Delivery Hero SE (DHER) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by. Welcome, and thank you for joining Delivery Hero's Q1 2022 Trading Update Conference Call. [Operator Instructions] Now, I'd like to turn the conference over to Christoph Bast, Head of Investor Relations. Please go ahead.

Hello, and good afternoon, everyone. We hope you are well, and thank you for joining today's Q1 2022 trading update conference call. We trust you have all received the press release and the presentation, which we published this morning. As always, the documents are available on our website. We would like to remind you that this call is being webcast, and a replay of the audio webcast will be available later today on our website. With me today, we have Niklas Oestberg, CEO; and Emmanuel Thomassin, CFO of Delivery Hero, who will take us through the most relevant aspects of our Q1 performance and share further details on our outlook and the path to profitability. After that, we look forward to answering your questions. And now, let me hand over to you, Niklas.

Thank you, and hey, everyone, and thanks for dialing in to today's call. We've had a fantastic quarter ever since and looking forward to sharing this with you in this call. Before going into the numbers, let's reiterate some in-depth material around our cohorts and GMV targets. But starting with our vision, as you know, we are very focused on always delivering an amazing experience. We are obsessed by building the best service, and we believe we have a superior service in most of our markets. If we then go to the next slide, this becomes clear. Those who invest into us, they invest into the largest food platform in Asia ex China; the largest food platform in Middle East and Northern Africa; the largest food platform in Northern, Eastern and Southern Europe; the largest food platform in South America ex Brazil; as well as the largest quick commerce player globally. From that perspective, we believe we are the safest bet for those believing in food delivery and quick commerce. Going to the next slide, we like to reiterate the incredibly stable cohorts in over 50 markets with up to 20 years of history. This slide gives an aggregated view. In my opinion, this is the strongest argument for being an investor in the food delivery industry. We have updated numbers to now include also the Korea business. Starting with the left-hand side. Every year, we add another cohort and every cohort generates a higher GMV per year as time passes by. If you look at the 2017 cohort in light red, you can see that the GMV spend of that cohort in year 2 was already 1.7x the size of that year. In year 3, it was 2.1x the size of the year one, et cetera. What you can see is that the newly acquired cohort generates a higher GMV than the customers we acquired the year before. You can see that in 2020, we acquired an exceptional number of users, which also was partially driven by COVID. We are now back to a more normal environment and yet continue to experience general improvements in cohort frequency and size every year. In the upper right-hand side, you can also see how every new cohort is performing better than previous one. This is amazing and pretty unique for our industry. What I love the most, though, is that the incredible straight lines of the cohorts, even slightly upwards if you look pretty closely, I would say. This word about competition will get or should get comfort by looking into these charts carefully as lines wouldn't be this straight with people multi-homed or moved away every time they get a voucher. As you can see, the cohorts are incredibly stable. The stability of the cohort also means we have an enormous amount of long-term predictability. When you asked how we come to the EUR 200 billion to EUR 350 billion GMV in 2030, this is the underlying data. We know the development of the cohorts, and we can predict acquisition levels reasonably well. On the bottom right part, you can also see how cohorts improved in frequency. Here again, you see the uptick during COVID, but it's consistently and fundamentally upwards. Also, I would like to add that the average number of orders per active customers of 4.9 on a monthly basis is higher than any global peer that have seen reporting. Still, this level of orders per month is very low and especially as we're launching new verticals and occasions. Next slide gives you a better understanding of the constant improvement of order behavior of our customers. Existing cohorts have stronger loyalty and order more frequently over time. If you look at our 2016 cohort, they were ordering 2.4x per month in the first year. In year 6, this number has already increased to 6.3x per month. So, a clear improvement over the years. What we also experienced is that the new cohorts especially exhibit a higher order frequency than previous cohort. This becomes visible if you look at the columns of the table. The 2016 cohort order 2.4x per month during the first year, whereas in 2021 cohorts ordering 3.1x a month. Based on this table, you can see how the business has a strong underlying growth ahead of itself as the cohorts matures. As we said, our 2030 GMV target of EUR 200 billion to EUR 350 billion is based on our cohorts. However, the graph below that we show here is just a sanity check that there is enough potential. In order to reach EUR 200 billion GMV, we would need to reach approximately 0.54 monthly orders per capita next 8 years. Today, already 7 markets have achieved this level. Top 7 markets had an average of 1.4 monthly orders per capita last year. The top 7 countries include markets from Asia, Middle East and Europe. I should add that none of these countries were at 0.54 only a couple of years ago, so they also keep developing. Before getting started with the financials of Q1, I'd also like to give everyone a view on the current COVID impact on the next slide. Here, you can see our estimated quarterly impact from COVID pandemic on our GMV growth. In 2020, we had a circa 5% tailwind on an annualized basis as per our own estimates. Some countries were positively impacted while others were shut down. In 2021, the year-on-year tailwind continued and added another circa 6% growth to our annualized year-on-year, in particular than Middle East as they were locked down in 2020. In particular, positive, of course, as mentioned is Middle East, as well know complete lockdowns timing as in 2020. In 2022, we expect very limited COVID tailwind, meaning -- and I hope so as well, I would say. But that also means that we lapsed 2021 positive effects causing a circa 9% headwind. As you all remember, Q2 and Q3 were particularly strong for us in 2021. On the right-hand side, you will find an estimate of normalized growth rates. The extraordinary high growth rates in 2020 and '21 are not unrelated to COVID, but can be explained by our hypergrowth strategy, including fast rollout of own delivery and significant push of affordability in Southeast Asia. During this year, we believe our COVID adjusted GMV will be more moderate, so between 33% and 36% in 2022, which is a reflection of the much larger size we have today versus the last few years. In absolute terms, though, we keep growing at record rates. Now let's move to Q1 key highlights. First part is on the GMV. We had a very strong GMV and revenue development in all 5 segments. We definitely gained market share in all these places. We achieved this while generating record high contribution margins across the group, both before and after vouchers. We also made a strong top-line growth while continuing to work on various profitability levers, minimum order value, dynamic pricing have been introduced in 90% of our markets. We have optimized our partial targeting and made further progress on upselling. We have successfully also tested services, which we rolled out in selected countries in coming months. I'm also very pleased to share that we reached breakeven on adjusted EBITDA level in our Asia segment in March. While this is before allocation of group cost, I still believe this is a clear milestone given the heavy losses in the past. This happened ahead of plan, and we are now in a much stronger position to build on our leadership position in the region. On top of that, we have introduced new pricing scheme for own delivery in South Korea with a clear positive impact on unit economics ahead. This was done, as you say, end of March, I think, 21st or 22nd of March for part of Seoul and actually in April 4 for the remaining part of the business. So this was not the reason for moving to EBITDA in the Asian segment, just to make that clear. But this will obviously create additional tailwind going forward. And we have also seen promising first results from our subscription offering pandapro in the APAC region. I'll speak more about that and why that is a good tailwind book or growth as well as profitability. Last but not least, we have strengthened our balance sheet, positioned by securing additional financing totaling EUR 1.4 billion, following the placement of a EUR 1 billion term loan. Our cash position stood at EUR 3.5 billion at the end of 2021 on a pro forma basis. In addition, we arranged a RCF of EUR 375 million, which adds even further flexibility. Moving on, and I should say, we feel very confident with that level as we're now moving forward with increased improvement in cash flow month after month, quarter after quarter. And we, therefore, think that we have a strong position with that cash balance. We are no longer reporting orders as this has become increasingly contradictory to our business objectives of increasing GMV and profitability. Historically, we focus a lot on driving awareness and usage. We have exceptional reach and it's more about driving profitability or profitable growth. Our focus is, therefore, to turn loss-making orders into value orders or cut them altogether. We have made the switch internally already 3 quarters ago. And we still want to make ensure that for transparency, we have provided information about basket size growth in markets where there has been a change. From there, anyone can back calculate our order growth. It's just that we don't want to measure ourselves against a KPI that goes against our focus to drive profitability. In terms of GMV, we had very strong GMV development in Q1. We believe we had around 3% positive tailwind from COVID, but 2021 had an even higher COVID tailwind. So overall, a very strong quarter. But in general, we will have to push much harder to -- or we have to keep pushing hard, as you said, to hit our growth targets, while EBITDA should be significantly easier. Revenue did even better in Q1. We totaled more than EUR 2 billion in the quarter, resulting in revenue growth of 52%. Now let me hand over to Emmanuel for a deeper dive into our Q1 financials. Emmanuel?

Yes. Thank you, Niklas, and good afternoon, everyone. Let me start with our performance of each of our business segments. Our platform business in Asia performed extremely well in Q1, reaching GMV of nearly EUR 7 billion and segment revenue of EUR 928 million, which corresponds to a growth rate of 35% and 50% year-on-year respectively. Part can be explained by the rollout of the basket size and delivery initiatives, which push our AOV by more than 10% compared to -- in Q1 2022. South Korea has made strong growth progress, while we also successfully faced other promotion campaign by late March in Seoul Metro and early April, nationwide for bringing own delivery service. Just as a reminder, as we offer promotional pricing on recently, our unique economics for own-delivery were negative. This should now [ improve quickly ]. In Taiwan and Hong Kong, we successfully introduced our subscription service pandapro, which results in higher order frequency, larger basket size, but also increased their customer loyalty and automatically strengthened leadership position. Last but not least, we have also reorganized our operations in Thailand and achieved gross profit after vouchers in January. In Q2, we will launch pandapro and we will participate in the government payment scheme. On Slide 14, you will see the strong growth in Q1 in the MENA region. GMV reached EUR 1.9 billion, a plus of 32% year-on-year at constant currency. Segment revenue came in just short of EUR 500 million, corresponding to our 50% increase year-on-year at constant currency also. Within this segment, Talabat benefits from healthy customer behavior and generates 37% GMV growth on a constant currency basis in Q1. And furthermore, the advertising revenues at Talabat already reached 2.3% of GMV. Hungerstation achieved a high penetration of vendor funded deals among their key accounts, which strength the leadership position in Saudi Arabia. The situation in Turkey remains challenging with additional headwinds from bad weather like snowstorm and weaker macro environment in Q1, but also fierce competition. Over the coming months, we will migrate Yemeksepeti to our Pandora tech platform, and this should improve the customer experience and also enable us to gain more traction as the new tech platform is significantly stronger than the current one. This means the user interface will be more customer friendly, and we will be able to offer a new service like pickup, dining, [ all range of ] programs and become more targeted in terms of marketing. Now turning to Europe on Slide 15. GMV in Q1 totaled EUR 707 million, only a modest 4 years growth year-on-year. However, this is significantly influenced by the divestment in the Balkans. On a like-for-like basis, excluding the Balkans and Germany, GMV grew by 17% year-on-year despite lifting of COVID restrictions. Advertising revenue, or NCR, grew by 52% year-on-year and now stand at 2.1% of GMV compared to 1.6% in Q1 2021. Now, I would like to briefly comment on operations in the Americas on Slide 16. Here you can see that the business had another successful quarter, growing GMV by 33% year-on-year to EUR 558 million, while segment revenue grew by 39% year-on-year to EUR 149 million. We continue to optimize our unique economics in the region, which includes minimum order value, cross-selling and delivery phase. And this [ propelled their ] average order value by 19% year-on-year to EUR 11.1 and drove gross profit per order to record high. Argentina, one of the largest country in the region, even reached breakeven on adjusted EBITDA level before central across the location. And we will continue to push for profitability and plan to roll out their services in Chile and Argentina in the coming months. On Slide 17, you will find the steady progress of the year, Integrated Verticals segment. Both our GMV and segment revenue more than doubled in size year-on-year, reaching EUR 410 million and EUR 374 million, respectively. As we constantly adjust and optimize our product assortment, the basket size soared by more than 22% in Q1 to EUR 13.7. Going forward, we will continue to invest in new Dmart opening, although at a smaller scale than in the past years. In Q1 2022, we launched 48 new stores, which is significantly less than the 213 stores opened during Q4 2021. The last 3 quarters have been the peak of our investments in terms of openings, impact both CapEx and OpEx. And this will clear strategy to be [ covered ] and now drive our GMV based store to drive margin improvement and allow the business as a whole to be more profitable. On Slide 18, in addition to that, what I just mentioned, we also work on the profitability to our own delivery business. You can see here that the contribution margin before vouchers remain above 6%. In MENA, the contribution margin has improved slightly, while Europe has turned positive again since the closure of Germany. Now, all 4 regional segments are showing a positive margin again, and we expect further improvement throughout the rest of the year. As you can see on Slide 19, the fully loaded contribution margin after vouchers has reached a new record high and are still positive for the whole 4 segments. We explained in the past that we constantly work on reducing our voucher intensity. And as such, we are happy to announce that vouchers as a percentage of GMV has declined by another 10 bps quarter-on-quarter to 3.2% in Q1 2022. For the remainder of the year, we expect the level of vouchering to decline even further. In addition, I would like to take this opportunity to refer to our advertising revenue of the so-called noncommission-based revenue or NCR. In Q1 2022, advertising revenue stood at 2% of GMV, up from 1.9% in Q4 2021 and 1.6% in Q1 2021. And all these numbers exclude Woowa still. These high margin revenues are excluded from the contribution margin, as you can see on this slide. So that's a very important element. Now on Slide 20, we provide an overview of our large investment portfolio in the global food delivery space. As we have discussed, there is strategic rationale quite extensively in the past, let's have a quick look at the valuation. So you can see that the value of the portfolio has slightly declined from EUR 2.1 billion to EUR 2 billion since last quarter. This is primarily due to the recent stock market tumult and the impact this had on the valuation of our public assets. This investment portfolio, we serve as an additional source of liquidity if and when desired. For instance, we have recently divested a partial stake in Rappi for worth $250 million, while continuing to hold a stake of around 5%. And furthermore, the portfolio consists of liquid assets worth around EUR 550 million. On Slide 21, we present an overview of our pro forma cash position following the announcement of the EUR 1.4 billion debt financing, including our revolver credit facility. At the end of 2021, we had cash and cash equivalents of EUR 2.5 billion, and having the proceeds from the term loan we issued this month, our pro forma cash stood at EUR 3.5 billion at the end of last year. The term loan offering was met with strong demand in the U.S. tech market and significantly diverse our funding source by setting a new source of long-term capital. Furthermore, we have agreed our RCF or revolver credit facilities of EUR 375 million with a group of our banks, which provide even more flexibility. Against the bank loan of more than EUR 3 billion in cash, a lot of investment portfolio that we can turn into additional cash if needed, and our plan to reach adjusted EBITDA breakeven in 2023 on group level. We view our cash needs as -- covered an overall planning horizon. And now let me hand back to Niklas. Niklas?

Thank you, Emmanuel. We already shared some information around our Dmart business a couple of weeks ago when we issued our term loan. However, as some of you may not have had opportunity to participate in these calls, we wanted to briefly touch on this topic again. We regard quick commerce as a massive long-term growth opportunity, which is highly complementary and synergistic to our platform business as outlined on Slide 24, I believe. First of all, the continued development of Dmart will drive even more new customers towards all our business verticals with upselling opportunities between them. This will then drive order frequency and customer engagement. In addition, we can already see today the positive network effects, as soon as existing platform customers try our Dmart service to start to order food more frequently. As food and groceries will be delivered by the same driver, we can also leverage our existing food delivery fleet, which reduces the average delivery cost. With a higher network density, we can also reduce the delivery time. On the next slide, I'll take you through the most relevant unit economics of the business. This slide shows what happens to economics when we increase orders per store. In those countries where we have reached over 500 daily orders per store, we start improving unit economics. Looking at the left-hand side, you can see that we then do less free delivery in markets where we have reached target order levels. We have also slightly longer delivery times. In a few comparative markets we operate, we obviously deliver much faster than what you see here on this slide. But you can see that we can adopt here. You can also see that we can offer a larger SKUs set as the store turnover is faster. Now more importantly, looking at the right-hand side, the product margin within -- in this case, the best-in-class countries and Dmart's overall is about the same. Both are still between 5% and 10% below long-term targets. You can see that we changed or charge slightly higher delivery fees, as we optimize a bit less for growth, and that reached our target order level. You can also see that delivery cost are 4.6 percentage points lower as we stack more. This also means a slightly lower delivery times as mentioned before. You then have picker cost, which are 4 percentage points lower as we have better utilization of store personnel. Then we also have lower other cost like store manager. There's also less waste as turnaround times and SKU predictability is higher. When you add all this up, we achieved 8.9% positive gross profit before vouchers versus negative 6.3% in the other markets. This is again why we need to invest to drive orders per store. This is also apparently looking at voucher share, which is 2.3% versus 5.7%, resulting in an even larger bottom line difference. During 2022, we plan to drive more countries towards the 500 orders per store. And in our best stores, we are already at 1,000 daily orders per store. So 500 is by no means the maximum, but it is the level at which we make good profitability. Now let's have a closer look at the margin development of the best-in-class Dmarts in the next slide. The business in these countries is on the verge of breakeven. This was achieved through an increase on the average basket size of 21% year-on-year, and GMV per store of 33%. And I spoke about orders per store, but what is truly relevant is, of course, a GMV per store or both that they have a utilization of staff to drive even more importantly that will drive GMV. So with an increase of 33% year-on-year, you can see how we have improved our unit economic share. The main -- the driver a larger product assortment, which possibly impact order frequency and the basket size. In turn, higher order volumes result in better purchasing condition and higher product margins. Better unit economics and scale then led to significant improvements in adjusted EBITDA before vouchers as mentioned. One important segment here, the best-performing country within this bucket already generates an adjusted EBITDA margin before vouchers of more than 6%. So clearly, within the 5% to 8% that we have in our food delivery business. And as mentioned here the Dmart already one country that we got in there, much faster than any food delivery country got there. Almost also partially driven by the -- or significantly driven by the scale that we already have. Coming on to the advertisement side. I'd like to highlight the advertisement revenue or what we call NCR, noncommission revenue, but advertise is a bit more accurate. As you already know, we have started to monetize our platform's audience through different ad products, which enable our restaurant partners to improve their visibility to some of the following products: First, cost-per-click, or CPC, these are various premium listing options to increase restaurant visibility on the platform with an automatic renewal of monthly ad booking. The vendor only pays if the customer clicks on the ad. The second project is Joker, pop-up banner or visibility banner with a discount offers displayed to customers. Here, the restaurant only pays when the customer places an order. This tool is highly focused towards new customer acquisition. And also the value of this is both for consumer, they get a good deal for the restaurant to get a new customer and for us to make more -- significantly more margin. Then we have various other products like offering restaurants, the opportunity to highlight their certain dishes and ad product features. So all in all, we believe there is really a huge earnings opportunity ahead of us outlined on the next slide. And what you can see here is that since 2019, advertisement revenues have more than tripled to EUR 288 million in 2021. However, last year, ad revenue stood at only 1.7% of GMV and this did not even include Korea in -- in Q1 2022, advertisement generated 2.0% of GMV. We plan to reach 2.5% of GMV, including South Korea by 2024, 2025. This would result in over EUR 2 billion of almost 100% margin revenue. On the long -- or in the long-term, we expect advertising revenue to represent between 3% to 5% of GMV. Now on to the next slide, where I'll share more details on subscription service, pandapro. We launched our subscription service pandapro in early 2021 in the APAC region. Since then we have managed to acquire already more than 1 million subscribers in over 10 countries. Pandapro customers pay a monthly subscription fee and in return benefits from selected free delivery, discounts and other attractive deals, both in food delivery and quick commerce. This customer exhibit a significantly higher order frequency and have higher baskets. Customer of pandapro become more sticky as a large portion of the subscribers order food only through us. This makes subscription a great customer retention tool. During the recent months, we managed to convert more and more subscribers from monthly to half year or even yearly subscription as a service offers great value to customers. Subscription services are being rolled out globally. On the next slide, we illustrate the impact from pandapro and customer behavior. And I must say it has taken us a long time to optimize our subscription program, but we are now super happy with it. The challenge with it is to make sure it's attractive enough for users, are still generating similar economics on GMV basis and not only per customer. We now believe we have a good program in place. And as you can see here, with that program, monthly order frequency of the customers increases by 59% after signed the pandapro subscription. In addition, they tend to order larger baskets, which results in GMV boost per customer of 81%. So these customers add a lot of value to our platform. On the right-hand side of the chart, you can see how fast we gain traction with this service. At the beginning of the year, we started at close to 0 and within a short period of time, we achieved more than 1 million subscribers only in Asia, and this excludes Korea. Going forward, we might also introduce subscriptions in other countries or actually, we are introducing subscription. In lowest market, there is some sort, but I think we are most happy with this program, and it's a -- I mean that we're implementing. Now turning back to Emmanuel for the outlook section on -- yes, Emmanuel.

Thanks again, Niklas. So now coming to our outlook. Q1 performance on GMV was, as you've seen, very strong. And we also end up ahead of our adjusted EBITDA compared to our budget. However, it's too early to change any guidance, and we maintained our previous guidance for 2022. So as I said, the business has done better than planned on EBITDA, so that we are more confident than ever that we will be breakeven for the platform business, including central cost allocation for the full year 2022. GMV is expecting to reach between EUR 44 billion to EUR 45 billion, and revenues between EUR 9.5 billion to EUR 10.5 billion. Furthermore, we expect our platform business to generate positive adjusted EBITDA on a full year basis in 2022. Having investments up to EUR 525 million in integrated verticals, the adjusted EBITDA margin on group level is expected to range between minus 1% to minus 1.2%. At this point, I think I would like to reemphasize once again that adjusted EBITDA is not a net factor for Delivery Hero, but rather an input factor. We are the ones that who decide on the targeted growth and associated investments. Therefore, we feel a very high level of confidence with this guidance metric. I would also like to remind you that in 2022, the vast majority of the negative adjusted EBITDA at Delivery Hero will come from less than 10% of the GMV. Let me also share some thoughts with you on integrated verticals. We understand that the Dmart unit economics very well and have a concrete plan for the business. So that means that we have visibility and control on the EBITDA burn here. And we're already profitable in some countries, so that we know how to get there. Given the market environment and as we have seen less competition that we're expecting in quick commerce and our Dmart are improving the unique economies rapidly. We may end up spending less than the up to EUR 525 million guide previously. But we will revert on this when we have more visibility. When it comes to global, the company confirmed that the business has continued to perform very well, despite significant GMV loss in Ukraine. GMV growth in Q1 was above budget and over 85% year-on-year. So consequently, the market share gains have been dramatic as well. And despite challenges in Ukraine, the team remains confident to reach GMV if the full EBITDA guidance is utilized. However, Dmart also choose to invest slightly less than previous guided, if markets remain constructive. For the first quarter, they performed slightly ahead of EBITDA budget as well. A further encouraging remark on EBITDA is that Spain has turned profitable before group costs, and it shows that the playbooks works. Finally, as a reminder, the transaction is not closed at this stage and we expect the closing to take place early Q3 2022. Now on Slide 25 -- 35, sorry. We also confirm our guidance of reaching adjusted EBITDA of between EUR 0 million and EUR 100 million on the platform business, including global in Q4 2022 after group costs. As well as a positive adjusted EBITDA for the entire group, including global in 2023. And our confidence in this is very high. And our group profitability might even come a few months earlier, but we don't want to guide towards this right now. So now on the next slide, and for those that who are wondering how the profitability will continue to develop after 2023. I will ask you to take a look at the -- on the Slide 36. Here we explain how we can achieve our long-term adjusted EBITDA margin target to offer 5% to 8% to GMV. As outlined in our previous calls, we have many levers to reach this target. Some of these are related to revenue, such as increasing delivery fees, minimum order value, introducing service fees and dynamic pricing, organic increase in commission rate, et cetera, et cetera. There's a lot of room to reduce cost per order by better digitalizing rider place for Dmarts logistic as a service or by slightly increasing stacking or negotiating lower payment fees with our payment providers and better purchasing conditions with our supplier. With the combination of increasing revenues and improved cost structure is planned to drive gross profit from the 5.1% achieved in 2021 to a range of 11% to 13% on the long-term. And here, I would like to mention that with our Woowa, the gross margin would already be above 7%. When it comes to marketing, we obviously have a lot of room to flex up and down, the latter the platform gets, the lower the customer acquisition cost. And given the size of the opportunity, we have remained on the marketing cost at 3% of GMV, while our largest markets are already below 1.5% of GMV. The OpEx include restaurant sales, which also drops as platform gets larger, and we see more self-signups. Also, a lot of CapEx will reduce as we drive operating leverage at scale. However, we kept overall OpEx at [ 3% ] as we expect technology and products will still remain a bit fortunate. But also here, our best-in-class markets are already below 1.5%. And this combination of all these efforts will bring our long-term adjusted EBITDA margin to GMV to the targeted 5% to 8% range. So we hope that you find this deep that helpful, and we are now looking forward to answering your question. Christoph?

Great. So a quick remark from my side here before we start with the Q&A. As we would like to give every analyst the opportunity to ask a question, I would kindly ask you to limit your questions to one only. Now let's kick it off for the Q&A. Operator, Please go ahead.

Thank you. [Operator Instructions] First question is from Giles Thorne from Jefferies.

My question is on Korea. And it's probably one you could anticipate, namely, I'd be interested to get some color on churn dynamics in both consumers and restaurants, as the promotional pricing has ended. I appreciate something you've touched on today and then the update on 4 of April, given its importance to the overall full year outturn, it would be useful to get the latest color. And then if you could weave into your answer, what your own delivery unit economics are looking like today? That would be helpful. Thank you.

Yes, so Korea, there's no churn in users because we haven't increased any fees for them really. And also for the restaurant side, we haven't seen any -- very minimal, I would say, churn there. So not more than we've seen on an average month. So therefore, has been very good results. We also didn't expect churn. So -- but of course, it's much nicer to see that being the case. Overall, we believe that we have slightly increased our market shares lately. But yes, we remain significantly larger in number 2, number 3. In terms of the unit economics, we don't disclose exactly, but I think I've stated before, that we increased the unit economics of delivery with roughly between 1.2% to 1.7% or 1.8%, if I'm not right per order. And overall, the unit economics is now slightly higher, if I'm not mistaken, then for the Marketplace business. Marketplace business is still the majority part of our business. So yes, that's still helpful.

Okay. So some of the inevitable negatives press we've seen coming out of Korea from your favorite English language newspaper, that [indiscernible] delivery, that's...

No. We haven't seen, I think Alaska team, and I was meeting with the team yesterday -- week or last week. I think they referred to exactly the same churn as we have in the restaurant industry of roughly 3% per month, but that is just because we have restaurants churn, because they turn the business or change the ownership are the reason, and there was no increase rates -- material increase. So we have not seen that. And press, of course, if you have a few 100,000 restaurants on the platform. And of course, it increased price, and if 1% of those restaurants are happy, then you're speaking about 3 or a couple of 1,000 restaurants. And surely, one or 2 of them would come through media. But overall, we think it's been, very well perceived. We have also done it very carefully. We have a very strong connection to the restaurant association. They are very happy with the change that we made. Everyone understands that we cannot operate on promotional pricing forever. So we have seen absolutely no negative effect on this. If anything, should we -- as mentioned before, we had roughly 3% COVID with positive impact, I would say, this year. Last year was a little bit more positive. So net effect is still negative impact from COVID, but there was a slight positive. And I would say Korea was one market where there was a slight positive impact in Q1. So -- that I would assume will go away, but not because of orders or pricing.

Next question is from the line of Rob Joyce from Goldman Sachs.

Thanks for that incremental detail in the release. Just one for me, sort of maybe sort of higher level one. Given the increased focus of your strategy and particularly the guidance on profitability now. Did you or would you consider extending the KPIs in the long-term incentive plan to operating metrics beyond revenue growth?

Look, I think in -- yes, so in general, we have not addressed to change our plan. And our business has come to a site where profitability makes a lot of sense, and driving additional levers make a lot of sense. So this was already our plan back, I don't know, a few quarters ago. So we have not reacted to the market environment. Of course, it's helpful, and we might guiding a little bit more towards that. But the plan was already to get to that profitability during this year, full year for the platform, including the Integrated Verticals. So I don't think we have changed. I think, overall, the incentive structure that we have is very much in line with our shareholders that we want to drive shareholder on a share price in the long-term. And I think we're doing that. I'm generally not opposed to being creative, but incentive structure, but I also find it a little bit careful, because it's very easy for us to produce EBITDA. And if you want to produce EUR 1 billion or EUR 2 billion EBITDA in 2022, I think we could do that. But it wouldn't be the right for the shareholders. If we would pull all the levers we can increase a little bit commission, drive EUR 0.50 more in pricing. So delivering 5 minutes or 10 minutes lower, that would save us another few EUR 100 million. So of course, we could easily drive that profitability. And I would find it a little bit dangerous if management teams are incentivized to drive profitability, because that can easily be received while driving long-term share value, it's harder and requires a little bit more diligence for the long-term. So therefore, I think the current setup is probably the right one. Having said, I would never think about -- I'm so in the boat of having shares in this company. And as you all know, I bought a significant amount of stock also during the downturn now. So I think we're all in the same boat with or without EBITDA incentive plan.

Next question is from the line of Joseph Barnet-Lamb from Credit Suisse.

Excellent. There are lots of drivers of order headwinds within the industry. One is proactive focused on economics. And Niklas, you alluded to a potential willingness to walk away from orders that were loss making. How much of your 2021 order base could conceivably fall into that bucket? Just some discussion or any color you can give us on the impact on group orders from proactive steps to the varying profitability?

So I think -- and it makes very little sense for the business to drive non-profitable orders. And yes, of course, sometimes you may accept one customer doing one order that you don't make money on, and we will continue to do that. But if you see customers continuously ordering with vouchers or with incentives or very small baskets and long distances or any other way that makes that order continuously loss-making or that customer loss making. Then for us, I think the only way to -- that makes sense is that you have to find a way to get that customer to either order with a high delivery fee or no voucher or lower delivery distances or you actually prefer that customers not to order at all. And that's a little bit more of a trade-off. I think in general, that's not material part of business, because we have also not done that in the past that we have strived bad orders, and that all going to be silly. But I would say there is probably a possibility that you're increasing GMV with, let's say, 5% because you -- some of them we push up to become value customers, and therefore, you push GMV a little bit. And you might lose -- so you might lose like 10% to 5% of orders, you might boost GMV with 5%, and maybe even 5% and 10% or so. So that means effectively, you might be in the order of minute 15% or so, as you're making that transition. If you look at this time, and there is by no means, I'm happy to be very transparent on that, it's just that we don't think it makes sense to measure ourselves in orders. But if you look now, we guided or we disclosed the increase in AOV for the segments where there was a change. So if you look at Asia, GMV has been increased roughly 10%, in integrated vertical it's more than 20%. I think the same was with Latin America, around 20%. So -- and so you can assume that -- or making the back of the envelope, basket size will have increased with 10% to 15%. So looking at growth, you would then also expect order growth to be substantively or consequently 10% to 15% lower. So I don't think that deviation will be much larger than that. And I think, yes, it's probably a little bit more of a transition. But in order of magnitude, maybe 5%-plus and GMV 10%-plus 10%-minus in orders.

Next question is from the line of Andrew Ross of Barclays.

Mine is a shorter-term question, just looking at Slide 8, and the helpful headwind you've provided from the pandemic for the year. It looks like my headwind is more in Q2 than Q1. So can we just get a sense of how much you're growing in April and kind of roughly what we should expect in Q2? I'm thinking of more low-20s, GMV growth looks like a sensible assumption based on my chart, but tell if I'm stupid.

Yes. So as mentioned before -- and I tried to be helpful and the reason why we put this graph is to make sure that we don't set a guidance or that people miss -- you said miss expectations for the next quarter or in Q3, Q4. So I'll try to add for disclosure there. So we grew 30% or 31%, I think, this quarter. I think there was probably a 3% -- and it was a 3% impact -- positive impact in COVID. As I said before, there was an even more negative impact, of course, from -- or positive impact in Q1 last year. So therefore, the net effect is that we probably had a 5% headwind. In the next quarter, we have roughly -- I'm judging here by the chart, so please forgive me if there is 1% or 2% wrong, but there is roughly 10%, 11%, 12%, maybe 12% impact negatively year-on-year effect on Q2. So that would expect that you should have a -- yes, an order of magnitude, it should be maybe 20% in that. So if you have 24% to 27% this year, Q1 was, of course, better, Q4 will be hopefully better and others will be slightly lower than this 24% to 27%. I hope this is a confusion more than it helped. But -- yes.

Definitely helped. Can I just confirm that you're actually seeing that in April? Or is that something that you may see in May and June?

I think it's a little -- April is a trickier month, because we have Ramadan starting 10 days earlier every year. So this month, it was completely in April. So -- but order growth was -- maybe slightly higher than what you said there. But yes, I think we can expect in May and June that there will be in that order of magnitude.

Next question is from the line of Monique Pollard from Citi.

So one for me, it's on the Dmarts. So really useful to get that EBITDA margin improvement over time for the best-in-class Dmart. Just wanted to understand now that you have that formula that you know you can deliver the breakeven profitability. Do you think the path to getting there will be successively quicker for the other stores? And I'm thinking particularly about that, given the product margin that you've highlighted for your best-in-class stores is actually slightly lower than the overall product margin that you have and that seems to be sort of sequentially improving.

Yes. I think in general, we have been very happy over the last couple of months with our development as Emmanuel also mentioned in his review. So I think overall, you probably have tracked a little bit faster than expected on our margins. And we don't want to change guidance here, we will do that when we feel that we have more visibility. But overall, we feel very positively about it. There are a number of levers that we can pull. And as I said, product margin is, of course, one that is still significantly lower when, what it should be long-term. We could, of course, increase the margin easily by just increasing pricing, but that's not what we want to do. We increase margin by actually pushing procurement and purchasing power. And that's the main driver for this 5% to 10%. And that, of course, takes a little bit longer time. The reason why I think we're also performing a little bit better than plan on margins and EBITDA is that the growth has been a little bit higher than what we anticipated. And as I said before, we're pushing hard to get to that, let's say, 500 orders per store, because then we can -- when we get there, we can be a little bit less voucher, a little bit less -- a little bit better on the pricing and so on, a little bit more stacking, et cetera. So as order levels were a little bit higher than planned, also means that we moved a little bit faster on orders per store or even more important is to gain deeper store. And that was also another -- that basket increased a little bit faster than what we anticipated as well. So the combination there, but the growth in order per store as well as the GMV has been positively helping our business. But it's a little bit too early to give a new guidance or rough guidance on this. So I don't want to say that it comes faster, quicker. I can say, though, that if -- as we -- the investments that we did last year was a lot about building new stores, building new coverage areas. The investments this year is really to drive orders per store. And the more orders we drive to 500 or more orders, the more orders will be in the profitable segment. So that should mean that the profitability will improve every quarter. And the cash flow will improve fast every quarter, because we get better procurement capabilities and purchasing powers and being able to move a positive working capital to a negative working capital. So EBITDA should improve every quarter, every month, every quarter from the Q3, Q4, Q1 level, so we were roughly at the same levels there. And that should improve quarter-on-quarter and cash flow improving even faster.

Sorry, just a follow-up, on the product margin, the improvement in that product margin, is that all driven by better procurement? Or is it some of it mix, like you pushing more SKUs that have a better product margin with suppliers?

Yes. So we do both. And that's why it takes us a little bit time, because you changed your SKUs that -- it takes some time to get the right SKUs, know what customers want, what are the items, and this is not on a global basis, but sometimes it can even be on a local basis, some product, of course, global, but many of them local favorites. And the more we know the right SKU set, the more we start building ability to improve some margins on those product items. And we can start also testing a little bit on price sensitivity on different products. So milk might have a higher price sensitivity than a chocolate bar, as an example. And the more with data we have, the better we can actually assess that price sensitivity by SKU. And of course, if we have 3,000 to 6,000 SKUs, that also takes a little bit of time to optimize to perfection. So I think a significant improvement there, a gradual improvement there. The other one is procurement, because the more orders per store we have and the better procurement capabilities we have, because for someone to supply our store on a daily basis, it is not very attractive for a supplier if you don't have enough orders or if you just have few items. So the more items we sell on the store, the more possibility we have to improve that. Additionally, we are building distribution centers, to further improve our procurement capabilities. So it becomes cheaper for our suppliers to distribute to one place and then we distribute our stores. And that way, we can significantly improve our procurement capabilities to be more in line with a retailer. And I would say, currently, we are -- our procurement capability is still somewhere between 6% and 9% of the same, worse than a wholesaler or a retailer, a strong retailer and that shows us that we still have more than 5% margin to improve just on procurement side. Yes, that has been a little bit of both.

Next question is from the line of Marcus Diebel from JPMorgan.

If we have only one question -- only one question it is. It's basically in a nutshell, what you were saying, you were increasing service fees in some of your markets. Overall, maybe with the section of Turkey, of course, would you say in your #1 market's market shares are at least stable or even marginally increasing. Is that really a right bottom line of previous comments?

Yes. I would say that there are 3 markets out of 72, it includes global. There are 3 markets where we have lost market -- sorry, 5, probably 5 markets out of 72 that we have lost market share. And out of the remaining market, I would say that we have gained market shares in most of them, and some of them have been stable market share. But we have -- Turkey is one market where we have lost market share. I think people think that we lost market share in Saudi Arabia. I don't think that is the case, if you look at the last 12 months, I even believe that it marginally increased market shares over the last 12 months, but then there are a couple of other markets. I don't want to go into detail, but a couple of smaller markets where we have lost a little bit of market share. But the majority have been increasing and some of them have been stable, I would say.

Yes. And just as a follow-up, you feel that obviously, you measure, again, increasing service fees is something you can do kind of without expecting that the competitors in those markets have to follow, you feel like -- through without losing market share.

Yes. I think in general, the market has become very rational. Everyone wants to increase service, everyone wants delivery fee, everyone wants to save a little bit, I don't know, this goes for everyone, this is true for everyone, this is true for couponing, and that's true for grab, that's true for - Just Eat, that's true for whatever the competitor have Uber, we compete in 2 or 3 markets as well. That's true for everyone. So I think in some places, they have been fast, in most markets, they have been faster implementing service fee. In some markets, we might be ahead, but I would expect that the competitor follow through. If they don't, then we might port as other markets for it. So we will be rational. But of course, we will also monitor what the competitor does. And if you feel like they try to take advantage of it, and we will adopt and adjust and move it in other markets. And there are enough markets for us where we can do it in. We don't plan or budget on doing this in all markets. And the fact is that we don't even plan to do it or trying it -- for it in any markets. Anything we do is actually incremental to our EBITDA. So we have no must of implementing service fees, and we will do it where we feel that competition is constructed.

Next question comes from the line of Miriam Josiah from Morgan Stanley.

It's just on the own delivery contribution margin. If I look at the chart, it seems to be sort of broadly stable, perhaps at Europe. So how should we think about the improvement going forward? Is most of that going to come from service fees? Or what are the sort of key levers there for the rest of the year?

I think there are many lever, and service fees is one, but we also long-distance delivery fees. And we have on the voucher side being more targeted, reducing, getting better, the AI part of vouchering making sure that there is no one that can order multiple times with it or any fraud protection you have there or anything else that would be also improved margin. There is definitely a possibility for further stacking. We continuously do that, to improve the cost side of things. You may have a little bit of commission work because we sign up more restaurants. And as we sign up more restaurant there going on slightly higher commissions. So there is a gradual continuous improvement. There was for a while a slight reduction, because we added some key accounts. But now those key accounts are on board and we had more independent restaurants and therefore, the average will gradually move up, especially since it is shown in the restaurant industry. So there are so many levers, I could continue probably for another few minutes. I wouldn't say that service fees is not the reason for what has gone up in the past, and it will not be the reason for going forward. But of course, it's very helpful when it does or when we do, additional bonus, putting this way.

And just as a follow-up, what the percentage of orders that you vouch today? Or where do you think that can get to you?

I don't have the latest numbers. I don't want to say something wrong here. I have a number from a few quarters ago. I apologies for this, but it's still low. And we deliver on average something like 23 or 24 minutes. I believe there's a little bit different market mix here, but in order of magnitudes, we deliver very fast, faster than I think anyone else does. We still manage to do some proportion of stacking. I would say it's low 2-digits. Where can it go long-term? Well, the more density we build and the more we grow, and you can probably get to a hopefully, maybe 50-plus, if you look at later on and others, they stacked probably more than 50%. And the stacking is not one order, it might even be 2 or 3 extra orders. So I don't have to say anything here, but I think it's quite substantial amount of stacking that can be done. If you use the late update, a couple of minutes lower on delivery.

[Operator Instructions] Might I kindly remind everyone to please stick to one question at a time. The next question is from the line of Andrew Gwynn from BNP Paribas Exane.

I was going to ask on Twitter actually, but I'm going to come back to the guidance. Obviously, the message here is things are going pretty well and indeed ahead of the guidance. So why not upgrade?

No, fair point. We don't want to upgrade, because basically, this is very early in the year. We see very good development, as I mentioned before. So this is our point of view, too early to upgrade the guidance. We did so for the integrated vertical, and as we -- and also mentioned in my statement before, also there, we moved from up to EUR 525 million. We've seen that the development and the development of the KPIs and the fact that we slightly reduced numbers of opening of Dmarts. We see that there is room for improvement, but this is quite too early in the year to confirm. So that's the reason why, but we had a very good first quarter all-in GMV revenues but also EBITDA wise.

Anyone who knows me and knows Emmanuel, there are no one hates more than to disappoint. And yes, I mean it turns my style, never want to disappoint. And we are probably both a little bit conservative and also what we tend to want to have a little bit flexibility. And maybe that's not the time to do that, but we don't know. We don't know what the future tells right now, everyone is constructive right now, the markets look at profitability right now, every company is doing that, but who knows what happens in one, 2, 3, 4 whatever quarters. And we don't want to be sitting in a position where we have to optimize the business in a way that we don't think is the right way. And therefore, we generally always keep a certain amount of flexibility and so on to respond, if need be. And if markets remain such as there are, and maybe there is through some room, but we are still only in April, so we like to making sure that we feel 100% certain that we can -- before exchanging any guidance, we would want to be 100% sure or even more so that we can.

Maybe if I can add a personal note. The last 2 years have been quite tremendous in terms of new events taking place. We talk about corona and that's why I think it's fair to be prudent. We had a very good start, but I think we should be prudent in that, as Niklas said, this is probably also our DNA. But that's why we mentioned it today, but we didn't feel like we have to be prudent as we start. We are just at the beginning of the year, looking at what happened in the last 2.5 years.

Okay. That's all very fair. Is it substantially better or slightly better?

I think what Emmanuel said is probably true, like I think we were EUR 23 million ahead of budget in Q1 on EBITDA. I think, on orders, we are on target, on plan and then EBITDA is EUR 23 million ahead. If you're going to be EUR 23 million held every quarter, is that EUR 23 million going to grow to more? Or are we going to use that EUR 23 million because there are comparative pressure or other things? I don't know. And that's why we're keeping the guidance as it is.

Next question is from the line of Silvia Cuneo from Deutsche Bank.

I have a high-level one on the market. Can you please share some thoughts about how you're seeing user demand evolving as these starts increasing due to inflation? Do you see the risk of lower demand as ordering food delivery becomes more expensive somewhat with service delivery fees, et cetera, that could partly explain the lack of guidance upgrades today?

Yes. I think, in general, if you look at the growth that we're having, I think it's still very strong. And overall, we wouldn't have that strong growth if our cohort or anything will be weak. And I think -- so therefore, we remain having very strong cohorts. There are GMV growth and even order growth kind of signals that, especially if you make it adjusted for COVID -- lapsing the COVID impact of last year. I don't think there is a lower demand. If so, we would have already seen it. The only lower demand is that we're lapsing and a year that was very good. So if you look at Europe, we grew 70% on a like-for-like basis. I think delivery grew something similar or maybe slightly less, I think a little bit less too. That is generally a little bit less than what we would normally have expected. But again, we had a very strong COVID tailwind last year. So, I think, the growth is still very strong even in places like Europe where it was the weakest of all. So, no, we don't see a sense. And the business, the cohorts, everything is so predictable. It's so straight that I could be pretty high certain to tell you a number or the number of -- amount of GMV this year, next year, the year after, the year after that and the year after that, assuming there is no COVID that temporarily increase or decrease orders.

And maybe on inflation because we have countries with high inflation before. If you look at Argentina, hyperinflation. Also, we've seen the situation in Turkey. And what we saw on the food business is that basically, yes, impact maybe the overall behavior of the consumer. But on the food business, very lightly, if any, because basically the transaction value for food transaction is not as important as you will buy for other goods. We think about -- for buying TV or other goods. So I think the expense that we made so far is that inflation or hyperinflation even had some impact on the macroeconomic. But for our business, the restaurant is passing the inflation of the interest of the price to the consumer. We take the fee. And basically people have to hit and continue to do so, inflation or not. So -- but it remains to be seen. This is another new view that we are facing. That's why we should remain cautious. But so far, we don't see any impact on that specific point of inflation.

Adding there, inflation is generally more in our benefit because we charge on the food value, while our costs are not increasing as fast as the food cost is increasing. Another question is, is demand going to be lower? Generally there is increased cost for consumers. Emmanuel mentioned a couple of scenarios, but of course, those markets have been used to inflation -- hyperinflation for a long time. I'll give you another example, take Greece during the crisis. The country went -- it was not inflation, but the country went through enormous amount of unemployment rate. The country went bankrupt, and our orders have never been stronger in Greece than it was during that time. People stopped going out to restaurants and start ordering because it's still cheaper to ordering than going out. That's the only explanation we have. So, even in such a disaster scenario, I think you've seen that food delivery is generally very strong and countercyclical in that sense.

Next question is from the line of Sreedhar Mahamkali from UBS.

Niklas, probably start with an observation or maybe a bit of feedback, stating the obvious, I guess, but also not looking to create a debate on this call. But most of us were surprised that the order disclosure was stopped, fully understand some of the rationale you outlined already. But it's fair to say that most of us struggle with it, but just a bit of a feedback. But my question is on Dmart, please. The top 7 countries versus all other 42, I'm curious if you can explore a little bit more. Is there anything specific about the nature of these 7 markets that you're able to get up to 540 orders per day, how long has it taken to get there? And I guess, more importantly, how much closer do you think you could get to for the entire segment, your orders per day by the end of the year? And what's the sort of path there? Is it end of this year, end of next year, how do you see that part?

So, maybe just a remark and we were a little bit surprised that it would be a big thing because we have, over the last few quarters, kind of reduced the importance of orders. We also excluded in some of the graphs in the past quarter, mainly because we don't really look at it ourselves. And we actually sensor the opposite incentive, what we actually want to do, which is to drive profitability on profitable orders. We have no problems, as I said, in that people know our order growth and we say average baskets in a couple of segments or in 3 segments where it was changed. So, therefore, everyone can back-calculate the order growth. It was, as I said, let's say, 10%, 15% lower than GMV as baskets increased. In terms of Dmart, so differences -- I would say, there is a slight difference in age of this, but there is also slight differences in number of customers that lives in those areas. So there is a little bit more density there. So, as you move out of density, that is slightly less. Having said that, the growth of those that are outside of that and also as we build -- we kind of split certain stores into 2 stores to cover a slightly larger area and so on. So, therefore, there has been over time then a little bit reduced customer density there. But the growth on orders are significant. So it takes a little bit longer to make drive it there, but it's clearly happening. Other than that, we know -- actually you've seen that the prices are not higher, the margin is not higher. It's just that we decide to do a little bit more promotions to get orders up. We have also not a quicker utilization, store manage utilization, that's one. But other than that, no.

And the overall...

And thank God, I should add, we have a lot of customers already. We don't have to acquire them very expensively. Of course, there's a slight cost of getting customers to be multi vertical. But at least, we don't have to go out and do advertisement and that has already happened in the past, and that's why it's also very synergetic to our business. Yes, sorry.

Sorry, I was just following up on like what's the path? How long do you think will it take you to get the portfolio of stores closer to 5, is it 12 months, 24 months?

I don't want to give an extra guidance, but if you look at the growth of our Integrated Verticals segment, of course, there will also be growth in the places where we already have high utilization. We might have to even increase number of stores, in this case, increase number of stores at 48 stores in this quarter. Let's assume it will be something similar in the next couple of quarters, but you see that the growth continues to grow with -- per quarter with EUR 60 million or so in GMV. Yes, so based on the slide, you can see how the growth is [ structuring ], and you can assume that the new store opening is probably more in line with Q1 than what it was in Q4 -- Q3, Q4. So that should hopefully give you some guidance on where the average order per store will move -- or average GMV per store.

Next question is from the line of Jurgen Kolb from Kepler Cheuvreux.

Just a brief one on the order data that you will not release. Maybe you could, on a quarterly basis then, indeed release average basket sizes per region as -- so that there is the quarterly trading statement at least has a little bit more regular data that you're showing to the market, which probably then includes also the idea that you have to go more after GMV and be more profitable, just as a suggestion. The question really on Dmart again, Page 26, when you say the best-performing country already generates an EBITDA to GMV margin of more than 6%, I was wondering if you could share maybe a little bit more details about this specific market. Was it really the first one that opened? Or is there any specific that you would call out of this country, which just maybe had a positive better impact than the others?

Right. So it's not the first one to open. The first store we opened in Turkey. And of course, Turkey is the only market where we have strong competition. And therefore, that is not the market where we are making the best-in-class. All the other markets, there is very limited offering from competitors, if not none. So it was one of -- the good markets are the one that are profitable or a little bit earlier stage in that sense. The main difference is actually not in -- for that country in particular. It's not that the margin is high, it's not that the delivery fee is exceptionally high, it's not that the cost is exceptionally low. It's then to have a lot of orders per store. The order attached to the store is high, still increasing, but it's high. And we don't have to promote it anymore because the awareness in that neighborhood -- or sorry, in that a country is a little bit -- is there. So there's less kind of [indiscernible], reduced delivery fee, there is probably a little bit less of this 10-minute type of delivery promise. But there's nothing else there, all markets will get there. I don't see a reason why not all markets will get there. The question is more what -- is this a service that everyone can have access to today? Or is it more of a premium service that we offer because we can -- there needs to be either a slightly higher margin than the store or some delivery fee in order to make the economics that we target. So yes -- and I think, in most markets, this mainstream product, but of course, in some low-cost countries, this could be more of a premium service, and therefore, a little bit less reach, at least for the next 5 or 10 years. So, the question is not if it's going to be profitable, not. It's going to be profitable and probably more profitable than our food business, it's more how biggest you're getting. And thanks for the feedback, by the way. We are way more open to think about basket, because basket -- increase in basket makes sense, increase in orders is actually against what we're trying to do.

Next question is from the line of Andrew Porteous from HSBC.

I guess just echoing those comments around sort of the order disclosure, it would be -- I guess I can understand why you've changed the way that you've run the business, and I understand the reasons for that. But at the same time, investors can sort of do the work and see the improvements you're making even if you continue to give us order numbers. So it would be helpful to continue to get them. I don't see what you gain from taking Hawaii. In terms of the question, I was really sort of trying to drill down on some of those 7 best-in-class Dmart countries. And could you give us a little bit more color there on the profile of those markets? I'm just wondering, are there markets where naturally you get sort of higher average basket and a little bit lower cost and maybe they're sort of non-economics that every market can get to? Are they more MENA focused, a bit more Asia-focused? I don't know. But if you could give us more color, that would be helpful. And then, perhaps if you could just talk a little bit about the operations. Are you -- do you still get -- or do you get direct supply from suppliers? Or do you still have to deal with wholesalers?

Yes. No, there is -- as I alluded to before, there is not that they have better margins or that the delivery fee, as you can see here is the substantial before the delivery cost is lower. It is a slightly lower delivery cost, but that is more because we deliver slightly lower and a little bit more stacking. The difference is really -- and as I said that, yes, we can have 1% higher delivery fee since we have a little bit more volume, can have a little bit more advertisement. We can have a little bit lower delivery cost because we can't deliver slower, [ but a little bit more stacking ]. More importantly, we have 4% difference because bigger cost, because you could double number of orders on the same number of staff. So that is the difference there. That's why that is half the difference, same with other costs, which mostly includes things like shrinkage where we don't have enough product items and so on. So those are the main differences. And the question, sorry?

Yes, that's what on direct procurement. I mean, you can access directly to suppliers or not. And I mean, I can start to answer, if you want that detail, which will depend on the volumes in the countries. What we see is that we have more and more inbound calls. So basically, large names, reputed brands that are calling us to get in touch with us to negotiate or to see how to partnerships. So yes, there are more even counts at the beginning of the offer as we started this industry, but there are also regions where you do have to reach very big volumes in order to be directly in contact with the suppliers. So this is not one rule for the market or globally. This is much more regionalized and sometimes you do have to get to a very high level in order to get directly in contact with the suppliers. But as I said, there are more and more inbound calls. And there are meetings taking place also like in Berlin to negotiate or to see how to collaborate directly with the big brands.

And I would say, we do most of the sourcing ourselves. We have discussed with partners who have bigger volumes than us who have massive retail distribution and bigger volume. They are willing to find deals with us that we will use their procurement capabilities. But of course, they were limited to their inventory or at least we need to buy more than 80% or so from their inventory and only have 20% kind of unique inventory. When we do that, it would be a clear benefit on price. That would save us somewhere between 6% and 9% instantly from today to tomorrow improvement in procurement capabilities in the cases where we have seen this, maybe not all markets, but at least in those markets that we have had those discussions. But of course, it limits us a little bit in our flexibility. I should also add that the 6% gross EBITDA margin country that we mentioned here and the best-in-class 7 countries, which are already significant gross profit contributing in line with our food business. This is before -- or like before a substantial advertisement revenue. We believe still that the advertisement revenue for this business is more between 5% and 8%. Our competitors are -- or one of our competitor at least in this space already claimed to be at 8%. So you have an additional 5% on that margin. Also, the product margin is 25%, as is shown here. We do believe that it can be -- that we can find ways to improving that to 5% to 10% as well. So you have essentially an improvement of 10% to 15% on the margin bottom line that can be improved over time by procurement as well as advertisement revenue. So, coming back to the rule, this is why we are so overly enthusiastic about this industry and why we think it makes so much sense for our business to do it. We understand that now we have built out the recent coverage we don't have to go wild on investment into more stores until we have proven the concept. Now it's more about making sure that we get the number of orders per store, as we can also prove to you and the rest of the world that we are incredibly good in this, and it makes a lot of sense for our business. And once we get there, this will be also highly profit contributing. The business model itself is, in my view, at least as good as the food delivery business if not better. There are even more margins and more levers to pull here in the rest -- so yes. I hope and of course, I'll be urged by it. I hope this is very similar to the logistics business a few years ago. No one has taken as much negative feedback and anger as when we start pushing our own delivery earlier and before our competitors because it was diluted to our profitability. We still did it because we thought it was the right thing for the long-term. The same with this business model, we think it's right for the long-term. It's not very long term either until we're getting there. And I hope I'm right. And you can judge by the results, and I hope you can prove this sooner rather than later.

That's really interesting color. What do you think -- tell us whether it's a premium niche business or something much bigger?

Yes. I think in most markets, it's probably not. I think in some markets, it's probably a little bit more of a niche business. In those markets, we also have the store concept where we don't have to build the stores. We have something like 100,000 stores today or vendors today. So -- and we have a good coverage of both grocery as well as other retail items in our platform. But since we think grocery is so important, we want to offer the best service. We think this is the best service, and it cuts a lot of cost as well. But it could be that in some markets, it's more of a premium service, in most markets, probably more on midstream, but it will take long time. The same with food. And the reason why we have been growing every year for the last -- we have some brands being for 20 years and has not stopped growth. It's because the cohort is stable. That means every month or every quarter, there is growth in our business. And that it will happen next year, that will happen a year after, will happen a year after, will not stop. The cohorts are what they are, and that's why there's a constant growth in this business, anything else is that you're executing poorly or there is a reverse of pandemic or something else. But if you execute okay, there will be a constant growth of the business for 10, 20 years. And food -- people ordered food also 20, 30 years ago via phone, and it's taken us this long even in U.S. It's taken in U.S., a market that is highly online, 20 years of good livery and still DoorDash, Uber, Grab or at least 2 of them are actually growing, if not 3, despite being around for so long. And they will still be growing in 10 years. It is a slow adoption and sales of living and behavior that has taken that long. Anyone believing that there is from today to tomorrow, everyone would order grocery online or the quick commerce instantly, they will be very wrong. This will still grow very fast in 20 years from now, hopefully faster growth than food delivery, but it takes time to build a behavioral change. And that's why it might also be a little bit niche for the next years, but this can be a mainstream in the long-term.

Next question is from the line of Sarah Simon from Berenberg.

Yes. Niklas, I've just got one question. I may be allowed one question. Is the -- basically you have talked earlier about not having to acquire the customers for the grocery business. So, can you give us an idea of what percentage of your food delivery customers are ordering through your quick commerce business today? And maybe you've got an example of your best-in-class market and then across the average in the country is obviously where you've got quick commerce that would be helpful.

I don't want to get through this new disclosure, but -- and I also don't have the exact number. But in order of magnitude, 20 in some cases where we've been around. But food has been around shorter and quick commerce have been around longer, it's even in the 30%.

So that was the last question, Niklas. Would you like to close with some final remarks?

Well, thank you, everyone, for attending the call. I know it was very long. I'm sorry for that. But we felt it was important to answer all your questions, and we're happy to follow up with further questions. We know it's very challenging times. We try to do our best by showing transparency on cohorts and user behaviors and Dmarts and subscription and ad tech and all of that as much as we can. We believe we have a fantastic business. I hope that the market will realize that too. And I think we're in a great direction, and I hope it will be a good year. And thanks for your support and listening today.

Thank you, everyone.

Thank you all for attending. Operator, you may now close the call.

Thank you. Ladies and gentlemen, the conference has now concluded, and you may disconnect your telephone. Thank you for joining, and have a pleasant day. Goodbye.