showed that earlier in the presentation, more of that investment is moving from the white space to the gray space, and with our leading technology here on the 800 volt DC and that DC architecture, we believe that this is a great opportunity for us to capture more of that growth and more of those opportunities with our leading technology.
That's why you have seen this significant investment from our side in R&D and technology, also some with partnership, that we're building up a complete DC portfolio when I think it's what I hear also from customers and our partners, it is clearly one of the leading or the leading platform and technology platform to build that new DC architecture. I think this is a great opportunity for us, and we want to be one of the disruptors here that we can take an even larger share in that market than what we have today.
Understood. Thank you.
Thank you, Eric. We follow up with a question here from the tool, from Andre. He says, "Can you please comment on your mix of exposure by customer group in the data center segment?" I assume he means mix of hyperscalers, neo cloud, colocators, et cetera.
Well, it is a mix for sure. We do not go out with the full mix, but majority of what we do is with a clear majority with hyperscaler and colocators. That is kind of the big majority of the side, but also some on the neo cloud solutions. As I say, by far, vast majority, it sits with the hyperscaler and the colocators. That is the mix, what we have so that without going into the number side of it, yeah.
I think we can squeeze in one more question, so we open up the line for Daniela at Goldman Sachs, please.
Hi, good morning. Thank you for squeezing me in. Just wanted to ask in terms of how should we think about balance sheet capacity from here and also your own management capacity to deal with further M&A and whether there are any divisions in particular where we see a bigger white space of things that are still possible to do. Yeah.
Sure. Let me start with the management capacity maybe first. I think this is one of the advantages and the strengths of our ABB Way operating model that the divisions that sit in our three business areas are the ones who are also having the integration capacity, and they also take the Rotork deal that sits with the automation team, are not affected, so to say, by the Mitsubishi Electric deal that we did now in the electrification space. Even there, it's limited really to two operating divisions who will run that program as we did on the GEIS side. So I think that's the strength of the model. That means we have management capacity to continue to do deals. That also what we, of course, take this into consideration when we make new ones.
But in general, I would say we have that management capacity as part of our ABB Way operating model. Maybe Christian, you talk more about the comment on the,
Yeah, we have the management capacity, and indeed, we also have the financial capacity. Now it's close Q3, we have a net debt over EBITA around 0.2, so of course, quite low, and we could feel like we could, as I said before, easily move up towards the 2x , staying in our rating area as we are, and that would, of course, free up, call it $13 billion - $14 billion. Then we have, of course, our general continuation of our strong cash flow that we get from operating activities. So yeah, we definitely have both management capacity and financial capacity.
On that strong note, I think we close this session. Thank you very much for taking the time to join us today, and, if not before, we'll see you in a quarter's time.