Deliveroo plc (ROO) Earnings Call Transcript & Summary

Hello, everyone. Good morning. I'm Will, Founder, CEO of Deliveroo. I'm here with Adam, our CFO. Thank you all for joining today for this morning's presentation of our full year '21 results. Just want to warn everyone upfront. This is going to be a pretty long presentation. We've got a lot to get through. So thanks in advance for bearing with us. Another thing I wanted to say upfront as well is, we have all been horrified by what we've seen in the Ukraine, and we hope to see an end to the invasion and the violence. At Deliveroo, we have a small number of contractors working in our tech org, and our priority has been to make sure that they're all safe and to provide any support that we can. We also have some colleagues with family in the Ukraine, and we've made sure they are supported at what is an extremely difficult time. As a company, we're making a donation to the Red Cross, and we're looking more at what we can do. Now we'll go through our results in detail and so why don't we start off and take a look at what we'll discuss today. We'll start with a very brief overview of '21, before we run through a summary of the key business highlights from the last 12 months. Adam is going to talk through the financial results in more detail. And after that, and this is where we'll spend most of our time this morning, I'm going to set out our path to profitability, which is part of our new guidance that you will have seen issued today. So let's get started. Overall, I'm really proud of our performance in '21. We grew very quickly across all of our markets, and we have continued to make good progress in executing our strategy. In terms of the key takeaways from today's presentation, there are a few. Firstly, full year gross transaction value, GTV, was up 70% year-on-year in constant currency. Gross profit margin was 7.5% as a percentage of GTV. We have continued to scale our differentiated consumer value prop with continued momentum in Plus, Editions, and on-demand Grocery throughout the year. We exit the year in a strong financial position with over GBP 1.3 billion of cash and cash equivalents on the balance sheet, which positions us well in an increasingly uneasy macro backdrop. Fourth, we are providing guidance for '22 today, guiding to 15% to 25% GTV growth, as well as negative 1.5% to 1.8% adjusted EBITDA margin for the full year. And finally, we are outlining today our longer-term path to adjusted EBITDA profitability, within which we expect to reach breakeven during the second half of '23 to the first half of '24. And by '26, we expect to deliver an adjusted EBITDA margin of 4% plus with further upside potential beyond that. So we're excited to take you through these today. First, let's turn to our marketplace performance. I know you've seen this slide a number of times, but I think it's really important to highlight our complex three-sided on-demand marketplace. We think of each side are riders, restaurants and grocers, and end consumers as customers of Deliveroo. We balance the interest of all 3 sides as well as Deliveroo. But ultimately, the strength of our performance is driven by how well our proposition is working for each group. Throughout '21, we've seen strong engagement from each side of the marketplace. First, our average monthly active consumers increased from 5.9 million at the end to 8 million at the end of '21. This is our highest ever monthly active consumer base. Rider satisfaction has remained strong at 85% across our global network of over 180,000 riders. We also continued to add great restaurant selection. We've got now more than 148,000 partner sites live on the platform globally at the end of '21. That compares to 102,000 at the end of 2020. And we have rapidly expanded our grocery rollout. We have over 11,000 partner sites live now at the end of '21. Again, that's up from 7,000 at the end of 2020. This all drove our financial performance in '21. We have seen 73% order growth year-on-year. Gross transaction value was up 70% to GBP 6.6 billion pounds, showing continued strength despite reopening effects and an increasingly tough comparison base throughout the year. Gross profit is up 43% to GBP 497 million. Gross margin, as a percentage of GTV, was 7.5%. This is down from 8.7% in 2020. And this is due to accelerated investments in consumer acquisition and retention to support future growth, as well as some of the reversal of the benefits from higher basket sizes during COVID-related lockdowns in the prior year. Adjusted EBITDA was a loss of negative GBP 131 million compared to the negative GBP 11 million loss in '22 as higher aggregate gross profit was offset by increased investment in marketing and technology to support future growth. So overall, we've seen strong engagement across each side of our marketplace. The U.K. and Ireland segment has delivered strong growth with consumer engagement remaining robust despite the reopening from COVID lockdown restrictions throughout the year. We saw order growth of 72% year-on-year, GTV growth of 71%. We continue to gain market share in '21, because of sustained execution on improving our consumer value prop neighborhood by neighborhood, and then layering on our growth initiatives, which have been geographic expansion, Plus, Editions, and Grocery. We'll come out on to Plus, Editions, and Grocery in a minute, but I think it's worth calling out here our geographic expansion in the U.K., which now allows us to reach about 77% of the U.K. population at the end of '21, which is up from 53%. What I think is exciting about this is that the vast majority of our '21 growth came from neighborhoods where we already were operating before '21. So we think about this expansion is planting the seeds for future growth. Now I know some of you have questions around profit potential in midsized and small towns and cities in the U.K. and we'll give a few examples on this later. Finally, I think it's also worth calling out. We executed really well in a very competitive market, I think, in a very thoughtful manner. We talked about the market share gains, but we also generated a geographic segment adjusted EBITDA. This is pre-central costs in the U.K. and Ireland of GBP 91 million in '21. So overall, we feel really good about the U.K. and Ireland in '21. We also saw strong year-on-year growth in our International segment with orders up 74% and GTV up 69%, despite a tough comparative period with many of our international markets experiencing lockdowns throughout 2020. We've been able to strengthen our position across a number of major cities and key markets by taking this playbook that we honed in Manchester, we talked about on previous calls into cities such as Marseille and Rome, as well as other cities in international markets such as the UAE. We've also continued to expand our on-demand grocery proposition into international markets. We rolled out with key partners such as Carrefour in France, Italy and Belgium. Also rolled out with Casino in France. We signed Park N Shop in Hong Kong, also signed Picard in France as well. So now we have approximately 6,000 grocery sites with major partners and smaller independent grocery partners across our international markets. So this has really strengthened our consumer value prop. Geographic segment adjusted EBITDA, this is pre-central cost, this was negative GBP 56 million in '21. This compares to positive GBP 6.9 million in 2020, as increased aggregate gross profit was offset by continued investments in growth and the reversal of COVID benefits. So overall, we're seeing good performance internationally as we continue to develop our CVP hyper-locally to drive future growth. Okay. Now moving on to Plus. I know none of you -- some of you have had a number of questions about Plus and I think we made great progress in '21. First, we launched our Plus Silver tier. This is a lower monthly fee, but with a GBP 25 MOV, primarily targeted at families. So that was launched in Q1. And in September, we launched a partnership with Amazon Prime, allowing all U.K. and Ireland Prime members to sign up for free delivery Plus Silver for a year. We saw really strong traction in the Plus subscriber base globally. That tripled in '21, the number of subs. And then the UKI subscriber base was specifically up 4x compared to the end of 2020. So sign-ups have been good, but I know some of you have questions on the unit economics, so I thought we would clarify things here. So what we put on the slide here is a matrix to give you a bit of color about how our various Plus tiers rank, including pay-as-you-go. So you've got 2 axes. One is average order frequency and the other is per unit economics. So for the -- in order of unit economics, the fee-paying Plus Silver consumers ranked the highest, because they have a GBP 25 minimum order value and that GBP 3.49 a month subscription fee, this is pounds. They also ranked second in terms of order frequency. So they're some of our most valuable consumers. And then consumers that are on Plus Silver, who aren't paying the subscription fee, such as those on a free trial, they rank slightly lower than those -- than pay-as-you-go in terms of unit economics, because we're not collecting that monthly fee, but there's still a pretty high minimum order value. These consumers order a little less frequently than both the paying Gold and Silver subscribers, but they do order more frequently than pay-as-you-go, which makes sense, as they have the benefit of a membership, but they're not paying for it. And the net effect of the unit economics and the frequency is still very positive for us. Now Plus Gold, these consumers order the most frequently, but due to their lower MOV of GBP 10, they have the lowest unit economics of the four, but their frequency more than offsets any unit economic dilution, and it drives significantly higher CLV than pay-as-you-go consumers. So if we take the U.K. where all 3 tiers of Plus are active, the 52-week CLV for all of our Plus programs is significantly higher than our pay-as-you-go CLV. This is measured across 52 weeks. So it's near term. With any unit economic dilution offset by higher frequency and retention. And this is fundamentally why we're so excited about the program, but we're also incredibly excited to drive some of the benefits to our user base as well. Okay. Now on to Editions as well. This is another one of our innovations that drive long-term differentiation of our consumer value prop. So just as a reminder -- quick reminder here, Editions are our delivery-only kitchens that unlock additional value for the 3 sides of the marketplace. Restaurants, what do they get? They get a way to expand in the new geos. They get to segregate delivery from in-house. It's a different P&L model as well, because you don't have front of house and the rent -- the implied rent is a fraction of what you would spend on the high street. What do consumers get? They get the best food content out there, the best brands and a better delivery experience. It's about I think, at this point, 4 minutes faster than the average restaurant. And then riders get more opportunities to work, because the wait time at Editions is low. The stacking possibilities are quite high. And as you can see from this chart, we weren't really able to roll out many of these delivery-only kitchens in 2020, because it was pretty difficult to secure sites during COVID. But really excited to say that we managed to accelerate the rollout of Editions during the course of '21. We added over 100 kitchens or so in '21 with nearly 70 of these opening in the second half. So we brought brands like Dishoom, Shake Shack, Five Guys and Pho in new neighborhoods across the world, which is really exciting. And this acceleration of our rollout meant that the vast majority of our CapEx spend in the second half related to Editions. And I think the cool thing about Editions is we have been doing this for 5 years now. We have a very, very good handle on the unit economics and the return on payback of the CapEx. And we're really, really excited about continuing to accelerate this rollout throughout '22. All right. Let's move on to Grocery. In '21, we've continued to rapidly develop our on-demand Grocery offering. The Grocery offering offers powerful synergies with the core restaurant platform. It's about 100% incremental demand to the restaurant offering, which I think makes sense, because it's very different occasion. And it also provides a very effective consumer acquisition channel, because these brands are really well known. Our partner brands are really well known. So on-demand Grocery now is around 8% of total group GTV. Like I said before, we've got 11,000 partner grocery sites live globally at the end of the year. Now today, I wanted to kind of explain some of the different models we have. The vast majority of what we do today is store pick. So these are orders that are picked and packed by our partners in existing grocery stores. And as I mentioned before, one of the issues with the store pick model is stock accuracy. We're working to improve stock outs, and we've made a huge amount of progress. But at the same time, we also launched a new rapid grocery delivery service called Deliveroo Hop in September '21. So Hop operates from delivering only micro fulfillment centers. I guess, some people call them dark stores. These are run by Deliveroo. We work in partnership with established grocers, such as Morrisons and it enables deliveries in as little as 10 minutes, but with a much, much, much greater inventory accuracy and a wider product range. Now these first Hop stores were established in London in partnership with Morrisons. And since then -- since the end of '21, we've actually launched a few more. We've launched further Hop partnerships in the U.K. with Waitrose. We've launched one in Italy with Carrefour. And I think it's really exciting, because obviously the great consumer experience. But when I look at our existing marketplace, I think we're just really well positioned, because, A, we have the consumers and we are the leader in on-demand grocery; B, we've got the existing logistics network, that is very efficient and then we have the strong relationship with grocers. So when I think about this, when I put all the pieces together, now we understand for Hop the consumer proposition is really good, both in terms of stock accuracy and delivery speed. And we've been really proud of what we've achieved on the retention and the speed side. It's really pretty amazing. But it's also not crystal clear to me yet that Hop or dark stores can be a profitable product. I'm talking on a fully allocated basis, CapEx, OpEx, everything at scale. And so we're monitoring this really closely. We're being prudent about our rollout plans, but the impact on consumers is definitely evident and it's been great. Now I think overall, if we think about really dense, urban areas, I do think that dark store models can work. And I think it's in these types of areas where we've been focusing our efforts. And ultimately, we expect this on-demand Grocery space to be served by a mix of dark stores, the store pick model and maybe some sort of hybrid pickup and delivery site. And we're excited about how well positioned we are to be successful with all of these models. Now nonetheless, I do believe the largest part of Grocery will still be the store pick model. And so we've been really focused on making technology and operational improvements. So let's go into a few examples here. So back when we first launched Grocery with Co-op, we did so by using the same in-app interface that we had built for the restaurant business. We sort of jammed Grocery into there. And you can see that on the image on the left here. Now we knew this wasn't the best consumer experience, but we wanted to move fast, which I think was definitely the right move. But since then, we've made really big advances in the platform for grocery. So this last year, for example, we launched a new merchandising navigation. You can see that on the right. where you've got improved search functionality, and it's just much easier to find. This is, I think, we call this aisle shopping, so it mimics what you would do in a grocery store. And I think this has transformed the user experience. It's enabled consumers to easily navigate large menus within the app, because these types of menus have 2,000, 3,000, 4,000, 5,000 SKUs as opposed to a restaurant menu that might have 20 items, right? So it's just really different. So I talked about the consumer side of Grocery. Now let's just talk a bit about the operational side, the part maybe you don't see day-to-day as a consumer. So what we're doing here is we're transforming the picking experience for our grocer partners, the in-store pickers. So what we've done is we are in the process of developing a new dedicated picking app that enables multiple pickers in store to pick different orders at the same time. So really helping our partners out on the efficiency side. So there'll be a new scanning tool within the app. This will allow pickers to scan items as they pick them, which will reduce accuracy errors. We'll also improve stock availability information by increasing availability updates much more frequently. And then with the pickers updating individual item availability themselves during the pick walk. We've also launched a new substitutions process as well. This allows consumers to have the option to update their unavailable item settings to either substitute or with the best available item, to remove unavailable items or just to cancel the whole thing. And our API enables our partners to see, which one of these options the consumers amended too, so we can change that order around. Now if you just look on the slide, on the left is basically what we launched with, which was again something built suitable for restaurants. And we've been developing technology to make it much more specific for groceries. That's kind of what you see on the right on this app that we're talking about. I guess, finally, before ending here, one of the areas we're most excited about is the opportunity for advertising partnerships with FMCG players. This is an area that is super early in development, in particular compared to our restaurant ads platform, which is live. But it is a proven large opportunity for online platforms. And it already represents a meaningful part of revenues for some players. So we'll touch a bit on that more on the path to profitability section later, but really excited about the advancements we've been able to drive in grocery. Now on to riders. So just quickly on the rider proposition itself. I know I mention this every time, but I think it's always worth reiterating. So Deliveroo's rider proposition offers 2-way flexible work. Given there's some focus on how we work with riders, I wanted to recap what this means. So what does this mean? Well, first, it means, A, you can choose when and where you want to work. You just log in whenever you want. You can work with multiple apps at the same time, so you can be working on different platforms and decide which orders you want to do. To that point, you decide for yourself, do I want to accept this order, do I want to reject this order, with no penalty. And then finally, you get paid per delivery completed. So we offer this type of work, because it is what riders tell us they want. Flexibility is the #1 reason that riders choose to work with us. And I think the thing that is really, really -- that we're really proud of here is that this year, we've seen clear evidence of the popularity of the work we offer. Because rider supply has remained extremely strong despite record levels of employment vacancies. And if you look at the chart here, you can see the evidence of that. Well, it's evident in most markets, but the chart here is about the U.K. specifically. So as the U.K. economy reopened, there was massive growth in employment vacancies. This was up 105% since Jan '21. Throughout this period, rider supply also increased with the number of active riders up 36% across the year. If you look in the fourth quarter, you see a spike there, and we -- this was no exception. So we usually expect to see more riders working with us as this is seasonally our busiest period. And the growth in riders in the platform is really all the more remarkable this year, though, because this is when employment vacancies were the highest. So added to this, despite employment vacancies available in Q4, we were still seeing average weekly applications of over 11,000 a week in the U.K. And not only has there been steady rider supply, we've seen very consistent rider retention over the last 12 months, which you can see on the chart. So overall, what we're seeing is high numbers of riders joining Deliveroo and then staying with us, making a positive choice to offer self-employed work, when alternatives are clearly readily available. And so I think we're just really proud of this, because again, I think attracting and retaining riders in such a tight labor market is something we're really, really proud of. And really quickly, before I forget, alongside flexibility, though, we have always prioritized offering rider security. And I'm proud that in Q3 last year, we strengthened our offer to riders. So all of our riders are insured. We extended this to cover periods when riders are unwell and they cannot work, and we also offer one-off payments to riders who become new parents, so they can spend time with their new child. So this is available today in several markets, and we aim to extend this benefit to more markets in '22. So I think back to what I was saying, the things that I'm most proud of in '21, I think it's attracting and retaining riders is in such a tight labor market. And I think it's quantifiable proof that people value -- highly value this type of work when there are so many other options available. Like all platforms, Deliveroo is engaged with courts and regulators across markets and issues concerning rider status. Where there are challenges to our model, we robustly defend riders' ability to work flexibly. We want to be transparent on this issue, and we're going to highlight what we consider to be key developments on these issues in '21. In the U.K. in June '21, the Court of Appeal confirmed that Deliveroo offers self-employed work. This was a unanimous decision and the fourth judgment to reach this conclusion. In France, in April, the Paris Court of Appeal confirmed that riders are self-employed. There are a number of ongoing cases concerning historical models in France, which we are defending. The French government has made repeatedly clear that they support riders being self-employed. They are introducing new rules for platform workers, but this is on the basis that their self-employed status is protected. And only this month in March, the Minister for Labor addressed platform workers through an online video and described the government's ambition as being "enable you to benefit from new rights and to carry out your activity under better conditions whilst respecting the self-employed status to which most of you are attached". I think it's significant in our view that a major European economy is thinking this way, offering riders the flexibility that comes with self-employment with increased security, which is exactly what we support and have been talking about for the last 6 years. Now why don't we turn to the European Union proposals. This is a complex issue, so I just want to go over the key points. But in short, the EU wants to put in a law -- put in law, a clear definition of a self-employed platform worker and EU themselves say that the majority of platform workers would remain self-employed as a result of this proposal. Now greater clarity is something we support. And overall, we welcome the proposals. First, the definition that has been outlined is broadly in line with our way of operating and working with riders. Second, we have had multiple core judgments in the EU confirming that Deliveroo riders are self-employed. It's worth saying that we are at an early stage in the process, and we are engaging on the ongoing consultation, but we are aligned with the intention of these proposals, and we are confident we are in a strong position across our EU markets. Linked to this, at the very end of last year, the Belgian government committed to introducing a policy that is very similar to that the EU is proposing. And based on what we know now, we don't think our model will be impacted by this new law, because our riders have already been found to be self-employed following a court ruling in December last year in Belgium. In Italy, Deliveroo continues to work in an industry-wide collective bargaining agreement with the local trade union, which legally recognizes riders as self-employed. There are a number of challenges, which are ongoing in Italy, which are subject to appeal. And there has been previous speculation over a sector-wide investigation by the health and safety authorities. This has been settled with each platform paying approximately EUR 35,000 each. Now tying all of this together, as we think about how we stand in our markets, we have to look at both the political situation, which is how governments think about platform work and the actions they're taking, and the legal situation, which are the results of court judgments concerning Deliveroo. On both of these accounts, we're in a strong position. We see major European governments taking action, which we believe will enable us to continue working with self-employed riders, and we have had multiple judgments confirming that our riders are self-employed. Deliveroo will continue to disclose any material developments in relation to rider matters, and we're happy to take questions on this or to provide more detail in the Q&A or individually. I'm now going to hand over to Adam to go through our financial performance. Thanks.

Thanks, Will, and good morning, everyone. I'm going to start with a quick recap of some of the full year numbers that you saw previously at our Q4 update and then move into the rest of the P&L. Let's start with average monthly active consumers. Across all of 2021, we have averaged 7.5 million monthly active consumers. This is up from 4.6 million in the prior year, representing a 64% increase. The split of the consumer base is broadly equal between our two geographic segments, and the growth has been consistent across both. We achieved global orders of 301 million, up 73% year-on-year, while GTV increased to GBP 6.6 billion, up 70% on a constant currency basis. Again, both geographic segments grew at similar rates, contributing fairly evenly to overall group growth. Gross profit was up 43% year-over-year. As we said at our interim results presentation, we started to accelerate the pace of our investment in many of the areas above gross profit at the back end of 2020, which has contributed some 2021 growth, but is also planting the seeds for future growth. This investment includes incentivization for new consumers to join the platform via our new user experience program, incentivizing consumers to move to our Plus program to drive retention and frequency, and continuing to build and maintain differentiated content. These investments are reflected in gross profit as a percent of GTV, which at a group level reduced to 7.5% for 2021, compared to 8.7% in 2020. At a segment level, UKI gross profit margin was 9.3%, compared to 10.4% in 2020, while International was 5.5%, compared to 6.9% in 2020. We started to see a margin decline in H1 2021, particularly in the International segment, which dropped from 7.2% to 6.1%. This was driven by investing in consumer acquisition and retention, mainly in France, Italy, UAE and Kuwait; driving differentiated content, for example, widening our grocery selection and investing in preferred partnerships in France, Italy and UAE, and AOV has also reverted towards its pre-COVID levels, as COVID restrictions eased throughout H1 with timing differing by market. As you can see on the charts, there was a sequential step down in aggregate gross profit in H2 2021. We saw the expected reversion of AOV towards pre-COVID levels, along with continued investment to drive consumer acquisition and retention and differentiated content, which is in line with the expectations we outlined at the interim results. In the International segment, in particular, if you look at this on a quarterly basis, we saw a GP margin decline from Q2 to Q3 in 2021, then saw Q4 margins back in line with Q2. And we've seen improvements continue into Q1 this year as well. So the blend across halves has masked the trend a bit. Now on to marketing and overheads. As this slide shows, we have stepped up investments overheads in 2021. As a reminder, we had a relatively slow investment pace in the first 3 quarters of 2020 due to both the uncertainty around the impact of COVID, as well as the ongoing CMA antitrust investigation that resulted in us needing to exercise a high conservatism around our deployment of capital until Q4 2020. Throughout 2021, we've been in a position to invest where we see the ability to drive durable long-term value. As such, marketing and overheads increased to GBP 629 million compared to GBP 359 million in 2020. Taking the 2 buckets separately, starting with Marketing. This increase in spend came across 2 distinct areas. First, we invested in growth marketing are activities that are directly attributable to the acquisition of new consumers. It is still very early in the industry shift online, so we want to capture that opportunity as consumers make that shift. And second, we invested more in brand marketing to drive awareness in both restaurant and grocery online food delivery and awareness of our brand. On the overhead side, the biggest driver of the year-on-year increase by quite some margin was investment in our technology and engineering team to build durable advantages and points of differentiation for the business. We think about the technology investment across 3 buckets. First, getting better at things we're already doing, but where we're still at the beginning of the journey on some of these things like revenue generation through our advertising platform or better personalization. Better network efficiencies are driving efficiencies through self-serve capabilities for consumers, partners and riders. Those are things we've been at for 9 years, but there's still a ton of opportunity ahead. Second, we think about enabling our long-term innovations and initiatives. A good example of this is our core Grocery business and the progress we've made on that, which you saw earlier in some of the materials that Will shared in terms of the improvements we made on technology and experience side for both consumers for the Grocery partners. Going forward, this includes things like our Deliveroo Hop stores. Finally, we provide the supporting infrastructure required for scaling the business effectively, for example, platform stability and continuous improvement on our models for forecasting consumer demand and rider supply. You also have some of the overheads of scale with growth, such as customer restaurant support that increased the number on an absolute basis, but where we continue to drive more efficiency on a per order basis in 2021 versus prior years. And finally, let's talk quickly about adjusted EBITDA. All of what I've just laid out means that we've seen an increased adjusted EBITDA loss in 2021, with higher gross profit more than offset by increased investments in marketing and overheads. As you can see from the chart on the left, we ramped up investment in the second half of 2021 as well as seeing the reversal of benefits from higher basket sizes during COVID-related lockdowns. You saw GBP 23 million of this shift coming from the decline in GP margin half-on-half that we've already talked about. The remainder is a step-up in overheads and marketing, the bulk of which came from investments in scaling our technology team that we talked through on the previous slide. Let's move on to our cash balance. We continue to have a very healthy balance sheet and are well capital to go after the opportunity in front of us. You'll recognize the left-hand side of this bridge from what we shared with you at the first half, which took us from a pre-IPO cash and cash equivalents balance of GBP 379 million to GBP 1.6 billion. We flagged at the time that we saw a positive impact in H1 from net working capital with GBP 132 million of the GBP 184 million inflow relating to timing of employee tax and social security payments related to the IPO. As expected, this unwound in the second half, contributing to a working capital outflow of GBP 188 million, with the remainder of the change due to timing of restaurant payments in December versus June. Before I finish, I want to take you through an update on our consumer cohorts. Historically, we've seen our consumers increasing their spend with us in every subsequent year, as the steady increases in order frequency more than make up for any incidental churn in our consumer base after the first year when retention is quite stable. As a result, we've seen a net GTV retention of more than 100% from our existing consumer cohorts each and every year. And we have been layering on a larger and larger new consumer cohort each year as well. The combination of those 2 effects: one, greater than 100% net GTV retention from our existing consumer cohorts. And two, supplementing with increasingly larger new consumer cohorts has been what drives our growth each year. But underneath this is a constant evolution and improvement of our consumer value proposition to give consumers reasons to transact with us more frequently year after year. This view is across the whole group. So I wanted to dive into a couple of specific markets as well. As I just mentioned, we've seen our consumers increasing their spend with us in every subsequent year, as they increase order frequency. This table shows how frequency has evolved for our U.K. cohorts dating back to 2015. Before the pandemic, the trend was for annual cohorts to increase average order frequency each year, as you can see, if you follow the line horizontally across for each cohort year. For example, the 2015 cohort had an average order frequency of 2.9 times a month in Q4 2016 and has increased every year to now order 5.4 times per month as of Q4 2021. While it isn't shown in the chart, consumers who joined our platform in 2014 in the U.K. have ordered on average, over 100 times in the last 12 months. As you might expect, in 2020, the rate of annual increase in average order frequency stepped up during COVID lockdowns, but encouragingly, each annual cohort has continued to see frequency increase in Q4 2021 versus Q4 2020, even as COVID restrictions that unwound in Q4 2021. Now let's look at the same data, but for one of our international markets, the UAE. While the U.K. is our largest and most mature market in terms of the time we've been in the market, when it comes to order the UAE is our most mature market. You can see on the table that in some cases, cohorts in the UAE have average order frequency have doubled out of the U.K., coming in at over 10 orders per month. Similar to the U.K., all cohorts saw a large rise in 2020 as COVID lockdown restrictions were put in place. And equally encouraging is that we are seeing no material impact on consumer engagement trends post reopening. And with that, I'll hand you back to Will to outline our path to profitability.

Adam, thanks a lot. As I said at the beginning, a very key focus for us in '22 and beyond is making progress on our path to profitability. We have set out new medium and long-term guidance today, which you can see on the page. We expect to reach adjusted EBITDA breakeven during H2 '23 to H1 '24. And by '26, we expect to deliver an adjusted EBITDA margin of 4% plus with further upside potential beyond that. I think it's worth noting through this presentation, references to profitability mean on an adjusted EBITDA basis. We, of course, recognize that being adjusted EBITDA profitable is just a step towards being truly profitable and cash flow generative. All right. So let me go into what we're going to cover in this section, the path to profitability section. So first, we're going to go into '26 adjusted EBITDA margins and the levers to get there. We're also going to go into the path to reaching breakeven by H2 '23 to H1 '24. Adam is going to take that section. We're also going to share some details on markets where we are already adjusted EBITDA breakeven on a fully allocated basis. And then I think we've been asked a lot of questions about hyper local profitability. So we wanted to share examples of profitability in different types of cities and towns, both in the U.K. and around the world. So with that, let's get started. First, we're going to dive into these longer-term margins and levers to get to 4% plus adjusted EBITDA margins by '26. Let's look at the slide here. If you look at the left-hand side, if you look at the purple shaded numbers, this is our EBITDA margins. This is what they've looked like historically. In 2020, we've had some exceptional impacts that drove profitability faster than before. But previously, we exhibited steady improvement over time. Now during 2020 and H1 '21, we had lockdown-related tailwinds, and we had conservative use of capital. Now you got to remember, this was due to the unconcluded CMA investigation where we didn't receive the cash. There were related layoffs associated with that. And then there was frankly just pretty big uncertainty across the industry at the beginning of COVID in 2020. I think people forgot, but almost all the restaurants were shut for dine-in and delivery. So people didn't know what was going to happen. Now all of that, that sort of lockdown related tailwind, the conservative use of capital, that drove reduced losses. So we had a negative 0.3% adjusted EBITDA margin for '20 and with a positive number in H2 '20. Now '21 saw a step up in investments for future growth, as well as the reversal of some COVID benefits. So that led us to a negative 2% margin for the year, with negative 3.2% in the second half, which we do expect to be the peak margin loss for the business from here on out. All right. So let's get into the levers that lift us from that negative 2% EBITDA adjusted margin in '21 to positive 4% plus by '26 that we just talked about. So what we've done in this chart here, we've laid out our P&L on a margin basis. And ultimately, what needs to happen both financially and operationally to get us to that 600 basis points uplift over time. Broadly speaking, we see this coming relatively evenly from both gross profit as well as marketing and overheads. So that's about 250 to 300 basis points each. There's obviously a lot more detail behind our plans than that. So let's just walk through the line items over the next few minutes. So if we start with net revenue. What is net revenue? That is, a, commissions for merchants, and b, consumer fees. And that includes both Plus as well as our pay-as-you-go fees and all their different components. And the third element is advertising revenue, which we've just begun scaling. These net revenue figures, they are inclusive of refunds as well as certain types of discounts, which we consider contra-revenue. Now if we look at commission revenue, what is that? Well, that is commission revenue from restaurants and grocer partners. That's the largest component of net revenue. And that's ultimately the product of basket size, otherwise known as AOVs and commission rate. Before this year, we really had an actively managed basket size, and really treated as an exogenous factor, but we're working on managing minimum order values more algorithmically. And also pretty excitingly, we shipped our first, I'd say, real scaled upselling initiative. In addition, basket sizes for Grocery are already slightly higher than that for restaurants. And so we do see scope to move into higher Grocery basket sizes over time here, as we add more SKUs to the offering. We've been doing that both in our store pick offering as well as Hop, and we have seen positive results there. Now in terms of commission rates themselves, we really don't expect to see much upwards or downwards movement on segment level commissions. They've been stable for quite some time. They really haven't shifted, but we might be impacted by some mix shifts. So for example, if we do more grocery over time, then we expect, obviously, the overall commission rate to change. Now why don't we move on to consumer fees. So consumer fees are another key revenue lever for us. You can think about this as the amalgamation of delivery and other consumer fees. So for example, the small order fee we have, the service fee we have, and alongside Plus subscription revenues that, that is all of consumer fees. Since we've started in 2013, these fees have grown at a CAGR of around 3%, but I would be the first person to admit that we have not always approached this in a very structured or scientific way. In fact, when I launched the business, I priced the delivery fee at GBP 2.50 and kind of kept it there for quite some time. And only recently have we begun the optimization process here. So I think we're quite at the beginning there. So ad revenue is a small part of our current model, but it's also a really, really big opportunity. So what this is, is, this is both the sponsored positioning product that we have for our restaurant partners today as well as advertising partnerships for FMCG companies on the Grocery side. We, today, have a product we launched relatively recently. It's a sponsored positioning product and also a sponsored carousel product. It scaled up really, really nicely, and we're really excited about it. This product itself, though, it is a proven opportunity already for online platforms. So for some online food delivery players, it's already 1% to 2% in GTV. And then for some players and other types of online marketplaces and Grocery, they're seeing a much higher number than that. Now we do see a lot of potential to grow this revenue channel, but it's also incredibly important to do this in the right way. Because consumers care a lot about the quality of the personalization and the merchandising and the app. And then you have restaurants and grocers and FMCG companies, who want to tell their story effectively, but also emotionally. And so I think the current product we have, that we launched recently, is scaling really, really well, and I'm really excited about it. But I do think it's pretty early in allowing merchants to communicate in that emotional manner, I'm talking about. Because if we do get this right, we can grow both this revenue stream and improve the consumer experience at the same time. And that is something I spend a lot of time thinking about. So if we consider this 250 to 300 basis points improvement that we just talked about, I do imagine a fairly significant proportion of that could come from ad revenue. So to summarize, I think the key opportunities in net revenue are, there's basket size, which we've only really started begun really thinking about, consumer fees, so optimizing on that and then thirdly, advertising revenue. And I think we're really early scratching the surface on all 3 of these. Now let's move on to the cost of sales. So cost of sales, which I would say is a few things. It's delivery costs, which is the biggest components. It's credit card fees and other direct costs. And we've consistently gained efficiencies in the past on this. Now delivery costs represent a vast majority of this line, so we'll just focus on that. I think some of you have heard me talk about RET before. This is Rider Experience Time. This is the amount of time it takes between accepting an order and then delivering it to the consumer. We see plenty of opportunities to reduce RET further, really by reducing rider's wait time at restaurant. That's the big key there. And that allows us to gain efficiency. It allows riders to benefit from taking on more orders and -- to increase the earnings, and also for consumers to get their food faster. I think there's also other operational improvements we can make. So for example, stacking more smartly and taking advantage of our density. And I think doing that without degrading the customer experience, that, of course, is critical. But these opportunities to drive efficiencies in the network are really, really big. I talked about that restaurant wait time. That restaurant wait time is still about 1/3 of RET. It's gone down a lot, but it's still a big number. So if you reduce that, that benefits all 3 sides of the marketplace. Riders get more orders, restaurants get their food out the door hot, customers get their food in a more timely manner and of course, it benefits Deliveroo as well. So if we think about net revenue and cost of sales, so that sort of aggregate gross profit number, we think it's realistic, we can deliver an improvement of 250 to 300 basis points -- 250 to 350 basis points. That basically gets us halfway to the target. So that you get there by basically taking up gross profit margins up to about 10% to 11%. Now we'll go into further detail soon, but we are either close or there already in markets such as the U.K. and UAE. And this is without the benefit of advertising revenue or any of those other changes that we've mentioned. The remainder of the 600 basis points, we believe, will come from marketing and overhead. So let's get into that now. All right. So let's talk about marketing and overheads. We're going to continue to make investments to support the growth of the business and technology will be the foremost investment. It forms a large part of our overheads investment today and will be the largest part going forward. Really, I think about this in 3 buckets. So how do we build technology to, a, get better at things that we're already doing. We're at the beginning of the journey, where there's a lot of opportunity ahead. So key examples on this are network efficiencies, so machine learning models that power that. The team -- the data scientists that work on personalization, for example. The product people and the designers working on our order tracking in the app. So those are examples where we've been working on this for quite some time, but there's still a lot of huge opportunity ahead of us. Second is a group of things we're working on longer-term innovations and initiatives. So these are things like Editions, things like Grocery, things like our Hop business or potentially nonfood. And then thirdly, we've got people working on the supporting infrastructure required for scaling the business effectively. So we grew 70% -- GTV 70% last year. Well, there's a lot of work that goes in the platform stability and continual improvement on forecasting models to make sure that we service the 3 sides of the marketplace as well. And clearly, the quality and effectiveness of the solutions, the product, the machinery that we build, that's a direct output of the quality and experience of the people that we hire and ultimately develop. And so on overheads, technology is going to be the key investment area. It is both foundational as well as the innovation in the business. And we will get leverage on the technology spend as a percentage of GTV, as the business continues to scale, but we will make these investments. I think on the rest of our overhead base, so really, I think, primarily nontech people. We expect significant efficiency improvements as the tech team builds tools to help our employees support the 3 sides of our marketplace, right? So we're really excited about that. And we have demonstrated this ability, right? We've gotten leverage on this line item consistently down over time. We brought overheads as a percentage of GTV down from 10.1% in '19, 5.7% in '20, and then 5.3% in '21, and we'll continue to target ourselves on that. Now let's go to marketing. So when it comes to marketing, we do believe we're still early in this industry when it comes to online penetration. It's quite low compared to other online categories. So we're building awareness at the top of the funnel for our offering across both restaurant and Grocery. This is really brand spend. And we measure the effectiveness of this brand spend via our econometric models. Now brand spend is one of those things that is just difficult to actually lead directly to specific consumers, specific orders. And so naturally, I think people can be skeptical of that spend. But we also know from our consumer surveys that one of the top reasons new consumers try Deliveroo is due to the visibility of our rider backpacks, right? So we're cautious about brand, but we also realize its importance, especially in these newer markets that may not have heard of us or even the industry that much. We also have proven digital channels. Now these are both on our platform and on major Internet platforms like Google and Facebook, right? So we call this growth marketing. So this comprises both your classic performance marketing as well as things like new user free trials and then winning back a customer on CRM. So we think about growth spend here, but doing that in a very high-quality way. So we're not just incentivizing a bunch of consumers that won't transact again, unless you keep giving them discounts, right? So there's a lot of work that's gone on that to making that much more efficient. So our focus on the growth marketing side, though, is, first, how do we continue to get more efficient at driving growth. And really, we do that through channels that are highly measurable that hit our internal return thresholds over a short period of time. And then we think about, secondly, are the things that we can be trying and experimenting with that ultimately can graduate into this proven growth bucket or lever over time? So we resource experiments each year to that effect. And then I think from a macro perspective, we expect some level of rationalization over time as the industry shakes out. Online food delivery becomes a higher portion of consumer food spend and the industry matures, and that split between new users and existing user spend shifts. So all of these factors, which is increasing online penetration food spend, us continuing to get more efficient on marketing spend, and then some rationalization and behavior across the industry, means we expect that balance of marketing as a percentage of GTV to decline over time. So between the marketing side and the overhead side, we see another 250 to 350 basis points improvement by '26, But we don't think that's the end. We definitely see further upside beyond '26 as well. So that covers off our long-term plan. So why don't we switch back and talk about medium term and our path to breakeven and Adam is going to cover this medium-term path. Thanks.

Thanks, Will. Using a similar slide to the long-term path to profitability, we'll now give you some color on how we get to breakeven in the medium term, which is what we see as the next key milestone on the path to achieving our longer-term profit ambitions. We saw a negative 2% adjusted EBITDA margin in the full year 2021. So what are the levers to making a 200 basis-point improvement, so that we reach breakeven at some point during the period of H2 2023 to H1 2024? We've shown our 2020 and 2021 numbers by half, because what you're missing when you see the full year numbers, particularly in those 2 years, is all the phasing of our investments and the movements from events like COVID. For 2022, we expect to cut that H2 2021 negative margin in half for the full year '22. And within that, we're going to see a sequential improvement from H2 last year into the first half of this year, and again from the first half of this year into H2 2022. Let me talk to you about not just the broader levers you saw of 2026, but the more specific levers that are going to drive improvements as we move towards breakeven. This year, we expect that the adjusted EBITDA margin improvements will come primarily from optimizing consumer fees across both Plus and pay-as-you-go, and from increased marketing efficiency. We're already seeing the positive P&L impact of both these drivers thus far in the year. Beyond this year, to breakeven, advertising revenue will become a bigger component of net revenue, and we will continue to optimize consumer fees. In the cost base, the investments that were in technology that we've already been making will drive continued network efficiencies as well as continued efficiencies on marketing overheads as well. Let me hand back over to Will now.

Okay. So we've talked about the levers. We've talked about what you need to believe for us to deliver on this guidance. And we're going to talk about a few proof points from a couple of our markets today. I think what these proof points show is that this isn't some sort of purely spreadsheet exercise, really, we feel that our aims are achievable, because we already have a track record of doing this. So we've got a couple of examples where we are breakeven or better on a fully allocated adjusted EBITDA basis, which means all country costs and all central costs are allocated. So if we start on the U.K. side, well, actually, before I get into that, let me just explain. So we've got the gross profit numbers, I think we talked about how we got -- what's above gross profit. What's in between gross profit and country adjusted EBITDA would be all marketing costs, all staff costs, anything that we can associate with the country that will be in there. And then below that, we allocate all central costs. So that would mostly be our people costs and non-people cost centrally. So engineers, nontech personnel, other types of services that we share. And finally, that gets you to the fully allocated segment or country adjusted EBITDA, so fully allocated. So why don't we start with the U.K. first. So what we did on unallocated central cost is that we based the allocation on the percentage of GTV. So for the U.K., that's 54% of our GTV. So we put in 54% of our unallocated central costs. It's just a pretty simple way of doing it. And so that what you can see is that the UKI was breakeven in '21 on a fully allocated basis, and then positive in 2020. And then the other market we wanted to call out was the UAE, which is one of our key international markets, and we've been positive at a country level, adjusted EBITDA fully allocated there for the last 2 years. So I want to show you one of our material segments, obviously, UKI and then one material International market. And I think what this shows is that we believe our aims are achievable, because we're already there in 2 of our major markets, and we're making progress on the others. I sometimes get questions asking whether we can only make money in large cities or even indeed just London. The answer is absolutely not. I'm going to show you here a number of examples of how we've got different levels of profitability across the spectrum of small, midsized and large cities. But first, before I get there, I think of profit pool potential as a function of population density, affluence, restaurant and grocery partner supply basically. And we build our local market share through relentless focus on CVP, neighborhood by neighborhood. And doing that is critical to realizing the profit pool potential. Now we also believe that the vast majority of neighborhoods in our markets have the fundamental characteristics to have decent to very good to excellent profitability. And improving and winning local market share positions ultimately yields outsized unit economics. And unit economics is the key to overall profitability. The unit economic side, in my view, at least, is, frankly, more difficult to do well than the marketing and overhead side. And so to illustrate the fact that we can be profitable in a number of different types of towns and cities, look at the slide. So what you see here is the whole spread of our gross profit margins city by city across U.K. and Ireland. I wish we could do this on a neighborhood level for you, but that would probably take too long. So these towns and cities, they are different sizes. There are different stages of maturity. And typically, the ones on the right-hand side are those where we are either at an earlier stage of maturity or we feel a need to focus on investing for growth. Now I'm not going to dive on all of these into detail, but you can see here that large cities like London and midsize cities like Brighton are indeed very profitable for us, but so too are other large cities like Sheffield or Liverpool. And we can also be highly profitable in smaller or midsized towns. So places like Sevenoaks or Harrogate. Those are really good examples where you have a moderate population density, but very good profitability. And then, not every market is great, right? So there's markets where there's room for improvement. You take a large city like Leeds to the North of England, we're super excited about the potential there. So we're investing to improve our proposition. And then finally, I do think it's important to note that not all markets are created equal, the profit potential for every market is different. And so we're always going to have a spread of different gross margins across our business. But we do aim to build the best CVP hyper-locally, thereby getting market share and then moving those less profitable markets up the curve over time. So we see a similar picture on the International side, where we've developed a really strong CVP, we've developed a really strong result in market share, places we've been in for a while. So you look at really -- you see really strong gross margins. So places like Dubai at 11%. Paris and Milan at 9%. And then we can also thrive in midsized population areas where you have still a decent population density with smaller cities such as Bruges in Belgium. And just like in the U.K. and Ireland, you have plenty of towns and cities in our International segment with room for further improvement. And it's even more the case in our International portfolio, because many are at an earlier stage of maturity that have consequently a higher level of investment and lower level of network density. But just like in the U.K. and Ireland, the overall margin level benefits from progressively moving them up along the curve. So I hope these last few slides have given a bit more granularity and real-world color to support the path to profitability we've laid out. And hopefully, it gives you a sense of how we think about the business, which is a very granular neighborhood by neighborhood, city by city view. So I'll now hand you back to Adam. He's going to walk you through our guidance, and then I'll wrap things up. Thanks.

Thanks, Will. So last thing for me today, I wanted to complete the picture on our guidance. We've talked at length about our path to profitability and our expectations for short, medium and long-term adjusted EBITDA margins. I did want to briefly explain why we've moved away from guiding to gross profit margin as a percentage of GTV to now give you adjusted EBITDA margin as a percentage of GTV. Simply put, having set out our path to profitability, it makes more sense to now guide on that basis, explicitly filling in the gap between gross profit margin percent and adjusted EBITDA margin. Now let's look at our GTV growth guidance. For 2022, we expect to deliver GTV growth in the range of 15% to 25% in constant currency. And within that, we will see a higher growth rate in the second half compared to the first half. This is because the toughest comps in 2022 will come versus Q1 2021, when many of our markets were in strict lockdowns compared to Q1 of this year, when our markets generally do not have any COVID-related restrictions in place. And in Q2 2021, when markets were starting to come out of COVID restrictions, but still had some form of restrictions in place for part of the quarter. This wider range of GTV guidance comes against a backdrop of quite a number of uncertainties. First, it is clear less than in Europe, all 3 sides of our marketplace will face headwinds this year due to inflationary pressures and an increasingly volatile geopolitical situation. Second, we acquired a lot of new consumers during COVID times, and although they were highly engaged through 2021, even after the removal of lockdown restrictions, we will need to monitor the behavior of these cohorts closely. Finally, we do expect new consumer acquisition to be more difficult and costly than during COVID times. Our 2022 guidance reflects our caution on these factors, but we are confident in our ability to adapt financially to a rapidly changing macroeconomic environment. For the medium term, we are maintaining our previous guidance for GTV growth in the range of 20% to 25% per annum in constant currency. And finally, you see on this slide, a recap of our path to profitability guidance, both in terms of timing to reach adjusted EBITDA breakeven, as well as our aim for 2026 adjusted EBITDA margins for completeness. Back to Will now to finish up.

So to conclude very briefly, it's been a really strong year for us. We grew really quickly across all of our markets, delivering GTV growth of 70%. We continue to execute on our strategy and scale our differentiated offerings with robust momentum in Plus, Editions, and Grocery. We exited the year in a strong financial position, which enables us to continue to execute on our strategy by investing to build long-term differentiation. And finally, we've outlined our path to profitability and look forward to giving you more detail and context on these plans over time. Again, I know this has been a really long call. It's been over an hour. So thank you again for bearing with us. There's a lot we wanted to get through, and we wanted to provide sufficient detail and we'll continue to provide more and more detail in the upcoming meetings. Finally, Happy St. Patrick's Day. And for those of you, who have made it to the end of this call, you should definitely have a well-earned pint of Guinness. So thank you all for listening. Look forward to the Q&A. Operator, over to you. Thanks.

[Operator Instructions] Our first question is from the line of Rob Joyce from Goldman Sachs.

I might try to squeeze 2 in, my apologies. I really appreciate the long-term detail you gave. And apologies if these are going to be more short-term questions, but I think there's been a fair bit of concern in the market about potential slowdowns, particularly in the first quarter. Consensus still has, I think, double-digit GTV growth for you guys in the first quarter of this year. Can you just confirm that's broadly in line with where you're tracking? Or give us a little bit more detail given we're quite far through the quarter? And then a very quick one on the cash side of things, maybe one for Adam, ex capital raise, your cash burn was about GBP 250 million in '21. Should we expect a similar level of cash burn leaving around GBP 1 billion on the balance sheet in '22?

Rob, I'll take the first one, and thanks for bearing with us on such a long call, but hopefully you found that useful. So I think on the quarterly side and sort of current trading side, what we expect, given -- we've got 2 weeks left in the quarter, we expect low single-digit sequential growth in GTV from Q4 to Q1, which would imply low double-digit growth year-on-year. So I think that's in line with kind of what you're saying. And we expect that to be the low point of the year. And I think the reason for that -- oh sorry, I guess, order growth will definitely come ahead of that GTV growth given the sort of AOV patterns we've seen over the last 12 months. But I think the point here is the comparison base is toughest in Q1, because if you looked at Q1 '21, that's really the peak of basket sizes and sort of the peak of COVID lockdowns. And then it gets sequentially easier quarter-by-quarter throughout the year. Now I guess that's a slightly different question than the sort of consumer behavior side. So we're watching consumer behavior very closely. Obviously, a lot of macroeconomic factors going on that are just happening very, very quickly. And I think we've reflected that caution in our guidance. I would also point out, it is not just about consumers, right? If you look at the other sides of our marketplace, restaurants, for example, in the U.K. are facing, that increases very soon, rising input costs, business rates as well amongst other things. So it's an area that, obviously, we're looking at very, very closely.

And Rob, let me take the second question, which I think was just around how should you guys think about cash cost, so anything below EBITDA. So maybe just worth kind of outlining what does sit below there and how we're thinking about that this year. So first, if you think about capitalized development costs, as we've said, we have been stepping up our investment in technology spend and kind of growing the technology team. We expect that to continue into 2022. So would expect capitalized development costs to be higher in 2022 versus 2021. And on the CapEx side, this is primarily Editions. So we talked about the growth in Editions kitchens in 2021 with that back weighted to the back half of the year, and we are continuing to invest in opening those Editions kitchens in 2022. And also, as Will talked about on Hop, that is something that we are still kind of opening Hop sites and waiting and seeing on that, but we are kind of being prudent and conservative and looking at making sure that we understand how that's performing, before we kind of go full bore on that.

Okay, Adam. So is there any -- overall where you broadly expect the cash balance to end the year?

Yes. We're not guiding on that specifically, Rob, but happy to kind of follow up with more detail later, if you would like.

Our next question is from the line of Andrew Ross from Barclays.

Can you hear me, okay?

Yes, I can hear you.

I'm going to also try and squeeze in 2, so I apologize. First one is on your 2026 model. Could you just give us a sense as to how that 4% EBITDA margin would then flow through to a kind of free cash flow margin? So just a sense as to how you see CapEx leases and other items in that kind of steady-state thinking? And then the second question, I see in the release, there have been a release of provisions of GBP 22 million, and it describes that as settlement reached for labor factors. Maybe one for Adam, but perhaps you could just talk us through what is that and an update on various legal debates?

Yes, Andrew, I'll take the first one, and Adam can take the second one. So I think it's a little hard to say, but here's the way I sort of think about it at a high level. So when I think about the bridge between EBITDA and free cash flow, I think that in the out-years, just the pace of hiring as well as the leverage we get from engineering will just be higher. So we would expect to see, in my view, a smaller bridge between EBITDA and free cash flow in future years. When it comes to Hop, we just -- we don't really know yet, right? We're trying to figure out if this product truly can make money within its 4 walls. And I expect sort of Editions kitchens, momentum to continue sort of as is. So overall, I would consider that the cash flow conversion from EBITDA to be -- that bridge to be smaller going forward rather than larger.

Yes, I think your second question, Andrew, is just on provisions and the change from the H1 number to the one that we released today. And so we don't kind of share details on any individual markets. That's just an aggregation of all the different things that we have going on at any point in time. But what you've seen is just as we've gotten more information, we're able to release provisions over time. And I think that any point, we have a number of different kind of investigations and challenges ongoing, and the provisions just reflect our latest view as to what's probable at any point in time.

Your next question is from the line of Andrew Gwynn from BNP Paribas.

One question then. Given obviously what the share price has done, is there a pressure on the business to deliver profitability even quicker than you're outlining today? That's it.

Well, thanks for the question. I think -- it's really what we think is the right thing to do for the business, right, Given the maturity of our business, given the industry is maturing, we obviously are extremely well capitalized and choose to invest a lot more aggressively, if we see. But ultimately, this is the path that we think is best for the business. And what we laid out today isn't just for investors, but it's also for our company, right, and making sure that we guide ourselves to the path to adjust -- to EBITDA profitability as well as free cash flow generation. So we're really excited about it. We view this as a very positive motivation for people internally as well.

The next question is from the line of Georgios Pilakoutas from Numis.

You've touched on quite a few markets today. I haven't heard so much on Singapore and Australia. I guess, just interested to hear, how you think about prioritizing deployment of capital and assets and just kind of general thoughts on what are the key factors that determine whether you're investing more or less into a given market?

Yes. I mean we obviously have a different blend of investments across, I'd say the U.K. and international markets. I don't think we're going to go into sort of country-specific levels. But I think the overall philosophy for us is, if we can get to a #1 position or a strong #2 position, we'll continue to invest. And keep in mind, our -- sort of the way we think about the world is on a hyper-local neighborhood basis. So it's really the amalgamation of all those local market shares, getting us to a point of that #1 or strong #2 position. And in our markets, we currently feel that we can do that. Now if that changes, as you've seen in the past, we don't hesitate to shed markets. We've done that in markets such as Spain and Taiwan, where it just became too difficult for us. So this is something we watch closely on the capital allocation side. And then there are -- certainly true that there are some markets we invest in more versus others, based on what we see the opportunities and risks to be. Operator?

Georgios, does that complete your question?

Yes.

So our next question is from the line of William Woods from Bernstein.

Just a quick question in terms of your midterm guidance, I suppose building on the last question. Does it imply either to meet your GTV growth that you need to enter into any new markets or to meet your profitability guidance, do you need to exit some of those markets? Are any of those things baked into your plan there?

No, neither of those things are baked into the plan here.

Next question is from the line of Navina Rajan from Morgan Stanley.

Just a couple for me. On the 4% Plus margins by 2026, can you just give us some color on what's implied for the U.K. and International and for the U.K., how you're thinking about your market share by that point? And then just on Editions, I think if you can just sort of give me an idea if I'm thinking about it in the right way and sort of my back of the envelope map. So if your PPE increased by around GBP 60 million and you rolled out about 100 million Editions kitchens this year, is that around GBP 150,000, GBP 160,000 spend per kitchen? And does that sort of compare in a similar ballpark to the cost for Hop?

Adam, why don't you take the first one? I can take the second one.

Yes. So thank you for the question. I think on the 2026 guidance. So I think the question, just so I understand it is implied in the 4% Plus group margin, how does that break down across our segments and how do we think about U.K. market share as part of that as well. So look, I think the way we think about this is clearly, there's room for improvement in both segments. I think we've showed you that the UKI segment is further ahead on both gross profit margin, but also overall kind of fully allocated adjusted EBITDA as a metric. But the international markets have definitely have room to grow, and we are kind of modeling growth in both segments as we get to 2026. I think in terms of market share in 2026 in the U.K., that's not something I think we're kind of talking about at this point in time. I think 5 years out, but I think we are kind of projecting continued improvement...

I think it's also important to say like the market share for us, I mean, at the end of the day, the market share we really care about is profits and free cash flow, right? And so we're going to take actions to maximize that profit share. Now when you look at GTV market share, obviously, that's important, and we've been scaling that -- I think we've gained a lot of market share in '21 on that basis. But I do think that not all market share is created equal. There is good market share, and there is definitely bad market share, and we're going to be going after the good stuff to drive our ultimate share of profitability and free cash flow in each neighborhood, in each city and in each market overall. And just to answer the question on Editions. I think your math seems generally right. I think the GBP 150,000 number is probably high, but I don't think it's that far off. And so I think that sounds pretty reasonable to me. Hop is different, though. So let me just explain the model slightly differently. So Hop the -- CapEx is actually significantly lower, but because we're running the warehouse, there are OpEx costs, right? And so Editions is a bit different, because restaurants actually bring their own staff and they're responsible for everything, whereas in Hop, we have a wholesale relationship with the grocer, but we manage all elements of that. And so the way I sort of think about Editions is, right away, you start generating cash flow. I think, on Hop, you need to build up to a certain order volume level to get that operating leverage from your OpEx base, but which are the people sort of picking and packing in the micro fulfillment center, but the CapEx costs are certainly below that of Editions.

The next question is from the line of Adrien de Saint Hilaire from Bank of America.

First question, if I may. So can you give us your thoughts on consolidation in the industry and what role Deliveroo wants to play? And secondly, if I may, can you discuss a bit competition, in which part of the world, is it getting more rational in your view? And in which part of the world is it getting worse?

Yes. Let me take that one. I think on the consolidation side, yes, certainly, we've seen a number of things happen in sort of the last 6 months. I think we've looked at some opportunities in the dark store space. Obviously, there's been some platforms that have sort of merged with others. My view on this is, I'm just hyper-focused on building the best consumer value prop neighborhood by neighborhood. And I think doing that with healthy financial returns, ultimately will take us into a good position, no matter what that is, right? I think that I do expect to see more consolidation over time in this space. I do think that makes sense. It is a rational thing. But I think for us, we're just focused on the 3 sides of our marketplace. And I think if we do that well, we're going to succeed no matter what. In terms of competition, you're asking sort of where are we seeing more rational or less rational behavior. I'm trying to think. So I think the '21 let's just take the U.K., for example. There was definitely a lot of, sort of, what you call it, sort of free delivery campaigns by some competitors. And I would say not highly targeted in that manner. But especially for sort of like fast food restaurants, we saw a lot of that. But that seemingly -- we've seen a lessening of that type of spend, and I would say that's pretty much across the markets that we've seen. There's been less of the sort of just all out sort of discounting. Now competition, of course, remains strong. There's no question about it. This is a competitive industry with a giant TAM. There's a reason for that competition. But overall, the sort of like really blatant, just buying orders or market share type stuff, I think that's sort of subsided quite a bit from what I can see.

The next question is from the line of Sarah Simon from Saint Berenburg (sic) [ Joh. Berenberg ].

I just got one question, which was, you talked well about the gross margin being impacted by basically adding additional content. And I think you clarified that as kind of more choice. Why does that impact gross margin?

Sorry, I'm not sure I fully understand the question. Do you -- are you saying does more selection add more gross margin? Or am I...

Well you said that one of the reasons for the pressure on gross margin in the second half was adding more content...

Right. I got you.

I mean more choice. And I'm just wondering why more choice is margin negative because...

Yes, yes, yes. Well, there are a few reasons for that, right? One is we added a lot of grocers, that's been a big area of investment for us. And secondly, we added a lot of exclusive content on the platform. So it's really those 2 things that I think -- sorry, sorry, I misunderstood your question, my bad. That's really what we're talking about.

So -- but when -- so Grocery, obviously, you've said the take rate is lower, but -- and so should we assume, therefore, that the gross margin is also lower?

Well, I think we talked about at the H1 that in the U.K., restaurant gross profit per order was [ 240 ], and Grocery was [ 210 ]. So it is indeed lower, but it is 100% incremental. And drives just higher lifetime value of the customer, because they're really sort of different occasions. I don't think people sort of say, "Hey, I want Nando's. Now you know what, I'm going to like cook a fish", right? So I think that's kind of like what it is.

Okay. So basically, in summary, it's really the mix shift between restaurants and grocery is the primary thing you're talking about there?

Yes, it's an interesting because the gross profit margin might take a slight hit, but then the overall lifetime value goes up, because it's the same customers buying it. It's also the same riders delivering it, right? So you drive more density into the delivery network as well.

Our final question is from the line of Marcus Diebel from JPM.

Well, just one last question left. I mean I appreciate the slides, also on cost. But in terms of inflation -- on rider inflation and could you just elaborate a little bit more in detail what it means on financial impact? I mean, it looks like you're in a very good position. But do you expect still rider cost to materially go up from current levels? And if so, is that something you just compensate with higher underlying costs for meal and you're taking a cut on this? Or is there a chance that we can expect or should expect a margin squeeze or more losses on that topic?

Well, I guess I'd say a few things, because really, there's a few questions. One sort of around wage inflation and the second one is probably around fuel costs, right? I do think there they're probably a little bit different. We have -- so we've been out of COVID in the U.K. for about a year, and you saw the slide where we showed our rider retention and sort of overall rider fleet versus job vacancies and what is the tightest labor market we've seen in a long time. We have not seen any sort of material adverse cost side from our rider fleet. So we haven't seen that happen to us. Now obviously, this is something that we have been cautious around for since, I'd say, a year ago. So we're always monitoring those, but we have not seen sort of any adverse cost impact. Now when it comes to fuel, it's actually pretty interesting. I get this question a lot. And I think one thing to remember is that 90% of our deliveries around the world are actually conducted on 2-wheeled vehicles with the majority of those being bicycles and electric bicycles. And so clearly, if you have a petrol scooter, you're going to feel a bit of a hit. And so we have launched a partnership in the U.K. We launched a partnership with Shell, which we didn't announce. We did it about 2 weeks ago, which allows riders to get a discount, if they purchase petrol at Shell stations. And it's just an area that we are clearly looking at very, very carefully. When it comes to sort of the inflationary environment, when it comes to food, it sort of is what it is. The grocers and the restaurants price what they do, and that's reflected on our marketplace. And so I think that part is -- it really depends on what -- how grocers and restaurants want to respond to input costs and labor costs.

Okay. I was more asking about the other bits, because DoorDash especially, they were highlighting obviously an impact on fuel, but obviously, their share of is probably much higher.

This concludes our question-and-answer session. I'd like to turn the conference back over to Will Shu for closing remarks.

Yes. I just want to thank everyone for bearing with Adam and I on a very, very long presentation. I don't think either of us actually enjoyed talking that much. But hopefully, it was it was useful for you all, and we wanted to be transparent on a lot of different details. So I want to -- we look forward to meeting with some of you over the coming days and weeks. I think we are doing that. And with that, I want to wish you all a good day and a Happy St. Patrick's Day. So thank you, operator.