Welcome to the Delivery Hero Q3 2025 trading update. Today's presentation will be followed by a Q&A session. For those of you who have joined the Zoom webinar, you can use the raise hand function at the bottom of your Zoom screen at any time to join the queue to ask a question, and you will be called upon during the Q&A session. If you have dialed in, please press star nine to enter the queue. If you want to withdraw your question, please lower your hand using the raise hand function in the Zoom app or via the telephone, press star nine. I will now hand over to Christoph Bast, Head of Investor Relations, to begin the presentation.
Hello and welcome, everyone. Thank you very much for joining our Q3 2025 earnings call. Joining me on this call today are Niklas Östberg, CEO, and Marie-Anne Popp, CFO at Delivery Hero, and together they will present the key highlights of our Q3 2025 results. Following the presentations, we will be delighted to address any questions you might have, and now over to you, Niklas.
Thank you, Christoph, and welcome, everyone, and thank you for listening in. We have a solid quarter behind us and entered Q4 with momentum. We have returned to order growth in Korea. We have excellent numbers in Saudi Arabia and very strong development in our multi-vertical offering. This current development gives us confidence to accelerate our strategy. We will be focusing our investments in three key areas: expanding our multi-vertical offering, enhancing our customer value proposition, and investing in strategically important markets. We have confidence this is the right strategy to build on our momentum and drive sustainable long-term value. Later in the call, we will share more specific areas where we plan to lean in. Now, let's start with an update on our global technology platform, which represents our greatest strength and a key differentiator in our industry.
What makes our global platform unique is that it is a single unified platform across 65 countries, exactly like Uber Eats, or DoorDash, with the important difference that we have enabled deeper localization for multiple brands across all verticals. This platform provides world-class logistics, ad tech, customer service, AI personalization, search, payments, quick commerce, partner integrations, and much more. Last quarter, we finished the integration of Glovo, demonstrating the power of our platform with significant operational improvements reflected in a better customer experience, increased delivery efficiency, and a stronger advertisement business. With the ongoing integration of Uber, we are working on the integration of the last brand to our global platform.
The first regions in South Korea have already been migrated with promising results, such as significantly increasing the numbers of deliveries per rider per hour while still delivering slightly faster, which points to significant delivery efficiencies that lead to higher rider earnings while also lowering costs for us and vendors. While business enablers are fully globalized, our global platform also allows for unprecedented deep localization of the customer experience. This enables us to fully embrace uniquely local customer preferences without any global one-size-fits-all compromises, leveraging our leading local heritage brands much better and faster than our competitors. To give you some examples, we are offering a fully customized Ramadan experience in MENA, Curbside Pickup in KSA , special Student Discounts in Greece, a Customer Loyalty Program in Turkey, Group Ordering in APAC, and so much more.
It is this unique combination of fully globalized business enablers and deep localization of customer experience that is the heart of our platform's competitive edge. Another pillar of competitiveness that we have been building for some time is AI personalization. We firmly believe in end-to-end personalization. We want to offer a customer experience that is individually optimized and curated for each specific customer, thus becoming much more relevant, inspiring, and engaging than any unpersonalized experience. Over the last two years, we have developed a proprietary global AI personalization platform that leverages state-of-the-art AI algorithms to process all of customer, vendor, and product information, continuously learn how to personalize the experience for each individual customer. With millions of customers in around 65 countries, every day this platform processes more than 10 trillion features. The result is that all relevant aspects of our customer journey and our apps are becoming personalized.
That goes for search results, recommendations, deals and promotions, ads, delivery fees, vendor and product lists, and so much more. This end-to-end AI personalization of the customer experience leads to significantly increased conversion rates, higher average order volumes, and improved customer lifetime values. The deployment of AI has also been impactful when it comes to cost. If we move to the next page, here we can see that we are laser-focused on improving our operational leverage by increasing efficiency through deploying AI automations. It is fully integrated throughout our business and constantly improving our performance. Over time, we have reduced SG&A and marketing costs as percentage of GMV from 7.2% in Q1 2023 to 6.0% in Q3 2025, reflecting measurable efficiencies and smarter marketing spend. Looking at the ratio G&A, including research and development costs to GMV, we are now world-leading in our peer group.
This improvement is supported by our AI and automation roadmap, which began in 2023 with self-service for customers, vendors, and riders. In 2024, we introduced Co-Pilots for service agents and productized incentives. By the end of 2025, we will have implemented vendor-funded deal optimization and AI content optimization. In 2026, we will focus on agentic services and sales along with AI-driven incentive optimizations. These steps position us for sustained margin expansion through automation and personalizations. Let me now hand over to Marie-Anne, who will guide us through the financial highlights.
Thank you, Niklas, and a warm welcome from my side as well. Q3 2025 was a strong quarter with continuous growth in our three main focus areas: Growth, Profitability, and Cash Generation, despite tough comparables. GMV in Q3 increased by 7% year-over-year on a like-for-like basis, excluding hyperinflation and FX effects. The slightly softer growth in Q3 was due to a strong Q3 last year following growth initiatives in South Korea and MENA. However, the GMV development is set to accelerate again in Q4, driven by Asia's recovery and robust demand across key markets. Revenue generated a plus of 22% year-over-year on a like-for-like basis, growing again markedly faster than GMV. In addition to the top line, adjusted EBITDA in Q3 2025 further increased, driven by stable gross profit margin as well as strong cost discipline.
Free cash flow continued to improve in Q3 2025 and is well on track to meet the full year guidance of exceeding EUR 120 million. Our capital position remains strong, with EUR 2.2 billion in cash at the end of Q3 2025, reflecting the convertible bond buyback of nearly EUR 900 million earlier this year and the net outflow for extraordinary items of around EUR 500 million during the first nine months. Let us now take a closer look at the individual building blocks of the Q3 performance. As of this quarter, we will show order growth on a group level. In Q3, orders grew by 8% on a like-for-like basis, with double-digit growth in all segments outside of Asia, outperforming GMV development. As mentioned, GMV growth for Q3 reached 7% on a like-for-like basis in constant currency and excluding hyperinflation accounting.
The slight deceleration compared to Q2 can be traced back to strong comparables following pre-subscription trial in South Korea and growth initiatives in MENA. In Q4, growth is expected to pick up again, driven by Asia's return to growth. Revenue growth came in at 22% and has consistently exceeded 20% at the group level for several consecutive quarters. The strong growth was driven by the ongoing expansion of own delivery logistics, especially in South Korea and Turkey, as well as the change in the rider model in Spain. Furthermore, the continued strong performance of our ad tech business, as well as the attractiveness of our subscription programs, have contributed to this development. Let's now dive into the Europe segment, where we temporarily scaled back GMV growth to manage the initial efficiency impact following the transition to an employment model in Spain.
Performance outside of Spain remained strong, with category share gains as well as healthy GMV growth in the majority of markets. Revenue growth in the Europe segment was driven by the expansion of our own delivery logistics, with OD share increasing by 8 percentage points year-over-year to reach 82% in Q3 2025. Furthermore, the introduction of the new rider model in Spain and the associated change in revenue recognition resulted in a higher take rate. In addition, we successfully adjusted the rider model in Italy in line with new regulations and completed the global transition to Delivery Hero's tech stack, resulting in improved delivery times, lower failure rates, and higher operational efficiency. We maintain a strong profitability outlook, with adjusted EBITDA expected to reach near break-even in Q4 2025. Let's move on to MENA.
We delivered very robust GMV growth despite challenging prior year comparables, which had benefited from heightened growth investments. Our competitive playbook in Saudi Arabia has worked out very well. We see strong performance with significant outperformance versus local peers. More on this later on. talabat sustained strong performance with GMV growth of 27% year-over-year in Q3 2025, driven by order volume growth across markets and verticals and supported by highly effective partner-funded savings, which reinforce a unique competitive advantage. In addition, the multi-vertical offering continues to thrive, with over 70% of the GMV now being generated from customers who order food and groceries. Furthermore, talabat's loyalty program continues to grow, and the share of GMV generated by subscribers is now accounting for nearly 50%.
With the increased competitive environment, we aim to apply a similar competitive playbook as in Hong Kong and Saudi Arabia by making some incremental investments into our service offering. We also believe that the regulatory environment in the MENA region addressing predatory pricing is moving in a positive direction. Turkey significantly improved profitability through a strong increase of vendor-funded deals and gross profit improvements. This led to positive adjusted EBITDA in Q3 2025, with further earnings growth expected in Q4 2025. Now on to the Asia segment. For Asia, the picture is slightly mixed. GMV trends in Korea were constrained by high comms in Q3 2024, given a full quarter of free delivery and free subscription trials. However, both the subscriber order share as well as subscriber frequency continues to increase at double-digit rates.
For the remaining part of Asia, we have seen a strong growth trajectory across the region, with Hong Kong being a large outperformer with elevated growth levels. Today, we're significantly larger than when competition started to heat up two years ago. Moving on to revenue. In line with previous quarters, the ongoing robust revenue growth of 17% on a like-for-like basis was primarily driven by the rollout of own delivery operations, which now account for 75% of orders in the Asia segment, an increase of 18 percentage points. Adjusted EBITDA in the Asia segment continued to grow year-over-year in Q3 2025. As you might recall, our operations outside of Korea under the Foodpanda brand have been generating positive adjusted EBITDA before group costs already for several consecutive quarters, and we saw further margin expansion in Q3.
Looking forward, the Asia segment had a strong start into the fourth quarter, with a number of orders in Korea returning to growth during October and trends further improving in early November. It is setting the stage for overall GMV growth in the Asia segment in Q4 on a like-for-like and constant currency basis. Now continuing with the America segment. The top line development again highlights the sustained momentum we have in this region, with GMV growing 19% year-over-year, driven by 21% order growth from both new user acquisition and increased order frequency. We continue to expand our quick commerce and subscription offerings to reinforce our value proposition by driving deeper customer engagement and broadening our multi-category offerings. Revenues in the segment grew 19% year-over-year in the third quarter, with ad tech outperforming the overall top line growth and offering additional upside potential going forward.
As profitability continues to improve, the America segment is demonstrating resilience amid current macro headwinds like currency devaluations in some of the segment countries, with adjusted EBITDA continuing to expand year-over-year during the last quarter. Now on to integrated verticals. Our quick commerce business continues its rapid expansion, fueled by 24% year-over-year growth in Dmarts and even faster growth in local shops, boosting annualized GMV to more than EUR 7 billion. Overall, the integrated vertical segment significantly enhanced its profitability, achieving its first-ever positive quarterly adjusted EBITDA in Q3 2025. The business remains on track to achieve adjusted EBITDA break-even for the full financial year 2025. We're big believers in the segment, and in 2026, we aim to add more stores and reinvest incremental profit contribution to drive customer experience further. Let's now have a closer look at the gross profit margin development on group level.
Overall, our gross profit margin on group level continued to increase by 40 basis points year-over-year to 8.0%, primarily driven by better margins in Korea and the scaling of the integrated vertical segment. Assuming the elevated competitive environment persists, we might consider to keep our gross profit margin flat during 2026 and to expand margins as and when the competitive environment eases. Our MENA and America segments are already achieving an attractive gross profit margin of around 10%, while leveraging profitability further to expand rapidly in the quick commerce space. Gross profit margins for the Asia segment continue to improve with an expansion of 90 basis points year-over-year in the third quarter of 2025. This was mainly driven by improved unit economics within our own delivery business.
The quarter-over-quarter decline is primarily due to the monsoon season in Southeast Asia, leading to temporarily elevated delivery costs in line with our planning. The Europe segment faced a temporary negative impact due to the rider model transition in Spain, which led to short-term elevated delivery costs, but has started to show gradual recovery. Further margin expansion for this segment is anticipated in the fourth quarter. As part of our continued efforts to streamline financial disclosures, starting 2026, we will report gross profit only in accordance with IFRS and on a semi-annual basis. Niklas will now take you through our case studies on Korea and Saudi Arabia.
Thank you, Marie-Anne. As you're aware, we made substantial improvements to our South Korean business over the past 18 months. We'd have clear focus on enhancing the customer experience and returning to a growth trajectory.
We improved our logistic capabilities, introduced our subscription program, simplified platform architecture, several product innovations, and drove UI improvements. All in all, we had to make some radical changes in the past 18 months, and I'm very, very pleased to see these efforts bearing fruits. On the left-hand side, you'll see order development presented as a 14-day trailing average growth rate. Based on this, we have seen an uptick in order volumes returning to year-on-year growth during October and early November, while maintaining stable category share. We remain on track to reach growth for the quarter. We now know that our growth playbook works, and growth will remain our priority in 2026. As a result, we would expect to maintain adjusted EBITDA around current levels in local currencies. We feel confident in doing these investments as we see clear, tangible results.
Some examples of those: we introduced and expanded our offering for low-frequency users, new users, and churned users, which led to new acquisitions going up by 8% year-on-year and 25% quarter-on-quarter, as well as reacquisitions up 12% year-on-year. Our subscription program: we continuously enhance our value proposition through new products like Meal for One and partnerships with Tving and YouTube, which are extremely popular in South Korea. Around 80% of Koreans use them regularly. As a result, the number of subscribers doubled year-on-year, with a 30% increase in order frequency since the beginning of the year. Delivery experience: own delivery service share increased by 30 percentage points year-on-year, of which we have begun to see the positive impacts of. This was only possible as we made significant improvement in economics over the past 18 months. Let's have a closer look at Saudi Arabia.
What you can see, Saudi Arabia exhibits a very resilient customer base despite increasing competition. The order growth increased by around 15% before and 14% after the Keeta entry one year ago, reaching now almost 50 million orders per quarter. The customer mix continues to improve, with the number of high-frequency customers growing the fastest at 19%, and medium-frequency and low-frequency customers growing at 10%, all of which exceeds last year's growth rates. Growth is driven by a very strong product offering, which is much larger than just food. Quick commerce customers' adoption went up 1.5x year-on-year, and the share of quick commerce as a percentage of GMV doubled year-to-date, with a huge upside potential going forward.
We have also seen subscription adoption triple year-on-year, supported by multiple new initiatives like attractive discounts on the top three items, reduced delivery fee, pick-up, Meal for One, or our loyalty program. Subscribers now account for 39% of GMV, with further room to grow. Current growth has come with some incremental investments, and we will maintain these investments during 2026. With this, Marie-Anne will now take us through our guidance.
Following the solid results in Q3, we're confirming our 2025 guidance, which was updated during the last trading update. We continue to expect GMV growth to come in at the upper end of the 8%-10% growth range year-over-year and on a like-for-like basis. Total Segment Revenue growth was upgraded to 22%-24% on a like-for-like basis, and we continue to see it in that range for the full financial year 2025.
The adjusted EBITDA result is expected to come in between EUR 900 million and EUR 940 million. In line with what we communicated in August during Q2 results, this includes around EUR 110 million of FX headwind coming from the second half of the year. Excluding this negative impact, adjusted EBITDA would be expected to come in somewhere between EUR 1.01 billion and EUR 1.05 billion, based on FX at the time of initial guidance from February 2025. Lastly, we continue to expect free cash flow to be more than EUR 120 million, including the previously communicated negative FX headwind of around EUR 80 million. That is it from my side. Thank you for listening, and we are now looking forward to taking your questions. Christoph?
Thank you very much, Marie-Anne. It's just a usual reminder from my side, so we would kindly ask you to limit your questions to one panelist, maybe one and a half panelists, because this way we can ensure that every analyst has the opportunity to ask a question. That is it. Operator, please go ahead.
Ladies and gentlemen, we will now begin our Q&A session. For those of you who have joined the Zoom webinar, if you have a question, we ask that you please use the raise hand function at the bottom of your Zoom screen. Once called upon, please unmute your audio and ask your question. If you have dialed in, please press star nine to enter the queue and star six to unmute once called upon. Our first question will come from Joe Barnet-Lamb with UBS. Please unmute your line and ask your question.
Excellent. Thank you very much for taking my question. So Niklas, you spoke about leaning in to accelerate your strategy and touched on some really interesting initiatives. You also dropped a few comments about ongoing investment in a number of regions. I'm wondering if this points to heightened investment into 2026. Could you talk a little bit about what these initiatives might mean in aggregate for 2026, and perhaps any color you can give us on adjusted EBITDA at the group level for next year? Thank you.
Thanks, Joe. At this stage, we're not ready to give next year's guidance. We typically provide our guidance in Q1. Having said that, as you mentioned in our prepared remarks, we are making some investment next year, such as continuing investment in improving growth initiatives in Korea, investments to push customer experience in UCC to maintain customer loyalty, multi-vertical offering is another one, and in particular, Dmart efforts. I would say, despite this and the elevated competition that we're seeing, we still expect to deliver both top-line growth and moderate bottom-line growth next year, while further improving cash conversion. We believe, or I believe, that this indicates the fundamental strength and resilience of our business model and our team's ability to execute.
Excellent. Thank you very much.
Thanks.
Our next question comes from Luke Holbrook with Morgan Stanley. Please unmute your line and ask your question.
Good morning, everyone. Thank you for taking my question. It would just be on slide 18, the order trajectory in South Korea. There seems to be a bit of a dip in that growth rate towards the end of October. It then suddenly reaccelerates. I'm just wondering if you could provide a bit more color on whether that's a function of more promotional campaigns and timing of that. Is it a function of comps? Then just how we think about that trajectory as we go through Q4. Thank you very much.
Thank you. Sometimes this can be that there's an extra holiday or something like that. I can't remember what it was exactly. I know we're speaking about it came down one or two percent, but it's often just one day that it's some special event. I can't remember exactly what it was.
You would expect it to continue at this current level or maybe even come up a little bit during the end of November, possibly also beginning of December. It might come down a little bit from mid-December again, as we did some big push last year during the end of December, early January. Yeah, the direction is clear and it's upwards.
Understood. Thank you.
Thanks.
Our next question comes from Marcus Diebel with JPMorgan. Marcus, please unmute your line and ask your question.
Yeah, hi. Similar on Korea, a question. I mean, the roughly 2%-3% that you're seeing, could you help us to maybe understand conceptually what it means sort of like longer term? Do you think there are some elements of sort of like, yeah, initial one-off effects? Obviously, you've done a lot, as you said, in Korea.
Do you think this is just about to start and we should assume at least conceptually a higher number than the 2%-3% that we just saw in recent weeks? That is my main question. My half a question, as you reflected, is just a follow-up on Joe. You said the EBITDA is going to grow at a moderate level. I did not fully catch that. Was that meant for just Korea or was that meant for the group? Sorry for being ignorant, but it is obviously important. Just wanted to make it clear.
Yeah. On the first one, yeah, no, there is no one-off effect or certain special things that we do. Of course, we continue to do promotions with our partner, certain vendor-funded deals. I do not know. That is an ongoing thing that will also be there in 2026 and beyond.
I don't think there's anything out of the extraordinary there. I don't know. It's just hard work over a long time that is bearing fruits. As we said, we'll keep leaning in in Korea and prioritizing growth for next year. We'll continue on that direction. There will always be a month here and there where things can go a little bit up or down. That's normal. The direction is clear and we'll keep driving that. Yeah, my view, I don't know, is more on an aggregated base for Delivery Hero. As mentioned, we are leaning in a little bit more in Korea, but we're also leaning in a little bit more in GCC, where we really want to protect the customer experience and maintain the customer loyalty that we have.
We also see that multi-vertical, as Marie-Anne mentioned, is an area that we truly believe in, that makes us very unique, and that is significantly improving the customer experience for our customers where we offer that. These are the things that really make us unique and why we can fight back so good when we compete against a player who spends hundreds and hundreds of millions and still does not impact our customer base. We will keep leaning in on those things and drive that customer experience because we think that is the cheapest way of maintaining share. That also means that there are certain investments next year. I would say, despite this elevated competition and despite the investment we are doing, we are still going to deliver at least a moderate bottom-line growth on the bottom line while also growing the top line.
But yeah, that's an aggregate, I would say.
Okay, thank you. I would like to ask you what moderate means, but I'm sure that's difficult. Okay, thank you.
Yeah, look, I don't have a number for what moderate is, but I would say moderate is more than stable or flat, putting it this way.
Okay, super. Thank you.
Our next question comes from Andrew Ross with Barclays. Andrew, please unmute your line and ask your question.
Hi guys. My question is from Saudi Arabia. Thanks for the slides with some more detail about Saudi Arabia. Why do you think conversations past with Shia is compared to a year ago?
Andrew, it's really hard to hear you. Sorry.
Yeah, again, the hard question is based on what we've seen so far.
Andrew? Andrew, there seems to be some connection issues.
Can't hear him.
Andrew, it seems to be some connection issues.
Operator, can we just move on with the next analyst and then we try Andrew?
Is that better now ?
No, we can hear you. [crosstak] Andrew, go ahead. Andrew, you are still there?
Our next question will come from Giles Thorne with Jefferies. Giles, please unmute your audio and ask your question.
There we go. I was muted. Look, I think Andrew's first question, I will ask it, was just category share for Hungerstation today versus where it was a year ago. I am sure he will come back around again, but I certainly heard that much. If you could answer that. My question was also on Saudi Arabia, and it would be useful to know what the contribution margin is looking like on orders at the moment.
The order growth that you're reporting certainly looks very impressive, but it'd be useful to know to what extent it's a tailwind or a headwind to EBITDA. Thank you.
Thanks for. We are not looking that much into categories here, at least not in this super promotional case where every cleaner and rider is getting money to feed their families. It's a little bit irrelevant. What matters is what is the category share of actual orders that would have happened without promotion. And there we would have to make a lot of calculation to get there, and it gets very complicated. But measuring category share when someone is buying orders is a little bit irrelevant. Also, what we are focusing on, and I think way more important, is are we losing any of our customers? If we're losing any of our customers, who are those customers?
Do we lose them for one order, or do we lose them for many orders? As you have seen there in the slides, we maintain our best users. We keep growing our best users. There is a little bit of churn on low-customer orders, low-value customers. I think you have also seen from some of our local peers who actually have a decline. I do think that we have a stronger loyalty in our product and service than what you have seen in the local peers. Possibly also some additional investments were done in our product offering. Based on that, we feel very comfortable. We feel very good. We are growing. Customers are loyal to us. I think we see a little bit similar, even more so in Hong Kong.
We took a big dip in the first 12 months, 6 to 12 months, where we lost a lot of low-value users. Year two was improving. We start, and we can only lose a customer once. That is good. We start seeing stabilization. Year two, we have seen rapid growth. Right now, it is one of our fastest growing markets. As Marie-Anne said, it is significantly larger now than when Keeta entered. That makes us also comfortable for Saudi Arabia for the next year or two. Initially, it means some incremental investment. It still means that our margins are good in Saudi. Maybe we took down margins a little bit, but also thanks to growth, it means that EBITDA was maintained more or less stable last year. I hope that helped a little bit to answer the question.
Yeah. On the contribution margin, please, that you're seeing currently in Saudi Arabia.
Yeah, it remains good. It's roughly in line, a little bit lower than what it was pre, as we are making a few investments in a few areas also on the consumer side. It's a fairly small difference. Most of our investments have not necessarily been on that or have been compensated otherwise. It's a small margin reduction, but thanks to growth, overall EBITDA is roughly stable, marginally down, but yeah, fairly stable, I would say.
Just to follow up, and I appreciate that's question number three, but one of them was Andrew's. Just to follow up, we're now quite deep into the Meituan phenomenon.
Do you think the body of evidence, Niklas, supports the idea that as long as you have product parity or product leadership against Keeta, the investment that they're putting in is effectively accelerating changes in consumer behavior? As an incumbent with product parity or product leadership, you actually benefit from that long-term? Is the jury still out on that idea?
I do not want to be too optimistic here, but I think there is a clear argument. Also, if you look at Hong Kong and we look at Turkey and some other places, you are taking a hit short-term because you are losing some of your least valuable customers. That means also that helps your gross profit. To your earlier point, it actually helps our gross profit that we actually also then reinvested in some high-value customer to make sure that we keep loyalty there.
As long as we have a strong product offering, and I think we do, I think we have a substantially better product offering than all other players in the region. We are no longer just a food delivery company. I do not think that we necessarily compete head-to-head. I think there is a valid argument to make that those customers who are low-value customers, maybe even some other customers who got hooked by delivery, will become real customers. Once this kind of initial voucher spend is kind of phased away, some of those customers will come back, and you will have a little bit of a tailwind over time. Again, let's not be over-optimistic. I do not know. We have a good development. We will keep that good development.
If there is some tailwind in two or three years, or maybe one or two years, that's great, but let's not bank it in yet.
That's great. Thank you very much.
Thanks.
Our next question will come from Monique Pollard with Citi. Monique, please unmute your audio and ask your question.
Hi, Monique, everybody. Thank you for taking my question. The question I have was just on Korea. So your own delivery share, you're saying now of orders is up 30% year on year. So am I right in thinking about three quarters of your orders now in Korea own delivery? And just to follow on, I guess if the rest of Asia GMV was growing 8% like-for-like in the third quarter, we can see kind of low single digit so far in Korea. So mid-single digit growth, is that reasonable for GMVs as we go into the fourth quarter?
I don't know what we disclose exactly on this other part of Asia, but I think your assumptions are sensible. As we mentioned already in Q2, Q3 was a very tough comp for Korea, given that we did free delivery for everyone, regardless if you had a subscription or not. We had free subscription, effectively. That is also why we've seen a kind of return to growth as we have an easier comp as we come into Q4 and going forward, effectively. The fact that we're growing now is not that we're doing a lot of activities in Q4. It's actually all the work that we've done the last 12 months, building subscription program, innovating, UI improvements, logistic improvement, moving OD to roughly what you said, three quarters. I think your assumption is fair.
Thank you. Very helpful.
We also expect APAC to grow. APAC is doing very well, both on top and on bottom line.
Our next question comes from Silvia Cuneo with Deutsche Bank. Silvia, please unmute your line and ask your question.
Good afternoon, everyone. Can you hear me? Yes. Yes. Okay, great. I have a question on take rates. In MENA, GMV growth was largely in line with revenue growth in Q3. Given the planned investments to respond to competition that you mentioned, should we expect these to potentially lead to a lower take rate and therefore lower revenue growth and GMV potentially next year? Can you counterbalance these perhaps with an increase on delivery share, if possible? Similarly related to this, also in America, we noticed steady take rates. I just wanted to ask if this is instead driven by perhaps the multi-category effect to the mix. Thank you.
Yes. We do not go into details on how take rate evolves, and we are a little bit more focused on gross profit. I cannot fully answer that question. I think, as I said before, with Hungerstation and the playbook that we have, that is very strong. We lost some low-value customers that actually lost money per order, so negative unit economics to GMV. We used some of that money to bind strong, loyal, good customers closer to us. Net net, there was still maybe a slight take or gross profit reduction. Thanks to the growth, the effective EBITDA was flat, slight down, or slightly down. For talabat, I do not want to guide for them, but you should expect something similar. In the markets where they compete aggressively, there might be a slight gross profit reduction net net.
As they are growing, that also means that, yeah, the total impact on EBITDA, I think, is fair to assume that is, yeah, I think we should be happy if it is flat, maybe slightly upwards next year. I think that would be a great performance. When it comes to Americas, yeah, I do not know, stable take rate to stable gross profit is what, yeah, we are happy where we are right now. We do not intend to increase it next year.
Okay, thank you.
Thanks.
Our next question will come from Jürgen Kolb with Kepler Cheuvreux. Please unmute your line and ask your question.
Fantastic. Thank you very much. Two, one-and-a-half questions. First of all, you mentioned this moderate EBITDA growth for potentially 2026. At the same time, you are trying to improve the cash conversion.
I was just wondering if we have to maybe look at some of the building blocks that increase the cash conversion. Are we talking about maybe lower CapEx? Is anything there going on with any of the other drivers to the bottom line of the cash flow that we should be aware of? First one, and the half one is basically housekeeping. Can you remind us what kind of special cash out we had in Q4? To reconcile the cash performance from the EUR 2.8 billion, I think it was in H1, to the EUR 2.2 billion in or after nine months. Thank you very much.
Sure. To your question on cash generation and cash conversion, I think the biggest driver of our free cash flow is our EBITDA performance, right?
I think that remains very much the place where we draw improvements in free cash flow from, right, as a basis. Then looking at the individual lines that go from EBITDA to free cash flow, I would say there is nothing unusual there right now to be expected. I think as we talk about investments, in particular on the quick commerce side, obviously, you will continue to see CapEx there, right? We will continue to have CapEx investments for sure. We will continue to have lease expenses going through this as well as we invest in some of our quick commerce operations. I think across all the other lines, you might have a small increase on taxation as more markets reach profitability. I think on the financing, there is nothing there.
I think you basically mostly deduct the free cash flow improvement and conversion from the operational performance.
Maybe some of the entities that are also increasing is profitability. We'll also have some losses carried forward. Small item there. One thing that Marie-Anne didn't mention, but she and her team, including myself, are also pushing very hard on the working capital when it comes to retail. When it comes to Dmarts and so on, everyone who knows retail is that it's all about cash flow. Extending payment terms, improving your turnaround on your stock inventory, and not hold them in stock more than we have very strict guidelines how fast we should get it out of the stores. We continue to see increased working capital in our Dmarts in particular. That is also helpful.
Sorry, the Q3, any specific cash out to reconcile EUR 2.8 billion- EUR 2.2 billion? Was it the EUR 450 million?
The main cash outs you will have seen this year, a lot of them were obviously in Q1 related to buying back convertible bonds, to having the breakup fee from Taiwan coming in. What you see in the second half of the year and some of that in Q3 is obviously the payment of the atomium fine, which I think we talked about in August, and at that time it was literally imminent. That has happened for EUR 328 million. We also discussed at that point in time, Spain, where we were expecting about EUR 450 million of payments related to the rider model, right? That has come through to a large degree.
That's maybe a place where we expect a little bit more as well, maybe 100 more or so over the next months. That timing of that is a bit going into Q4 and beyond as well. Okay. That will be the main building blocks of the, let's say, the non-operating cash elements.
Got it. Fantastic. Thanks very much.
Our last question will come from Joe Barnet-Lamb with UBS. Joe, please unmute your line and ask your question.
Excellent. Thank you very much. I'm going to follow up on my own question earlier, if that's okay. Just thinking about how you're able to keep growing profitability despite substantial reinvestment. It draws me to a question on Korea. Your OD penetration in Korea has increased substantially, as Monique referenced. We know that OD margins in Korea have been below marketplace margins, but improving rapidly.
As such, Korean profitability has been working against both FX headwinds, but also blend headwinds throughout 2025. How materially different are Korean OD margins versus marketplace today? How materially have they improved? What does that all sort of mean for Korean profits into 2026? Thank you.
Thanks. There are many things at play there. When you look at the OD margin, it is lower than the MP margin still. It is improving, but it is still lower. In particular for subscribers, because it is free delivery versus for non-subscriber, it is a better economics for our OD than for our marketplace. Therefore, you have both the difference in marketplace OD, and you have the increase in subscription, and you have the increased margin that we continuously work on, that will continue to improve. All this together, it makes kind of the average gross profit.
We do not guide now overall on every single line items or anything, but as I mentioned before, taking it all in together, we want to prioritize growth. We are on a good journey. We know exactly what needs to happen and what we need to do and what are the growth levers. That is what we are prioritizing for next year. That means that, yeah, next year, it is fair to assume that it is flat development on EBITDA in local currency. I do not know what happens in local currency. That is another topic.
Excellent. Thank you very much.
Thank you very much.
This concludes the Q&A session. I will now hand back to Niklas Östberg for closing remarks.
Yeah, thank you very much, everyone, for listening in. And in particular, all heroes for your very hard work.
We are now having great momentum, and I'm very excited about the plans for 2026 and beyond. Thank you, everyone.
This concludes today's call. Thank you, everyone, for joining. You may now disconnect.