Hello, ladies and gentlemen. Welcome to Nokia's third quarter 2026 results call. I am David Mulholland, Head of Investor Relations. Today with me is Justin Hotard, our President and CEO, along with Marco Wirén, our CFO. Before we get started, a quick disclaimer. During this call, we will be making forward-looking statements regarding our future business and financial performance, and these statements are predictions that involve risks and uncertainties. Actual results may therefore differ materially from the results we currently expect. Factors that could cause such differences can be both external as well as internal operating factors. We have identified such risks in the Risk Factor section of our Annual Report on Form 20-F, which is available on our investor relations website. Within today's presentation, references to growth rates will be on a constant currency basis, and other financial items will be based on our comparable reporting.
Please note that our Q3 report and a presentation that accompanies this call are published on our Investor Relations website. The report includes both reported and comparable financial results and reconciliation between the two. In terms of the agenda for today, Justin will go through the strategic highlights, and Marco will go through our financial performance before we move to the Q&A. With that, let me hand over to Justin.
Thank you, David, and hello, everyone. Team Nokia continues to deliver on our commitments and execute on our strategy. Growth is accelerating in the areas where we have chosen to focus and invest. That growth is also beginning to translate into stronger operating leverage. In Q3, we grew net sales by 9% and expanded operating margin by 220 basis points to 11.7%. AI & Cloud was again our largest growth engine, with earlier design wins converting into revenue and IP Networks and showing continued strength in Optical Networks. Mobile Infrastructure grew in the quarter as our focus remains on profitability and improving return on invested capital. Technology Standards continues to execute its strategy and licensing. We are tracking to deliver above the midpoint of our full-year outlook.
Before Marco takes you through the financial performance in more detail, I will give you an update on the progress that we are making against our strategic priorities. In AI & Cloud, net sales grew 123% year-over-year and represented 12% of total group net sales in the quarter. Order intake was strong again, up 118% year-over-year to EUR 1.5 billion. Demand from AI & Cloud drove IP Networks to grow 43% year-over-year as previously announced design wins began to convert into revenue. Optical Networks grew 18% and still contributed the majority of our revenue in this customer segment. However, IP Networks is now playing a more prominent role. In both businesses, we continue to see demand that exceeds supply. The market continues to develop, with demand accelerating across all segments, including scale across front-end networking and data center interconnect.
Further, we also see demand extending into our telco customer base as they invest to support demand from AI and cloud customers. Optical Networks and IP Networks are now on track to grow somewhat above the 18%-20% assumption we provided for the full year. For 2027, we now expect these businesses to grow between 25% and 30% on a combined basis, reflecting the strength of the underlying market and our position to capture that demand. The build-out of AI infrastructure is driving demand across Optical and IP Networks. Demand for our 800G coherent pluggables continues to accelerate, with shipment volumes increasing more than twofold sequentially. We are securing orders from both hyperscalers and major telcos. We now have design wins with 29 customers. We are also investing in our next-generation optical roadmap, developed in close collaboration with our hyperscale customers.
In line systems, an important growth area where we have historically had less share, we are making meaningful progress. During this quarter, we secured a significant win with a large hyperscaler. Our Multi-Rail technology, launched this past March, delivers 25% higher density than competing solutions in the market and has secured three design wins, with the first customer now receiving initial volumes. In IP Networks, momentum remains strong. We secured two additional hyperscaler design wins, including an important next-generation data center gateway win. We also secured significant telco wins across edge and core routing, IP aggregation, and 5G backhaul in Japan, Europe, and the U.S. Customers are modernizing their networks to support growing AI traffic.
With demand for intelligence becoming more distributed and localized, this creates more data centers in more locations, which requires more connectivity to deliver services with the latency, performance, security, and locality demanded by AI workloads. That means more AI compute in more places, requiring more optical and IP capacity to connect them. For inference and workloads, security, trust, and sovereignty are foundational for operators. We are the market leader in telco edge routing, and we have a rich optical portfolio which uniquely positions us to deliver solutions to meet this market need. An example of this is helping our customer, Zayo, a major internet backbone provider in the U.S., to create an architecture and service blueprint for connecting distributed inference across data centers for multiple partners. We are also seeing encouraging traction in out-of-band management, leveraging our passive optical technology from our Fixed Networks business.
Out-of-band management provides data centers with an independent optical lifeline to critical IT infrastructure if the production network becomes unavailable. Compared with traditional Ethernet-based architectures, our solution can reduce energy consumption by 50%, reduce points of failure, and improve network resilience through separation from the production network. We announced our first design win last quarter and are now actively engaged with additional customers. In addition, in our core Fixed Networks business, we are focused on growing our optical line terminal business through continued innovation and differentiation while we bring greater discipline to our customer premises equipment business. In optical line terminals, we continue to lead the market with approximately 45% port share globally, excluding China. Last week, we announced an industry-first 16-port line card that supports 10G, 25G and 50G PON on every port.
SoftBank completed Japan's first commercial implementation of this new solution, demonstrating that GPON, 10G XGS-PON, and 50G PON can coexist on the same fiber. It gives them the opportunity to invest in higher speed capacity as demand grows without requiring replacement of equipment. SoftBank is one of several customers in Japan moving from its legacy EPON base to this technology, creating a new serviceable market for our portfolio. With AI -RAN, we are transforming the radio network into a platform for physical AI. This is critical for emerging applications such as autonomous mobility, industrial automation, robotics, and augmented reality. The technology industry has a long history of moving workloads from proprietary silicon to general purpose computing platforms. For example, over the past decade, we have seen this happen in data storage and in mobile core networks. RAN is entering the same transition, driven by the unique requirements of serving intelligence.
We are positioning Nokia to lead this shift in partnership with NVIDIA. Nokia's AI-native anyRAN software running on NVIDIA's Aerial RAN Computer provides a software-defined foundation for integrating AI-powered radio functions that improve network capacity and performance. This means our customers can adopt new capabilities as their needs evolve, including a software upgrade path to 6G. This past weekend, T-Mobile completed their first live voice call over AI -RAN in its production network. In addition, Elisa yesterday announced that they have run voice, mobile data, and AI workloads simultaneously on the same AI -RAN platform in their commercial network. Together, these milestones show that AI -RAN is nearing commercial deployment. In live environments, we've already demonstrated spectral efficiency gains of more than 20%, increasing capacity and improving network economics against our roadmap of 100% improvement in spectral efficiency gains in 2028.
Our general-purpose approach allows new applications and services to be built on the network. At our Radio World event in September with our partners DeepSig, ZaiNar, and Cohere, we showcased real-time sensing of the physical world to enable precise network-based location capabilities using only existing wireless infrastructure. This creates the potential for new telco services and a path to monetize AI. Accelerated computing moves radio networks beyond the architecture that the industry has deployed since 3G. Our customers are telling us we're on the right track, and engagement is accelerating, with 19 customers from around the world now working with us. Alongside our progress in AI & Cloud and AI-native networks, we're working more closely with customers to shape the next generation of network technology. That collaboration is translating into meaningful wins. Our partnership with Google in support of their EUR 13 billion investment in Finland is a great example of this.
It shows how our Optical Networks and IP Networks will help to connect the next generation of AI infrastructure within and between data centers. We are also making progress with our Nokia Defense Incubation unit. Last month, we announced a strategic partnership with ICEYE to build secure, sovereign, low Earth orbit satellite communication systems for governments and defense clients. Alongside these efforts, we continue to execute our focused capital initiatives, investing for growth and simplifying our portfolio. In optical manufacturing, our new San Jose fab remains on track to start production before the end of the year. We've completed 90% of the process development and the process recipes and expect 24/7 operations to begin by the end of this month. At the same time, we've completed the sale of our fixed wireless access business to Inseego and signed an agreement to sell our enterprise campus edge business to AURELIUS.
Finally, we remain focused on delivering sustainable returns. In Q3, operating margin expanded by 220 basis points, and we are delivering on our cost savings commitments. We are accelerating restructuring in Europe and expect the integration work in China to be 90% complete by the end of the year. The Infinera integration is largely complete, and we are on track to achieve the EUR 200 million synergy target by the end of this year, one year ahead of schedule. These actions taken together provide a more durable operating cost base for the company. In closing, growth is accelerating, operating leverage is becoming more visible, and our investments are strengthening Nokia's position in the technology transitions that are reshaping connectivity in the AI super cycle. We remain focused on delivering our full-year commitments and building the foundation for sustained growth and long-term value creation for our customers and shareholders.
With that, I'll turn it over to Marco to take you through the financial performance in more detail.
Thank you, Justin, and hello from my side as well. We had a strong quarter three. Net sales grew by 9%, supported by both Network Infrastructure and Mobile Infrastructure. Gross profit was EUR 2.3 billion, and gross margin was largely stable at 44.6%. The improvement in Mobile Infrastructure gross margin was offset by product mix in Network Infrastructure and the costs associated with ramping new products. Operating profit was EUR 605 million, and operating margin increased 220 basis points to 11.7%. The improvement mainly reflected higher sales and continued operating leverage across the group. Financial income and expenses benefited from a positive venture fund revaluations during the quarter. This supported both net profit and earnings per share. Free cash flow was EUR - 99 million. We ended quarter three with a net cash position of EUR 2.6 billion, maintaining both a strong balance sheet and financial flexibility. Looking now at Network Infrastructure.
Net sales grew 15% in the quarter, reflecting continued positive momentum across Optical Networks and IP Networks. Optical Networks grew 18%, and the growth was supported by continued demand from AI & Cloud customers. But we also saw healthy demand from telco customers investing in transport infrastructure. IP Networks grew 43%, and the strong order momentum we saw in the second quarter is now translating into a significant acceleration in revenue growth. Fixed Networks declined 22%, reflecting the portfolio actions we are taking to prioritize higher-margin products. Sales of our optical line terminal products were stable, with the underlying business remaining healthy. Gross margin declined 150 basis points to 41.2%, and this was mainly driven by product mix and the ramp-up of new products for AI & Cloud customers in IP Networks. We currently expect gross margin to improve in the fourth quarter.
Operating margin increased 230 basis points to 10%. The improvement was mainly driven by operating leverage as revenue increased. Turning now to Mobile Infrastructure, where net sales grew 5% in the quarter and Radio Networks grew 10%, benefiting from some contract deliveries moving from the fourth quarter into the third quarter. Core Software declined 13% as we de-emphasize certain applications, and we also saw some operators delay projects due to the supply chain constraints for server infrastructure. Technology Standards grew 8%, supported by the signing of several licensing agreements and the recognition of some catch-up net sales during the quarter. We have seen a strong performance in the business year -to -date in signing new deals. In quarter four, we expect slower deal activity and therefore a lower revenue level than what we have seen in the last few quarters.
MI gross margin was 47.5%, largely stable compared with the previous year. The quarter benefited from acceleration of revenue from quarter four into quarter three in Radio Networks, operating efficiency, and higher contribution from Technology Standards. Operating profit increased, reflecting higher gross profit and lower SG&A expenses. This was partially offset by other operating income, which had benefited from a positive hedging impact in the third quarter of 2025. Turning to sales by customer segment. AI & Cloud was again the fastest-growing segment, with the net sales increasing 123% year-on-year. Sales to telecom customers were stable, while mission-critical enterprise grew 7% and technology licensing grew 8%. As you heard from Justin, we are confident in the growth potential for AI & Cloud given our backlog and pipeline.
At the same time, our expectations for a stable telecom market remain largely unchanged, with some pockets of growth in IP and Optical driven by data center connectivity. Turning to restructuring and integration costs. As part of our strategy to drive long-term operating leverage and to ensure we are investing our resources in the areas that drive the greatest long-term growth opportunities for Nokia, we have further accelerated some of our restructuring actions. We are focused on simplifying our processes, streamlining our operations, and consolidating locations. We are also driving a faster integration of our China operations, as Justin touched on. We now expect restructuring charges of approximately EUR 1.1 billion in 2026, and restructuring and associated cash outflows will now be between EUR 800 million and EUR 900 million. Turning now to cash flow. Free cash flow was EUR -99 million in the third quarter.
Cash generation was impacted by restructuring payments and a further build-up in working capital. This included higher receivables and inventory as growth accelerates. Given the strength of the demand opportunity and the growth we now expect in 2027, we plan to invest an additional EUR 400 million in working capital compared to our expectations at the start of the year. As a result, we now expect free cash flow conversion for 2026 to be between 35% and 55%. While this outcome is partly due to acceleration in growth, we continue to focus strongly on cash flow and ensuring we can accelerate our cash conversion as some of the working capital pressures ease. Finally, turning to our outlook. There is no change to our comparable operating profit guidance. We continue to track above the midpoint of our comparable operating profit guidance range.
The strength of the performance we saw in quarter three reflected some pull forward from quarter four as we work to improve linearity. Looking at the fourth quarter specifically, we currently assume a sequential increase in net sales of between 20% and 24%, although the pull forward will dampen the normal seasonality we would see in Mobile Infrastructure. Overall, quarter three was a solid quarter, and we remain focused on both capturing growth and driving the operational rigor required to maximize the long-term value creation potential of the company. With that, let me hand back to David for Q&A.
Thank you, Justin and Marco. As usual for the Q&A session, as a courtesy to others in the queue, could you please limit yourself to one question and only a brief follow-up? Jenny, could you please give the instructions?
Thank you. Ladies and gentlemen, we will now begin our Q&A session. If you have a question and are using the Zoom app, we ask that you please use the raise hand function at the bottom of your Zoom screen or by clicking the three dots on the black bar at the bottom of your screen. Alternatively, if you have joined via Zoom browser, please click the reactions button at the bottom of your Zoom page and select raise hand. Once your name has been announced, you can ask a question. If you want to withdraw your question, please lower your hand using the raise hand function. We would like to give as many attendees the opportunity to ask a question, so please limit yourself to one question and re-raise your hand in Zoom to rejoin the queue if you wish to ask another.
I'll now hand back to David Mulholland, Head of Investor Relations for the Q&A.
Thanks, Jenny. We will take our first question today from Rob Sanders from Deutsche Bank. Rob, please go ahead.
Yeah. Hi. Maybe if you could just give an update around your 6-inch ramp in indium phosphide. How are yields progressing? Are you confident of the ramp as you go into next year? And how many more milestones do you need to cross before getting really high conviction on that ramp? Thank you.
Yeah. Thanks, Rob. I think I touched on this a little bit. Fab 2 is our 6-inch indium phosphide fab. Fab 1 is 3-inch wafers. In that regard, we are very pleased with the ramp. I touched on this in my comments. 90% of the process development is done moving into 24/7 operations. Obviously, the key milestone is seeing us enter production by year-end, and then continuing to see production ramp through the year next year. So we have got a set of milestones that start to move into obviously the volume and yield, and those are the things we will be watching. But obviously, we will probably give you another update in January, but we are quite pleased with the progress and obviously touched on it today. Did you have a quick follow-up, Rob?
Maybe if you could just address the opportunity in satellite in North America, just given recent market developments. Thanks a lot.
Yeah. Two things on that one, Rob. One is, our focus in the partnership with ICEYE is going after sovereign communication networks. These are really networks, particularly what we see in Europe initially is networks that are being upgraded from legacy Satcoms that has been deployed. The way I think about those are 2G or 3G wireless communication, Satcom solutions that need to be upgraded, one, to get to modern technology, to meet the demands today for Satcom, but two, to also provide the capabilities for the future. That is a big focus for us on this initial partnership with ICEYE. Beyond that, look, I think there is a lot of discussion in the U.S. about Satcom.
I think we have been really clear that with the amount of connectivity demand coming from physical AI, we think connectivity is going to grow, demand for connectivity is going to grow, demand for persistent connectivity is going to grow. We think satellite is going to be a part of the solution. The work that we are doing is looking at how we enable that, and that is a key part of why we are also investing in AI-RAN because AI-RAN provides spectral efficiency that will also be necessary to support that increasing demand.
Thanks, Rob. We will take our next question from Simon Leopold from Raymond James. Simon, please go ahead.
Yes, thank you very much. Appreciate it. I will ask both up front. The first one is, you have talked about the order patterns of the IP and optical together. I would appreciate it if maybe you could double-click and drill down on the IP, the routing and switching in particular, and help us gain a better understanding of where you are in terms of expansion and verticals beyond the telcos. It sounded like that got better. I would like to make sure I understand that. I will give you the follow-up right now to hopefully help out. One of your competitors has offered a forecast for the optical business growing on the order of 30% in each of the next two years. I appreciate Nokia's customer base is somewhat different, but how do you see this particular market shaping up over that kind of timeframe? Thank you.
Thanks for the question. First of all, on the order patterns and sales funnel, I would say two things on this. One is, what we're seeing, and I touched on this again a little bit in my comments, is that the demand is in three areas, for Optical Networks and IP routing and switching, and it's really in these areas around front-end networking, data center interconnect, and then obviously scale across. Scale across for us today is much more of an optical play. The others are much more a balance of Optical Networks and IP. What we're seeing rapidly evolve, and you don't really see it in the quarter because there's some other moving parts with what we're doing with de-emphasizing non-profitable parts of our ONT business. A little bit of the discipline we're getting in the portfolio that Marco talked about in Core Software.
When you look underneath our numbers, what you do see is more investment by the telcos, also in providing effectively connectivity for east-west traffic. And this is data center interconnect, long haul, et cetera. And that investment continues to come. Obviously, we're watching the CapEx guides very closely as you guys are from our customers, but we're quite optimistic that that's going to be a place where they're going to deploy capital and see a very attractive return. Because despite a lot of the sentiment in the market, the demand for connectivity in AI & Cloud is massive, and it's insatiable. And it's driven by the fact that if compute is constrained, connectivity becomes another way to maximize the access to the assets or the utilization of the compute assets. And so that hopefully gives you a little bit more color.
But we're seeing, obviously we started from a lower base. We talked about this a little bit last quarter, lower base on IP, and now we're seeing more of that come through in revenue. We expect to continue to see order strength in IP Networks as we go forward from AI & Cloud customers. And then in Optical Networks, we're continuing to see that demand on both ends. And again, we're very supply constrained. And then as you think about the growth, maybe just to go to that one for a minute. Look, we're providing a view of IP Networks and Optical Networks combined of 25%-30%.
I would just say that, our view right now is that, as we build momentum in our portfolio, when we build out our supply capacity, including the investments we talked about in Fab 2, we have some optimism that Optical Networks can be a bigger contributor to that growth. And look, for me, I think the only other thing I'll say about 2027 versus 2027 and 2028, I think this is the third time I've given you an update on our growth assumptions since November, since the CMD last year. I want to give you a concrete view of 2027. Rest assured, we see a very durable market here. We wouldn't be investing at the pace and the scale that we are or in the portfolio if we didn't. And that extends well beyond 2027.
Thanks, Simon. We'll take our next question.
Thank you.
From Artem Beletski from SEB. Artem, please go ahead.
Yes, good afternoon, and thank you for taking my question. I would like to ask about the highlighted several design wins what you have received in the quarter, so related to Multi-Rail pluggables and also IP in particular. Is it fair to assume that momentum is strengthening what you see on that front, and hence we should be assuming that these wins should be converted into orders, let's say, maybe even later on this year. Just in terms of Multi-Rail and the 800G pluggables, as we are still in early phase of this product ramp-up, how do you see you have been managing compared to competition on that front?
Yeah. Two things I would say. Obviously, if you look at our assumption around growth this year that we've updated, then what we're saying about growth next year, then just the order and momentum we've had in the year, I think it's fair to say we're quite confident that things are moving forward. Things are accelerating, and we continue to build momentum. Obviously, as I said in Q2, we had a huge intake in Q2, and we said at that point we thought this would be a little bit lumpy. But when I look at it, I look at what we see in the pipeline and the funnel, what I continue to see is tremendous demand. As I said just now with Simon's question, our focus is on investing to meet that demand.
Then in terms of pluggables and Multi-rail or just the broader line system market that we play in, this is a place where we were, I'd say very transparently, we were under-penetrated, both as Nokia and Infinera. It's been something as we've come together the team has done, I think, a great job of doubling down on in terms of getting the portfolio right, and now building the capacity. It takes a little bit of time to show up, but I'm probably one of the least patient people for seeing that show up. But it does take a little bit of time to show up, and I'm really pleased with the way this team has come together as one company, one team, to build the portfolio and ramp production.
But that doesn't mean that we don't have more to do, and as I said before, we're not demand constrained today. So if we could ship more, we would certainly do it.
Thanks, Artem. We'll take our next question.
And maybe just-
Go ahead, Artem.
Just a quick follow-up. Related to IP Networks and the nice growth acceleration in the quarter to more than 40%, do you think that such growth pace could be maintained short term, or are you already starting to be constrained on that front?
Yeah. I think we feel pretty good about the assumption we shared in the update, and I will let you guys work through the math. But again, on IP Networks, we are starting from a pretty low base. As I just touched on in Simon's question, as you think about 2027, right now what we see is Optical Networks contributing a little bit more. That is not to say there is not more potential out there for us in IP Networks. It is just what we see today.
Thanks, Artem. We'll take our next question from Daniel Djurberg from Handelsbanken. Daniel, please go ahead. Daniel?
Hi, Justin, Marco, and David. Thanks for the question. I just wanted to ask you a little bit. You mentioned that you saw some increased demand from investments from CSPs on AI traffic, and you also saw a strong Mobile Infrastructure and especially radio demand or revenue in the quarter, with Japan, India, and EMEA being strong, partly perhaps pre-buying. Can you give us some more insight into this and if we should expect this strength to continue into 2027, or if it is a trend with some investments on back of the AI demand?
Yeah. I would say, I think I made some comments on the AI & Cloud demand that's driving telco. Obviously, a fair amount of that is in the U.S. because it's tied to where the data center investment is, and it's driving interconnect. But it's not just limited to that. We're seeing investment in subsea, in AI & Cloud. That also means that players in subsea are more, what you might consider more traditional service provider players in subsea are investing. Obviously, that doesn't have the density that things like scale-across or even DCI have. But as we talked about too, I think this is going to be a place where we expect to see more investment, and I touched on the work we're doing with Zayo, who's one of the operators in this space. On the other side, on mobile, look, I would just say the following on mobile.
I think we touched on this, Marco referenced it in his comments. The trend right now, all of the things we're seeing in demand in AI & Cloud is putting pressure on the supply chain. That supply chain, that's driving impacts, as we've talked about previously, for mobile. It's not just pricing. Obviously, commercial terms and costs are increasing. We're addressing that through greater commercial discipline, as we talked about. But it's also a concern around supply and visibility. So the conversation that I have with customers is around forward visibility. Give us line of sight of your plans over the next 12 - 18 months. Give us early visibility so that we can plan. Some customers take that and say, "If you've got availability, I'll take it," and you saw a little bit of that acceleration. Others are working with us to just give us better visibility.
That is a shift from certainly where we were when I started in this role just 18 months ago, where there was a bit more of a, I would call it a more of a normalized hand-to-mouth ordering cycle, with something like six-month lead times. That dialogue we are having with our customers and obviously, I think we have got some great partners and customers in telco, and they are working with us around that, and I think that is what we are going to have to continue to do. A lot of people talk about the supply constraints in 2021, but that was very narrow, very targeted in some ways to some specific nodes. This is much more broad-based, touches all the leading-edge nodes, touches memory, also touches some of the source materials, PCBs, et cetera.
It is not just a shortage, it is also a structural cost increase that we are seeing in the supply chain.
Perfect. May I just have a follow-up on that? It will be on your Nordic peer, Ericsson, commented about gross margin headwinds in 2027 based on these constraints on the price increases that were a bit harder to work around. Is that your view as well with regards to Radio Networks again, perhaps?
You want to take that one, Marco?
Yeah, I can take it. Hi, Daniel. Just like Justin alluded to as well, there's broad-based constraints when it comes to supply and price increases. We see that as well, and that's why we are working on different fields to try to mitigate those impacts for us as well, and try to secure that we can deliver for those demands that our customers have. Just like Justin said, the telco side is quite aware of this now and getting more and more aware that lead times are slipping when it comes to component supplies, and prices are increasing.
Thanks, Daniel.
Okay, perfect. Thank you and good luck.
Thank you. Thanks.
We'll take our next question from Sami Sarkamies, from Danske Bank. Sami, please go ahead.
Hi. Can you still try to unpack the gross margin dynamics we're seeing at Network Infrastructure? You were at the 43% level in the first half of the year, now at 41%, even though you probably benefited from lower ONT volumes. How material headwind are you seeing from new products? What will you be thinking about Q4 when you had 45% last year?
Yeah. Thank you, Sami. I would say that it's always when you have new products you're ramping up, it takes some time before you get the right cost base and profitability on those. That's why we saw also in quarter three now, specifically IP Networks side, that we have this situation. We believe that this dilution that we've seen is temporary, and also that in quarter four we should see improvement in the gross margin levels. AI & Cloud bring very strong incremental gross profit, and is accretive to operating margin as well. We believe this is very good business for us going forward. Your comment on ONT, we are working throughout the whole year now to prune the portfolio and basically securing that our portfolio and offerings going forward are the best suitable for us.
That's why we are taking down those low-margin customer premises equipment offerings. But because of some longer-term contracts which are working throughout the year, it takes throughout this year to be able to do that. That's why it takes some time before you can see the impact as well.
Yeah. Sami, I would just maybe make two comments on this as well. One is, if you look at the Fixed Networks business in Q3, effectively the decline was the reduction in ONT volume. Then obviously the growth came across Optical Networks and IP Networks. But it is a bit of an addition subtraction. It does not move the margins that much because we are not talking about business that was loss-making, right? We are just talking about business that was dilutive to our profitability. So I think it is a little bit of that probably in addition to what Marco highlighted, that it does not move the needle as much as we would like.
But I think over time, as we scale what we are building in IP Networks, and obviously continue to scale in Optical Networks and, as Marco said, play through the transformation that the team is driving in Fixed Networks, I think that is where you will start to see the healthier margin profile.
Also remember that size-wise, Fixed Networks is much smaller than Optical Networks and IP Networks. ONT business, then customer premises equipment business is even smaller.
Thanks, Sami. We will take our next question from Richard Kramer at Arete. Richard, please go ahead.
Thanks very much, guys. Justin, you just mentioned visibility, and you've given a roughly EUR 5 billion AI & Cloud order book in the last five quarters, but we don't know the timing. Could you maybe shed some light on roughly what portion of those orders are for 12-month delivery versus long-term deals? When, given Marco's comments about near-term dilution, do you expect that AI & Cloud, which is now 12% of sales, translate into neutral or even accretive margins? Is that 2027 event or is that further down the road? Thanks.
Yeah. Hey, Richard. Why don't I take the first part, Marco, you take the second. In terms of the growth that we see in the order book, what we continue to see is the majority of orders we've seen are in a 12-month timeframe. There's some portion that are moving outside of 12 months, but I would say still largely within 2027. The demand, and again, the constraint here isn't demand, it's supply. Some of this is obviously our customers would take more if we can deliver it faster. Obviously, it was implicit in my comment. But for now, the way I would think about it is we have a lot of backlog towards building towards 2027.
Yeah. When it comes to profitability, as I said, we definitely believe this is accretive for the company, and this is why we're investing in this as well, and believe that this is very accretive for our shareholders also. It's always so that, especially if you look IP now that I mentioned specifically for quarter three, we are starting from quite low levels. So it takes some time before we get that volume up, and when we get the volumes up, then, of course, you get more operating leverage as well and better profitability. So that's why you saw quarter three little bit lower margins. But as I said, we expect quarter four to be better than quarter three.
Thanks, Richard. We'll take our next question from Emil Immonen from DNB Carnegie. Emil, please go ahead.
Hi, guys. Thanks for taking my questions. Maybe to touch a bit still on Mobile Infrastructure where especially Radio Networks you are seeing quite good sales. Just wondering if your customers are seeing cost inflation, are they then doing a lot of pre-buying now, or what is your view on that market going into Q4 and then 2027?
Yeah. Emil, it is a good question. I do not see people pre-buying, and the reality is, given the constraints, I do not think we could pre-supply. If there was a huge bubble, one, I think we would all know about it because you would see it in the numbers. Two, given the supply constraints, it is not like we can just respond and support it. But what we do anticipate is that we are going to see more stable forecasting. I think the other thing that I do not see is I do not see an imminent investment cycle. I think the market continues to be flat as we look ahead, and we have been pretty vocal and transparent about this. I think that is consistent with what others see. You may have a certain market that drives demand up or down.
We have talked about certain markets that things slide, and that will continue, but overall, we expect the market to be flat. However, given the cost inflation, we expect pricing to go up.
Yeah. There are some pockets of opportunities, as I mentioned earlier, IP and optical side on telco side, but the traditional telco, we expect to be quite flattish market. Also back to your comment about pre-buying. The quarter three movements that we had from quarter four, specifically on the radio side, I would say that it is not pre-buying, it is just that sometimes projects are proceeding faster than expected, and that is why we were able to deliver and conclude those deliveries already in quarter three instead of quarter four.
Thanks, Emil. We will take our next question from Felix Henriksson from Nordea. Felix, please go ahead.
Hi. Thanks for squeezing me in. I appreciate the initial guide for Optical and IP growth for 2027. How would you describe the expected operating leverage on that growth? I know we have discussed about gross margin already a little bit, but are there any larger incremental OpEx needs to enable this growth going into next year? Thank you.
Yeah, thank you. I can take that one. Of course, the product mix has an impact on gross margin. As I said earlier, we definitely see improvement in NI. What comes to all the different mitigation actions we are doing and focusing in not only in NI, but the whole group, to secure that we are getting the operating leverage and mitigating potential cost increases on component side. We definitely believe that we have opportunities for all parts of the company to improve the profitability going forward. Also, what comes to the capacity investments, we invest in those areas where we believe that we can create shareholder value in a long term. If those opportunities occur, we will definitely take those opportunities and invest in those. This is something that we do actually throughout the whole year on the capital allocation decisions.
It is not a once per year decision. This is actually ongoing running discussions that we have on a weekly basis, basically, that where should we allocate our resources and capital, and we make those decisions on a running basis, basically.
Maybe I'll make. Can I add one comment to Felix's question? Couple of things. One is we touched on the fact that we're starting to see operating leverage in Q3. Again, just to remind you, if you look at where we are, this operating leverage that we're starting to see is also hidden by the fact that we've got a bunch of these portfolio moves. The ones that we've talked about quite openly in Portfolio Businesses, and obviously the ONT one we were talking about earlier with Sami. But also the continued cleanup of some of the areas where we're focusing on profitability, even inside Core Networks and in Radio Networks. That hides this a little bit in the short term as we make those decisions.
Also if you look back, where we're highlighting the operating leverage of the quarter, we're also saying we're now moving to above the midpoint of the range, which is a signal up from where we were last quarter, because we're starting to see that potential improved profitability. That's a sign from the set of actions we're taking, both in cleaning up the portfolio as well as some of the more formal restructuring elements we've talked about.
Thanks, Felix. Let's take our next question from Jakob Bluestone from BNP Paribas. Jakob, please go ahead.
Thanks for taking my question. I have a question around expansion of your fab capacity. You obviously talked about the ramp in San Jose. I was wondering if you could talk a little bit about the ramp at Chandler. I appreciate it hasn't closed yet, but if you could maybe share a little bit, what are your plans? What will it focus on? Is it just indium phosphide? I guess, given how much capacity you're adding, can you maybe comment on how closely are you working with your customers to support the decision to have so much manufacturing capacity sort of in the midterm? Thank you.
Yeah, maybe I'll start on that side and then work the other way. I would say we talked about this a little bit at ECOC, or the team talked about this a little bit at ECOC because I wasn't there. But one of the things as we think about is we've got a set of building blocks and capabilities in optical that give us a lot of flexibility in terms of building solutions. What's becoming more and more apparent is that optical penetration and demand is going to continue to increase. It's a factor of the tremendous increase in bandwidth demand that's going to touch every part of the network that we serve today, and then obviously the power efficiency for solutions. You look at things even like what we're doing in out-of-band management as an example of that.
The reason I say that is that's enabling us to have earlier discussions around solutions and really co-innovate and co-develop these platforms. If you go back to OFC, we announced products that we're going to release in 2027. That's a sign of the left shift that's happening and the longer-term visibility that our customers have on the demand. The key thing there is building the fab in Chandler, the investment, the acquisition we're making from NXP of that facility. Right now, I'm not going to give you specific metrics. What I will tell you is we're positioning that to give us the flexibility to support the demand as it emerges. We've talked about that being a 2029 timeframe solution. But we think we've got lots of different avenues to monetize that facility through indium phosphide manufacturing.
Obviously as we get closer and we complete the transition and we work on it, we'll talk more about where we are, and we'll give you similar milestones to what we're giving you on Fab 2. But in terms of near-term execution, the focus right now 100% is on making sure we get Fab 2 up and running and supporting the growth and, of course, hitting the yield and output targets that we expect.
Thanks, Jakob. We'll take our next question from Janardan Menon from Jefferies. Janardan, please go ahead.
Yeah. Hi. Thanks for taking my question. Yeah, two questions. One is just a follow-up on those opportunities that you are seeing in the medium term on the Optical Networks side. You have come up with some products for optical solutions inside the data center, outside of your core area of scale across DCI, et cetera. I am talking about things like your LPO pluggable and your external laser, et cetera. I know it is very early days, but can you give us any indication of how the response to those kind of products has been from potential customers? Do you see those kind of products as being a reasonable contributor to that medium-term outlook that you talked about? I will just give my second question also right away. It is again on margins, but this time just looking at the restructuring where you are accelerating it to EUR 1.1 billion.
When should we expect to start seeing the results of this on your profitability? Does this restructuring start to taper off next year, and then gradually we see the effect of that on profitability? Would that be coming more on the Mobile Infrastructure side where some of that is more focused, or will we see some of that coming through on the NI side as well? Thanks.
I would just say on the Optical Networks side, I think I just answered the question. Right now what I will tell you is that the assumption that we provided for 2027 doesn't assume anything outside of the portfolio we have and the orders we have. Of course, as I touched on, we see broader opportunity and right now we just want to continue to work with those customers to design the right solutions that they need.
Yeah, what comes to the restructuring, if I start with China, where we basically do the integration after we acquire the remaining part of the company. We said originally that this will take up to three years, but now we are actually accelerating, and we are able to do that in 12 - 24 months, the whole integration. That is why we get more and faster the savings as well, and synergies there. This is, of course, aligned with restructuring costs. As you know, in China, as a Western company, it is quite limited what you can capture as a market share there. That is why we are adjusting basically the company to that reality that we see. The other parts that we are doing, of course, we are finalizing the 2023 to 2026 program, and here we have actually maximized what we set the range.
We set EUR 800 million -EUR 1.2 billion, and now we are at the top of that range, what we expect to get. In addition, in quarter two, we announced that we see further opportunities to streamline the company. Also, what comes to securing, we consolidate locations in Europe and other parts of the world. Just securing that we continuously seek for more efficient way of doing the business we are doing. This will definitely help, I would say, the whole company, not only MI, but of course, considering also what you have heard from other companies said that there is component strains and component price increases. This is one way to mitigate those also going forward. Also in addition to what we already do when it comes to, of course, discussing with our customers, redesigning our products and so on.
There is definitely benefits in different parts.
Thanks, Janardan. We will take our next question, and possibly last question, from Sandeep Deshpande from JP Morgan. Sandeep, please go ahead.
Yeah. Hi, thanks for letting me on. I have two questions, if I may. Firstly, regarding IP Networks, you are talking about two hyperscaler design wins. You already have multiple hyperscaler design wins in the IP Networks business. Are you expanding the number of customers that you have in IP Networks? Thus, do you expect that through 2027 and 2028, IP Networks will be a bigger percentage of your orders and sales in the AI & Cloud segment? Secondly, when you look at the kind of typical Q4 revenue growth that you have, you are already seeing some operational leverage in the business, in Network Infrastructure associated with the growth in Optical Networks and IP Networks. How should we look at your guidance for the year? Can you go to the top end?
Can you reach the top end of that EBIT guidance that you've given the kind of now leverage you're seeing in this business?
Will you start or-
You want me to start? Okay. I think you were trying to ask a 2028 question and a profit warning question, Sandeep. Neither which I'm going to give you. Look, I would just say in terms of the design wins, the way I think about it, I don't think about it at a customer-by-customer level. I think about it at a design win-by-design win level. Because each one of these is almost the magnitude of the product. Each design win in each customer is like its own product revenue stream and product life cycle. That's why you hear us talking about it that way, and that's the focus. Obviously we're given the information we gave you on 2027, and the ongoing demand that we saw in the order book we saw in Q3 alone on the back of what we did in Q2.
We feel very good about the long-term potential of this business. The feedback we get from the customers that we're working with is that we're quite competitive in our design work. So, obviously that gives us optimism to continue to invest. Then in terms of operating leverage. Look, my view, and Marco can comment, is this is something that, as you guys know, takes a few quarters to work out. It never comes right away. There's a lot of work to get Nokia to the level of lean operating machine with the kind of different discipline portfolio. We're not there yet. I'm incredibly proud of what the team is doing because I think they've done a great job and they've moved faster than I thought in terms of portfolio clarity. But it doesn't mean that our work is done.
We are going to continue to push hard and stay focused on providing that ongoing operating leverage as we drive scale, because that is going to be fundamental.
Yeah. When it comes to the full-year guidance, just like Justin said earlier, we changed a little bit the wording on the full-year guidance. But the range is what the range is that we have said as well. Just consider also that linearity, especially on the MI side, is a little bit different considering that we saw these, some of the movements from quarter four to quarter three. But I think that we have given as good, I would say, path for you to see and calculate what you expect that where we would land.
Thanks, Sandeep. Ladies and gentlemen, this concludes today's call. I would like to remind you that during this call today, we have made a number of forward-looking statements that involve risks and uncertainties. Actual results may therefore differ materially from the results currently expected. Factors that could cause such differences can be both external as well as internal operating factors. We have identified such risks in the Risk Factors section of our annual report on Form 20-F, which is available on our investor relations website. Thank you all for joining us.
This concludes today's call. Thank you everyone for joining. You may now disconnect.