Ocado Group Plc

Very good. Good morning, everyone. Welcome to Ocado's first half results. Tim and Stephen are going to lead you through the operational and the financial milestones of the period, and then we'll go on shortly thereafter to talk about the progress that we've made in improving cash flow and growing profitability. not only in technology solutions but in retail as well, and then how we've been improving in the rolling out of CFCs around the world. What we're pleased to be able to say, again, is that online is the fastest-growing grocery retail channel. And, you know, as Ocado, as you well know, we have... Proven technology, constant innovation that we can feed into clients, those clients looking to find solutions for growing e-commerce. As a board, what we've been particularly encouraged to see over the past half year and before that is the work we are doing to help our clients manage that tough but crucial transition to being a growing, profitable, multi-channel retailer. And that transition actually requires incredibly patient and close collaboration. And as you all know, we've taken some very significant organizational and process steps that are absolutely essential to help manage that transition. As a result, we actually have a very high conviction that it's going to pay off, that we will see accelerating growth and profitability over time. Meanwhile, I think from a Cardo standpoint, we still bring the cash and the cost discipline, and you combine that along with the economics of the CFCs that are in existence, we have a growing confidence of a clear trajectory to cash flow break-even in the second half of 2026. And with that, Tim, I'm going to hand over to you to take us through the results.

Thanks, Rick. Okay, so let's move on to the highlights. Clicker, here we go. Thank you. And right, so financial progress, strong. Group revenues up 13%. EBITDA nearly tripled up at 71 million, up 55 million year on year. Strong revenue and EBITDA growth has come across all the divisions. Strong cost and capital discipline in the group. underlying cash flow up 101 million in the first half. With guidance raised, we had guided to 100 million for the year. We've now increased that guidance to 150 million. That is following from 356 million, if my memory serves me right, last year. So 500 million improvement over financial years 23 and 24. Um, that is on the back of operational progress to technology solutions, uh, driven by the resilient recurring revenue growth and increasing cost efficiently cost efficiency. So obviously the cost efficiency is improving at a faster rate than we had, uh, suggested. Um, the three CFCs we have going live in the second half of this year, um, And our build for McKesson in Mark's area is also on track. And Ocado Retail has resumed its mantle as the UK's fastest growing grocer and building back its EBITDA. We're making strategic progress. Our partners' success focus. We're doing work with a lot of our clients. We've put investment into resources. We've put resources in country. Our reimagined technologies are successfully starting their early deployment. I can see a few faces in the room that have seen the robots picking in Luton. Our OIA pipeline, Mark will talk more about, is growing strongly, and as are ongoing discussions with potential future OSP partners. So overall, a strong first half. At this point, I'll hand over to Stephen, who's going to run us through the numbers, and I'll come back and talk after that.

Thank you, Tim. And morning, everybody. So thank you for joining us for today's first half results for Ocado. So it's been an encouraging first half, I think. Revenue grew by 13%, group revenue, up to £1.5 billion. Group adjusted EBITDA trebled from £16 to £71 million. So very good progress there, more than trebled in fact. Underlying cash flow improved by £101 million. Tim just referenced that, already slightly ahead of our guidance on a full year basis. And we're raising that on the full year guidance. And then liquidity remains strong. We've got £747 million of cash in the bank and an undrawn accessible revolving credit facility of £300 million. The growth stream by tech solutions up 22% and Ocado retail up 11%. An improving trend in Ocado retail Q2 versus Q1. Important narrative and we expect that to continue in the second half. We have three resilient operating models. in technology solutions, contracted modules, visibility around module rollout, and a high margin business model, as we'll see shortly with tech solutions. Similarly with Ocado Retail, over a million active customers now. shopping regularly with Ocado, and growing customer numbers at the same time. And then we have Ocado Logistics, which is a cost-pass-through business, relatively modest EBITDA, but reliable, consistent EBITDA, and generator of cash flows for the group. So there's the £101 million underlying cash flow improvement, driven by EBITDA growth, capex reductions, that's unit capex reductions, not just capex itself and less activity, and cost control, as you'll see in the numbers. That's how it all adds up to the profit before tax number. After adjusting items, £154 million. It is less than half that it was this first half last year. So whilst it's a loss, it's an improving trend. We do get asked a lot about this by the media, and we do expect to turn EBT positive sometime within the next four or five years or so. Next slide, technology solutions. I'll start with average live modules. An important number, it's live modules that drive a majority of the revenue in that revenue line. Around 84% of that £240 million first half revenue is ongoing recurring revenue. That's linked to the average number of live modules, that 112 live modules that you see there. We're going to see a chart shortly that shows the improving revenue and contribution per module as well. Those two will be key drivers on our route to cash flow positive. The contribution is affected with the gross margin from that revenue. 71% a healthy and improving trend on gross margin as well, on contribution. We'll see that shortly. The P6 exit rate of the direct operating costs for the revenue of technology solutions is improving quarter versus quarter. Contribution margin. Technology costs and support costs are in line with the previous half, and that's after digesting the investment in partner success of around £7 million in that space to drive the success of our partners. strong growth in EBITDA, now at a 15% EBITDA margin. I'm on the back of the visibility that we have around the second six months and the expected progress we expect to make on cost management. We're raising our guidance today on a full year basis from previously over 10% to mid-teens. Growth in recurring revenue. So this chart on the left here shows that recurring revenue, that 84% of total revenues, and how it has grown from half to half over each of the last three years. Driven by the module numbers that we see there. And we now have 112 live modules at the end of the first half. You can see the UK international split. Then the annualised recurring revenue per module, that's growing as well. It's growing due to the indexation that's part of the contracts that we have with our partners around adjusting the fees for local inflation at the start of each year. It will also grow as we roll out reimagined technologies to our partners. We expect to be announcing the second half of this year orders for reimagined products by our international partners. That will lead to incremental annualised recurring revenue. That will be another reason for that number to grow even further. So watch out for that in the second half. We've got good visibility on the module rollout to fiscal 26 as well. This is another part of the attractions of this business model. We have good visibility around the sites that are going live in the current year, and we expect to go live over the next two years. The right-hand side summarises that. So the second half of this year, we've got Sydney and Melbourne for Coles, and we've got Madrid for El Campo. We expect those to go live in the second half of this year. We also expect some further incremental drawdowns in the lapping of the second half 23 module go-lives to lead to that higher number of at least 120 live modules. by the end of this year. So those two factors there in the top right-hand corner, adding up to in excess of 20 modules on a four-year basis to get us to that, sorry, over 10 modules to get us to that 120 number by the end of the year. Going into the next two years, five CFCs, going live over 25 and 26. Phoenix, Charlotte and Warsaw, most likely in 25. Tokyo and Busan in 26. Cost control in this business remains key, getting our fixed costs down. It's something that Tim reminds me about all the time, particularly in support costs. We're doing pretty well here. First of all, though, let me take you to the variable cost line, direct operating costs. Now, you can see that 1.56% in the far right of that table there on the left-hand side. An improving trend. We expect to improve that even further. What is this cost? It's the local engineering cost to maintain the grid and the bots internationally. And it is also the cloud computing cost as well. That's a rough mix around three quarters, one quarter in terms of the total cost. In fact, it's in the cloud computing cost that we've taken most cost out over the first half. I expect, though, and the visibility that we have of it, due to the reliability of our MHE, that the engineering cost, there's an opportunity there for that to come down considerably. And that's part of the reasons, I think, for the guidance that we're giving around the improving margin in tech solutions. The technology cash costs. This is the total cash that we spend on technology in the half, £145 million, capitalising £98 million of that, expensing the rest through the P&L account. Those costs are broadly stable, but as we've guided previously over the next couple of years, those costs will start to wind down as reimagined technologies get completed and ready for market and rolled out. Support P&L costs, that's the classic finance, human resources, legal, group IT, and so on, those costs. We've had good progress, as you can see, taking those costs down over the last two or three years. More is to come. But at the same time, those numbers accommodate the investments in the partner-facing teams that I just talked about. Kydo Logistics. I won't dwell too much on this business. It has Ocado Retail and Morrisons as its two customers, serving those in the UK, a reliable generator of EBITDA for us, and the cash flows as well. Costs are well controlled here. Fulfillment and delivery costs are up 3%, whereas orders were up 7.8% across those two customers. That's a really healthy dynamic. The role of this company is to bring its unit cost down as a measure of volume. That's what this company does because that means more money and profits for Ocado Retail and Morrisons. So on that note, this is the important chart here. The logistics measure itself against probably all measures. This is the one that looks at very, very closely. Units per hour. How many shopping items is it picking in a labour hour? You can see the trend, 170 through to 221 in the first half of this year. In Luton CFC, which we opened recently, that delivered 250 units per hour in the last couple of weeks and expects to improve from there as well. What a higher units per hour means is lower labour cost. Going back to Tim's earlier comments. The drops per eight-hour shift is slightly lower at 21%. That's largely a consequence of the Hatfield closure and moving those orders to Luton. So a slight increase in the stem times between the warehouse and the spoke for delivery to the customer. Ocado Retail, healthy revenue growth, 11%. Good EBITDA flow-through and an improving revenue trend. Nothing other to highlight in those bullets beyond those comments I've just made. Word on inflation, the average selling price increased by 1.5% in Ocado Retail, in the half. That compares with the 4.4% Nielsen quoted grocery price inflation. So well below the price of inflation. That's this investment in value point to grow the customer numbers, which are happening very nicely, as you can see, growing by 8%. KPIs have improved and stabilised here in the shopping items per basket. See where we were at the COVID peak. Active customers, just reference that, growing nicely. This is a key point. You recall the low EBITDA numbers, or in fact negative territory that Ocado Retail was in, as it was operating with excess surplus capacity, now at 80%. and heading towards 100% usage of capacity. The key point there, though, is that the revenue growth for that business, where we have excess capacity, has a very high EBITDA margin flow-through. It flows through at about 19% or so to EBITDA. Ricardo Group, our cash flow. So in the first half, our cash outflows reduced from £320 million to £138 million. That includes the auto store settlement, but there is a £101 million underlying improvement, as we'll see shortly, in those cash flows driven by EBITDA and CAPEX. A word on our maturities. Three maturities on the horizon. I'm sure I'll be asked about this in the Q&A. We plan to address those maturities December 25, October 26, January 27. We do not expect to, or we're not planning to, allow those to go current. So we'll do our maturities well in advance. We've got a very clear plan. We're confident in delivering that plan. So, underlying cash flow continues to improve. That shows the trajectory of the last few halves. £197 million in the first half of this year. That's an improvement of 101 versus the 298 of the same half of the period before. A word on underlying cash flows, just as a reminder. It is cash flows basically after all capex, EBITDA, all other costs, etc., excluding, though, either the positives or the negatives of adjusting items. In old language, exceptional items. We are improving our guidance from £100 million improvement to £150 million improvement. That's on the back of the EBITDA growth. There is some capex in there as well, some capex reductions. We're using more capital that's existing capital on our balance sheet to fund certain capex projects. You may know that on our balance sheet we do carry a constructions in progress amount. We can draw down on that and deploy that in future go lives. And we're doing that quite successfully. The word here, we're on track to turn cash flow positive during fiscal 26. Tim referenced it, but we have very good visibility now that during the second half of 26... we expect to turn cash flow positive. This next slide takes you through the key building blocks to deliver that outcome. Number one is the growing contribution in technology solutions and the most important driver of that deliverable of turning cash flow positive. growing recurring fees from our clients as live modules increase, and good visibility, as you saw from that earlier chart, around the CFCs that are going live and the modules that we expect to be added to the live modules that we have today. The revenue per module that I talked about, reimagined will drive that, plus the indexation factor, and the lower OSP direct operating costs. That's a continuation of that trend that I showed on that chart a little earlier. robust cost control and efficiencies, keeping costs flat today. The support costs we have reduced, but we've guided previously on this, technology investment costs will reduce as the reimagined products are rolled out naturally. We're guiding to £240 million in fiscal 26. That's the number we've consistently guided at, and we're holding that number. At the same time, Ocado Retail is expected to return to cash flow positive territory in fiscal 26 as it delivers high mid-single EBITDA margins and as we reach that improved CFC utilisation number that I talked about earlier. Logistics will continue to generate cash for us. And finally, we will get some new cash flows from Ricardo Intelligent Automation. Mark's going to talk about that shortly. But just in case you're wondering whether all the sort of heroics is in this bottom right-hand corner, that's far from the case. It's going to be a low tens of millions of pounds type number from that business in fiscal 26. Guidance, kept guidance pretty much where we put it six months ago. Three areas, though, where we're highlighting or improving guidance or raising guidance. Cash flows we've talked about, now £150 million improvement. Technology solutions I've referenced, the mid-team margin from previously greater than 10%, and capital expenditures, £425 million, and previously that was £475 million. That's it from me. I'm now going to hand back to Tim for the strategic and operational review. Over to you, Tim.

Thanks, Stephen. Okay, so I think the first point that we want to highlight really is that the shift to online grocery is resumed. And so what we're just highlighting here is both our clients and some of their competitors' total like-for-like growths and their online like-for-like growths. You can see online being the driving force in many grocery retailers around the world's growth at the moment. And also here, we can just continue to see that as a channel, it's forecast to continue growing And so we've got a positive view of future opportunity for the market. Let's just look at our own numbers for a moment then. So here we're talking about modules. So we've shown this graph before, just wanted to update you on where we are. We'll end financial year 24 with over 120 modules live. In those warehouses that are already live, we've got another, we have previously stated 55 modules from their design capacity. What I'm just highlighting is it's now over 70 modules. So we've increased the design capacity on a number of existing live warehouses. So we've got over 190 modules of capacity in the live warehouses. We've got 25 modules plus of capacity in the ones under construction and 40 in the announced sites that currently aren't under construction. Going live this year, we've got Madrid, we've got Sydney and Melbourne, all to happen in the second half. We've obviously confirmed the order for Eon 3, which is a very large warehouse, significantly larger than an average warehouse. We've paused, as you know, Sobeys 4. That was actually a smaller warehouse earlier. But we expect to get back to that in the coming period. And we've obviously got live discussions, or we do have live discussions with a number of potential partners about future warehouses in new markets. This 55 to 70 is both important for us and it's important for our clients because it helps to improve everybody's asset turn, helps improve the amount of revenue they're going to get from the fixed cost investments that they put into those sites and on an ongoing basis to leverage the rent rates and fixed costs over larger volumes that those sites are now capable of. And in the case of our client in the UK, it will provide a critical growth avenue as they fill up and with their good growth that we're seeing at the moment as they fill up their existing site. So they'll be able to move that utilization number down from 80 in terms of reflecting the potential capacity in those sites. Our partner success is beginning to help drive utilization and we are starting to see incremental drawdown of capacity at live CFCs. So partner success is key for us. What are the objectives of it? To help drive the growth of our clients, their sales and their shopper numbers behind those sales, as well as to improve our partners' operating metrics, both in their warehouses, in their supply chains and in delivery, and to help those partners win in the online channel. We're leveraging group skills and resources, people that have been in this space for 20 plus years. We're putting people in country and dedicated individual teams to each client. We're reassigning a lot of our best talent to deliver that expertise and execution abilities across our network. We're seeing encouraging progress. What's really encouraging is how closely our partners today want to work with us on this. We've obviously expanded the program this year, appointing John Martin as the new CEO of Ocado Solutions and bringing his incredible depth of experience and previous success to that area. And we've got no structural impediments to achieving the targeted economics. We're not seeing anywhere something in a market that says, oh, this market won't work. That's just not the case. There's a few comments on the slide from a couple of our partners around their growth. But I think I've also got a short video from one of our partners who's going to say hello this morning. Oh, before we get there, can we just play the video?

Hello, I'm Rodney McMullin, CEO of Kroger. I'm delighted to join you today to talk about the progress we are making with Ocado, who is an important collaborator in our seamless strategy. Together, we are bringing our customers an amazing delivery service. The Ocado model is built on proven technology and is one way we deliver a customer experience with zero compromise on value, selection, and convenience. In the three years since our first CFC went live in Monroe, Ohio, our work with Ocado continues to evolve. Today, we have eight CFCs serving customers across the country. Our customers love the service, which stands out from other delivery offerings, as demonstrated by the outstanding MPS scores we continue to receive. We are also seeing strong delivery growth from CFCs, where we recently announced that sales had almost doubled in the last quarter over the same quarter in the previous year. We are pleased with the progress happening at our Monroe site and appreciate Mark Bentley's contributions. And we are excited about the talent both companies are bringing to make further improvements across our network. We continue to learn and improve at our sites. A great example of this was our decision in the first quarter to close three Spoke locations to reallocate that capacity closer to our automated fulfillment centers where we have higher customer density and better order-level economics. Spokes are a useful tool, giving us the flexibility in how we serve customers from the CFCs and grow volumes most efficiently. We plan to open a new spoke in Florida later this year that is substantially closer to the CFC than the initial spoke location that we closed. As the online grocery channel continues to accelerate, we remain optimistic about our continued close collaboration with Ocado and the opportunities ahead. Thank you.

So talk a little bit about partner success. What is it that we're trying to do? So as I said, the first part is trying to help drive our partners' growth. There's a few parts here. We're working to help our partners learn best practice based on the experience that we've had in the UK over the last 20 years in terms of how to acquire and retain digital customers and to do so cheaply or cost efficiently and to drive them through what we call the nursery journey till they become loyal shoppers. We're working with our partners in terms of talk about their ranging and webshop functionality to help drive online basket size. Range obviously is a key driver of basket and margin and the OSP facilities allow you to carry an unparalleled range in size at the most cost efficient way and without affecting the cycle time it takes to pick an order. So we really want our partners to take advantage of that range. It adds margin, it improves acquisition, retention and frequency of customers shopping. It also adds media opportunity as well as basket size, which obviously leverages the cost of the delivery. Just a little chart here where we actually had specific data points, but I know how clever you all are and that you'd all try and deduce which client each one of them was. So we turned it instead into a best fit line between the six clients just to show you how having a larger range drives basket size order growth from the first to the fifth order. So across those six clients, so we have this data. There you are. As the basket size grows, the order volume grows across those clients. across those orders. And obviously, this is the kind of data we can now share between the clients and show them how others who are leveraging the capability that OSP is bringing are driving that growth. Not only it's a growth in revenue, it's a growth in basket size, which, as you know, converts very well. We've got coming up Swift Router, but we've actually got Swift Router coming in stages and we've started running with two of our partners trials on some of the earlier parts of Swift Router. We're not yet able to do all of the functionality Swift Router needs to do, but we've got some very encouraging trials going on where we're seeing very positive learnings from short lead time orders with a significant and strong sell-through of those shortly available slots. And our clients are seeing that those are significantly coming from new customers as well as from existing customers taking advantage of those earlier slots. Right. Another place obviously we're working on is in helping our partners to reduce their operating costs and their operating metrics, giving them real-time views of key operating metrics and helping them to optimize those settings, et cetera, to drive results. You've got some ideas here on the chart. This is waste, for example, bringing waste down in a partner's site, bringing inbound productivity up in a partner's site. We've seen partners improve their productivity by over 50% overall since we started these programs in some sites. We're helping them on the supply chain with our supply chain software and also how to use it best, enhance demand and inventory projections, manage both customer availability, so have higher availability of products to sell to customers, as well as improving wastage. And we've got a lot of resources going on to improve last mile efficiency. You might have noticed that the last mile efficiency in Ocado Retail had not gone up. Just worth reminding ourselves that until we switch Ocado Retail over in the first half of next year onto OSP, that that part of their business is still running on the old systems. They can't take advantage of any of the enhancements we're doing in delivery operations until they move on to that platform. Whereas obviously the warehouses are all on OSP and you've seen a significant uplift in their productivity in the warehouses. We're also helping our partners in terms of ease of use. So for those of you who have been in our warehouses, you can see that actually being an operator on the frontline, it's incredibly easy. The systems, the screens, the process is very, very well engineered to make it incredibly easy. We are now taking that up through the chain of command to make it easier to run these warehouses and manage the flows inside these warehouses and make it as easy as it is for the frontline operators. I'm going to hand over to Mark.

Thanks, Tim. It doesn't happen very often. Hi, everybody. So you'll recall we set up OIA, as I'm going to refer to it from now on because otherwise it's quite a tongue twister, in order to take our existing technology, hardware and software, and sell it into the substantial market outside of grocery e-commerce. Our grocery background means that we come complete with a very capable logistics platform. That's because groceries are a particularly hard use case, very complex, and that means that what we have to sell now is applicable for a very wide range of use cases and markets. At the same time that we launch this, the market for logistics automation is growing very strongly. Oh, sorry, someone's telling me I need to click. Thank you very much. The market for logistics automation is growing very strongly, partly because warehouse labor costs are rising inexorably, but actually, possibly more importantly, because that warehouse labor is increasingly electing to earn its living doing something else completely. And that means that an investment in automation for many companies is their only option, actually, if they want to continue to grow, because otherwise they just can't hire enough people to run their networks at full capacity. We are in an era where investment in automation is as much about ensuring continued growth for your company as it is about reducing OPEX. Now, in that market, there's a fantastic opportunity for Ocado Intelligent Automation to sell its existing technology in a capital-light model, which realizes its margin at the point of go-live. The solution itself brings all the same benefits that apply in grocery. But in the wider market, some of those benefits and advantages are actually amplified. So, for instance, where high-value goods are involved, then it's particularly important that there is good physical security. Also, in the pharma industry, accuracy and physical security become even more important than they are in grocery. The traditional strengths of dense storage and very high productivity, of course, apply absolutely everywhere. And then whenever ultra-high throughput is a factor, Ocado has a particularly strong story to tell. But it's not all about the hardware and the software. One of our early surprises in the market was learning just how attracted potential customers are to transacting with a company who not only create and manufacture automation but also operate it. automated warehouses on a large scale and can demonstrate a very deep knowledge of how to get the best from automation and how to optimize the outcomes for a facility that includes an OSRS. Our first customer project for McKesson in Canada is on track, and Ocado is due on site in that building at the end of the summer. And there we will install our own equipment and software, but also we're going to integrate a little bit of third-party equipment, and we are deploying some pharma-specific functionality, which means we're taking a solution which is already a very good fit for this market and then sharpening its appeal in this vertical. We began marketing, believe it or not, we only began marketing OAA at the beginning of this year with the launch of a website, social media campaign, some direct marketing, but most importantly, attendance at trade shows. And at those trade shows, we took our latest generation of grid and robots. We generated a huge amount of demand or a huge amount of interest from potential customers and from integrators. And we detect in the market a real hunger for an alternative cubic ASRS solution. Now, from the hundreds of leads that we generated at the trade shows and in the other marketing, we're now building a substantial pipeline of future potential projects. And in the shorter term, we are engaged with multiple late-stage project bids right now. Another lesson from the year gone by, actually, is how well we play at larger scale. So it doesn't mean that we won't do small projects, but when the throughputs rise and the scale of storage rises, Ocado has a particularly strong story to tell. So what about our target verticals? You might not be surprised to learn that we've had significant unsolicited interest from the pharmaceutical distribution and healthcare industries. But in addition to healthcare, we also think we play particularly strongly in apparel, CPG, and third-party logistics. And in that last vertical, third-party logistics, we already have many successful contracts with 3PLs because they're customers of the Chuck assisted picking AMR product that was developed by Six River Systems and which we incorporated into OIA's product set when we acquired Six River Systems one year ago. Another exciting target for us is is the world of large-scale moving cases around as part of an internal supply chain. So we think that with some small additions to our product set, we can play quite well in that space. Our solution already handles cases. We can already ship in the same consignment a mixture of cases and individually picked items. But by taking the robotic picking technology and extending it to be able to pick up cases and auto-load cases from a pallet into and out of our storage grid, and then to take the Chuck AMR technology and extend that to be able to automate the movement of pallets, we think we have a very compelling offer into the case handling and internal supply chain market. So where are we focused right now? So we are growing appropriately, we think, scaling our business appropriately. We are leveraging our existing technology. We are targeting specific verticals where we think we play hard. And we are exploring an extension into a very large adjacent vertical with some small additions to our technology product set. Also in the year ahead, marketing activity will increase. And in particular, we will attend more trade shows and we will have a bigger presence at those trade shows. And that includes, for the first time, attending Logimat in March of next year, which is the largest trade show in Europe. I hope maybe I can see some of you there. And, of course, we are focused on delivering for McKesson. So we're going to deliver that project on time, and we are chasing down those late-stage project prospects that I talked about earlier. Thank you very much. I'm going to hand back to Tim.

Thanks, Mark. Now, this is, I think, the most confusing slide of the day. I am not the CEO of Ocado Retail. Hannah is, who's pictured. I am the CEO of Ocado Group. Name up there. OK, so let's just look at retail for a moment. So what are we seeing here? We're growing ahead of the competition. You can see the growth in our growth. And that's through Hannah and the team doing a great job on improving our unbeatable choice, our unrivaled service and our reassuringly good value. And as Stephen mentioned before, our average selling price has been increasing less than the Nielsen data suggesting that grocery prices are on average increasing. So we've been improving our value and that is driving us to have leading sales growth. We've got growing loyalty. You can see active customers up 8.1%, but with our mature customer base up 9.7%. We have improved our marketing, optimizing the channels, lowering our vouchering year on year, and improving our nursery journey. If we move over to volume, which of course is the critical one, we want to grow volume. You can see coming out of COVID how we were suffering year-on-year negative volumes, still suffering those at the end of the third quarter of 2023. And you can see how the volume growth is steadily improving and up in the second quarter of 2024, up 11%. And so we're very, very pleased with that trend. We would obviously like it to continue. Obviously, behind the scenes, all the volume at Ocado Retail is powered by our CFCs. As we mentioned before, our utilisation is driven by that growth. Our utilisation of the currently paid for and live modules, so in all the OSP sheds and in Dorden, is now up at around 80%. As I mentioned earlier, we'll be able to deliver more modules without Ocado Retail having to spend capex on new buildings in the existing warehouses to give them more growth beyond what is implied there. We've also seen an improving trend on CFC efficiency. Pleased to say that we have now exceeded 250 UPH in our Luton facility and we are now more than 25% of the each is in that facility being picked by our new on-grid robotics, which we will roll out across our client network to come. And also we talked about the last mile metrics and how the last mile metrics there, we've kind of back at 21. We've seen slightly higher stem times from the location of Luton compared to Hatfield. And a lot of the work that we're doing to drive optimization in this area, we can't take advantage of yet, as I spoke about before, because the routing that is driving that is still on our legacy systems, not yet on OSP. Whereas all of this warehouse is running on OSP with the exception of the Dorden facility. We have a clear pathway to get to high mid single digit EBITDA margins in the medium term in Ocado retail. We are building profitability through our perfect execution that is driving growth in customers, orders and volumes to leverage an efficient cost base. We are also driving greater efficiencies through the rollout of those reimagined innovations. It is still in the early days across the network in terms of AFL, our automated frame loading, as well as our OGRP. We will get more benefits As we roll on to the OSP platform, in terms of the short lead time orders, I am particularly excited about the growth opportunities that short lead time orders will deliver for Ocado Retail next year and improved cut-off times. improved cutoff times, that's when we say to a customer, you need to be out of the shop by a certain time. And the new OSP, short lead time software, will allow us to let all our customers add to their baskets right up to the close. Certainly, everybody will be allowed to add up to midnight. And we know that as we extend lead times, customers put more in their baskets. And obviously, that has an incrementally higher conversion rate. So We're excited about that. And as I said, we expect the EBITDA margins to continue to build and create that clear structural advantage as the channel shift has resumed and continues growing. So our key conclusions and our outlook, a strong first half with a clear trajectory of improving EBITDA and improving cash flow. Very importantly, well placed to continue our financial operational and strategic progress. Still plenty to do. On track to meet our midterm capex and cost targets and to turn cash flow positive during 2026. We have an absolute focus on helping our partners to become the leading online grocery retailers to best serve their shoppers and to leverage that structural growth opportunity with the resumption of the channel shift. And now we'll shift to questions. Go ahead, Marcus.

Hi, it's Marcus with JP Morgan. Three questions. Maybe the first one for Stephen on the numbers. I mean, we are seeing, are you guiding for 50 million less capex? You're guiding for 50 million more cash flow largely as a result. You're also guiding for better EBITDA in solutions. Yes. So what's the delta, i.e. kind of like how much is this additional EBITDA dropping through? That's the first question.

No, the delta is in working capital. Ocado Retail is a growing business. Its working capital requirements are growing, and that is the key delta in that maths, actually. So I highlighted the two key drivers, but that's the key difference.

Perfect. And then maybe second question. Now, as we really have good visibility of new CFCs ramping, what can we do in terms of additional disclosure? For example, a lot of clients are basically saying, what are the online sales of your clients really doing? How many of the current CFCs that you have are profitable? Because I think that would just really help hugely to really extrapolate and draw the line. I guess it's a wider discussion, but maybe if you can have some words on this.

We have a challenge here, which is that we do appreciate how smart you all are and that every time we've tried to work out how to disclose more around our total sites utilization and stuff, we can see the way that you can apply it and read into our clients' numbers. And our clients are very clear that their numbers are theirs to disclose and not ours to disclose. I think as we get bigger numbers of sites live with more clients, then we might actually be able to be more transparent with it without you trying to back out, well, what's Kroger's share and what's this one's share and what's that one's share? And I know that obviously everybody would love to see more disclosure around our clients and their facilities. Their profitability is their profitability. It's not for us to discuss or disclose. Obviously, we do disclose the margins that we generate from the sites and we do disclose the costs for us of serving the average site. As I said, it's absolutely not for us to sit here and start discussing whether we think one client's warehouse is profitable for them or not. We don't even have disclosure. We see cost lines and things like that, but we don't always see their full operating, their margins on their cog. We can estimate certain data, but we don't have all that data, and it's not ours to disclose.

Just as a thought, maybe an aggregate. I'm not really asking for specific numbers, but what can be done?

It's challenging in aggregate at the moment, given the relative size of Ocado Retail, which we do disclose, and then Kroger. And so the question is how easy it is for you to then deduce that and try and meet and end up in the position of disclosing something that's someone else's data.

Okay, fair enough. And then lastly, maybe to Kroger.

Obviously, we've seen the statement. You're highlighting that the Starlet Phoenix are going to come in 2025 and 2026. Sorry, I'm highlighting what in 2025? So basically the two more CFCs for specifically Kroger's. are coming. Just to redraw the line, you basically don't have any indication that there are any delays in the Kruger CFCs by where we are today?

No more than we've previously spoken about. We're not talking about CFCs 11 and 12 and 13 and 14 and 15 and 16, etc. So we hope to be talking about those in the future. But right now, what we're seeing is very strong growth, as Rodney spoke about. I think he said near doubling in the last quarter versus the same quarter last year. I think that's an important number to highlight. Fastest growing part you can see from the data of their business. and obviously with that near doubling will come a demand for more modules in the existing sites, and it won't be long, I think, until some of those sites, despite the fact that we can offer extra capacity in them, will still nevertheless become full. So I think it's a watch this space, and let's see what happens next.

Mark, as a point of clarity as well, I think you said 25 and 26, Phoenix and Charlotte. They're both 25. Okay, even better.

Okay. Yeah. Thank you.

Sorry, yeah.

Hi, Sarah Roberts from Barclays. Thank you for taking my question. So just firstly, you mentioned you have good visibility into the module rollout up until FY26. So there are 120 modules by the end of this year, then an additional 20. It would just be helpful from our perspective to kind of get an indication of what gives you the confidence that you have that visibility into the midterm, given that what we're seeing on our side is obviously Kroger and Sobeys are slowing out their rollout. So I suppose the question is, what gives you that confidence that those additional modules aren't going to get pushed out beyond the FY26 timeline?

I think if you look at the numbers that we're putting out, they are... in effect, somewhat conservative in terms of you can see what's coming in committed modules in live sites with unsigned deals that are going live. And then the addition behind that is some of incremental modules in sites that are already live where we can see the growth. And so we are already rolling out modules, incremental modules, at some sites that have already achieved the utilization of the modules they're already paying for. And so obviously we can see it site by site and say, okay, well, that one is going to need more than it already has. uh and it's going to hit it in three months can hit it in six months nine months 12 months and we can add that up we're not banking we we expect but don't need to see a a large reaction to any forthcoming um enhancements that we're bringing to the platform in for example you know we're not suddenly saying oh well short lead time orders is going to give us some huge win and that's what's going to get us to those numbers i think we've got upside if if it does um But I think we've got very strong confidence based on the rollout of sites that we know are going live, plus modules we can see getting drawn to get to those numbers.

That's helpful. Thank you. And just very quickly on the OIA, I appreciate it's early days, but just kind of the sales pitches that you're kind of participating at the moment. What is the initial feedback from clients that you're pitching to and how does your technology compare against peers such as AutoStore would be helpful incremental colour?

Yeah, so the feedback generally is very positive. I guess, obviously, we're not winning every bid that we enter, as you'd expect. And so in some cases, people... Well, I guess I made a point earlier about the fact that we play particularly strongly in the larger projects, and we're going to do some smaller projects, but we're not as strong there yet as we will be in the future. And AutoStore in particular do a lot of projects that are really, really small, kind of right at the bottom end of the market. And at the moment there, I would say we're not quite as competitive as we'd like to be. That will come with a bit more volume And so, in general, by the way, I guess to put some color on some of the positive feedback, what people really love about the solution is that it is widely deployed already. It is easily demonstratable at huge scale, like really enormous throughput scale. And Ocado, I think I mentioned this when I was talking, we do speak with tremendous confidence about what the solution is capable of because we We operate it and we can take you somewhere where it's doing exactly what we say on the tin. And we can offer some of that expertise to make sure that you get the same result. And so, yeah, people are responding very, very positively to a kind of a deep and wide knowledge of how to operate automation at scale.

It is extremely rare in automation projects to find people that can achieve what was written on the tin when they signed the contract. And here we are and we've explained this morning that we're taking up the potential modules in all of the live sites beyond their original design criteria. To be able to demonstrate and show that is just is really unusual.

Great, thank you.

Hi, Tin Tin Stormen from Deutsche Numis. On OIA again, just picking up on that last comment, and is that a comment on pricing in terms of competitiveness in the smaller warehouses? And maybe if it is, if you could give a little bit of color in that. And is the effort, the sales effort in OIA always going to be, for now, a direct effort, or do you see yourselves partnering more? with other solutions, if that makes sense, into a warehouse. And then whilst I just have the mic, I might as well chuck it in. Any chance you could give us the utilization rate range of the CFCs that are live?

I mean, on the last one, some of them have very recently gone live. So they're extremely low, right up to, you know, needing to draw down modules because they're 100 percent of their drawn capacity. And I know it's a it's a wide range. And I think that they've been public in saying they've drawn down now, I think, five and a half out of the original seven in one of the Kroger sites. So that gives you an idea. And I've been drawing that down in quarter modules and continue to draw down. So that's at the high end. UK site utilization is obviously high. And then it varies. And some of them are low, but they're growing.

Okay, so on the OAA points, I'll do those in reverse order as well. So your second question was about whether or not we will always sell direct, I think. The answer to that, I think, is no. So our intention was always to develop an integrated channel, reseller channel. I would say the impression we made at the trade shows generated a lot of demand for that. So we have been in weekly conversations with a wide basket of of integrators and resellers ever since we exhibited at the first trade show in February. Some of the late stage project bids that we're currently involved in are projects that integrators brought to us and asked if we could deploy our automation into their project and so I expect going forward there will be a healthy mixture of both I'd like to still do direct projects if we can because Ocado has that capability and we bring something special to bear especially when the projects become very large but we definitely will nurture, I think, quite a big integrated channel. And in the very long run, I would expect more projects to be done by integrators than by Ocado Direct, for sure, probably by quite a large margin. On the point about the pricing, yes. So if you like, the fact that we're more competitive at large scale than small scale is a pricing issue, largely. And it's born out of the fact that we grew out of grocery where volumes are very, very high. And so some of the elements of the Ocado solution are optimized for very high throughput. And in some of these very small deals, the like of which, you know, AutoStore does many. Throughput is less of a concern, if you like, or certainly the kind of throughputs that we have been able to generate in grocery. And so Ocado is already working on some lower cost versions of one or two of its peripherals that are more suitable at that end of the market compared to the very, very high volume throughput like the grocery world we came from.

Quick question for me. Kroger reported that their sales using delivery system grew nearly 100%, while the overall growth was 17%, suggesting that the other two channels they pursue, in-store fulfillment and using companies such as Instacart, are barely growing. Why is that the case? Why are you able to grow so much faster than the other two channels, and how sustainable is it?

That's my first question. I guess, Alice, I can only talk about our part of the growth. It's not for me to talk about the rest of the company's growth. As Rodney pointed out, the shoppers, they love the service and the NPS scores are very strong. The fulfillment rates are higher than you can achieve in store. We think there's still a long way to go on Kroger's fulfillment rates. They're not up as best in class yet alongside some of our other clients. There are some elements of the end-to-end solution where they use their own software. That's having a bearing on that, which will come. We'll see those improvements come. But overall, you know, the US grew up in a different way, as you know, in terms of a market, you know, with more kind of crowdsourced, independent people bringing the groceries in their car, having walked the supermarket to collect them. The numbers of substitutions and... missing items that customers in the U.S. are used to compared to what this service is delivering and the quality of them arriving in a white glove service with a chilled van arriving outside your house with a dedicated employee of our client Kroger delivering to you in a uniform is a different experience for many U.S. consumers. And I think the feedback has been very strong and they're seeing very strong growth in a number of these sites, very, very strong growth. This is durable.

growth rates?

Look, yes, I mean, it's definitely going to grow strongly. I don't know exactly how strongly, if you see what I mean, but we're expecting to enhance the TAM of what can be addressed with it through the short lead time orders. I think the impact of that will be strongest felt in the US market than elsewhere because the US market grew up through the very heavy labor model of Instacart, which has a disadvantage in every major factor in terms of range, pricing, freshness, accuracy, but had one selling point, which was immediacy. And with shortly time orders coming, then our model will address the immediacy whilst at the same time maintaining the chill chain, maintaining the choice, maintaining the accuracy. And because it's much more cost effective to operate, we believe our partner can offer that at a materially lower price. So some of our own teams, just feeding back to us, have done pricing comparisons of ordering Kroger delivery versus ordering on Kroger's site and getting an Instacart delivery or ordering Kroger on Instacart. And there's material cost difference. And the US, I think, is the least price sensitive market that we know. I think if someone came out with a service in the At the price premiums that you see Instacart at in the U.S., they would get no traction here at all, outside of a minor immediacy business, if you see what I mean. But the U.S. consumers ultimately will notice the price and quality difference, and I think it will continue to grow.

And then the spoke closure in southern Florida and opening up a new spoke. Is it just a matter of distance? It's closer to a customer and therefore better economics? That's the entire explanation.

Yeah, I think Kroger opened three sites that they themselves have said, you know, with the benefit of hindsight, they maybe shouldn't have opened them. They were too far away from the processing sites. And based on the strong growth that they can see where else they're trading, they just don't need to be open in those geographies to fill those sites up in the near future. And therefore, why serve longer distances where you've got longer cutoff times, significantly higher costs in terms of trunking those distances when you can sell it all closer? And if you do sell it closer, you gain the benefit of the increased density in those geographies as well, where you've now got more ability to sell it And so those were smart moves to learn from the experience of not needing them and to close them. And yeah, from what Rodney's saying, he's going to open another one in Florida to obviously leverage the site that he has in Groveland, but in a location that's materially closer to the CFC than the original spoke site.

My very last question for Mark. In terms of the size, there is a confusion about the size of the contracts that you're attempting to sign for your for ACADA technology, ACADA Intelligent Automation, what is the range of contracts that you're pursuing, or the range in pound sign or dollar sign of contracts that you're pursuing? For the main business, not Chuck, but for the main business.

Yeah, I'd say, okay, I can give you, I think, a real range. Probably the smallest project that we've taken seriously so far is around $5 million, and the largest is probably topping out just a little over $100 million. So it's a pretty wide range, actually. That's not quite the whole range of ASRS deals. There are in the market a few smaller and a few bigger than I just said, but that covers most of them.

Thank you. It's Luke Holbrook from Morgan Stanley. My first question is just on the refinancing. Just interested to hear why you've left it so late to refinance the first December 2025 600 million bond.

I don't think we've left it late. I think, you know, my view is that, you know, we've got a pretty good existing debt structure and a pretty low cost of capital and let that one run. And I think, you know, I think when I look at the markets today, well, they are what they are, but we've got a strong equity story. The markets, we believe, are very much open to, you know, generally and to us and confident of executing before they go current. That's our logic.

Understood. And the second question is just on E.ON. So you've signed the third CFC deal. Just be interested to hear if the terms of that deal, that fee structure, is the same as the first two or is it under a different fee structure?

Same. The same.

Okay. And then my final question would just be on the non-grocery side. You'd guided to three to five deals, potentially per annum. Just interested to hear whether that still applies for this year, given the pipeline that you're suggesting.

That's definitely still a possibility for this year, but I obviously don't have a deep insight into whether or not the business that we're pitching for, we will win this year. So I can't tell you how many of those deals we might win, but that's still definitely a possibility. Understood. Thank you.

William Woods from Bernstein. The first one is just on Phoenix and Charlotte. Do you actually have the go-ahead from Kroger to open those in 25, or do you need to wait for Kroger to say go?

I mean, ultimately, the day before a site opens, if a client says it's not going live, it's not going live. So I can't make that call. We have strong belief that those sites are going live in 25.

Understood. And then if you look at, you've obviously said that online grocery penetration is increasing, but if you look at the international modules, you've had, I think, one or nothing drawdown in the last six months. Why haven't we seen any drawdown there? And would we expect to see any above and beyond the new CFCs that are coming live in the next six months?

So we actually have had some module drawdown in the last six, and you will see module drawdown in the remainder of the year.

So is that that some modules have been closed?

There's one module that was to do the exclusivity in Canada that they were able to return, which is the only one in the world that was in that situation. And so we've had a net switch over. But we have been drawing down. I mentioned already that Monroe is now at five and a half modules and we expect to see continued growth there and in some other sites as well. So expect some more growth in the second half and expect material growth more growth compared to that in 25. As I said at the last time we present, if you open one that's at 40%, 50%, 60% of its endgame, when you turn it on, if you are doing this, and some of our prospective clients have got quite large businesses and they could just take a site and put 80% in immediately, a bit like we did in Luton. That's obviously an ideal scenario, but where people are building new businesses, obviously they've got to start to build up And ideally, we'd like to build them up in a linear fashion. Actually, it's more realistic that they'll build up in some exponential fashion in the way that most businesses grow as a percentage. We always talk about percentage of last year. But when we look forward at these sites, we can see a number of them starting to want to draw down.

Excellent. And then just going back to the cash flow seminar in 2022, I think you were talking about getting to 300 modules in the midterm. It now looks like the slide is showing 130 modules by FY25 with potentially no exponential improvement into 26, 27, et cetera. But you've kept the guidance for tech and support overheads flat versus that seminar. And you've said that the free cash flow, there's a clear roadmap to that by FY26. What are we missing in those pieces? Because that doesn't feel like it's logically consistent.

Yeah, we put a lot of range around that 300 free call. That was like a midpoint. We also had sort of, you know, we showed models at sort of 200 and sort of 400, 500 as well. A lot of it depends as well on the MHE capex that you assume in that year as well. We have visibility around that capex because we have visibility around the 26, 27 go lives. So those are the two key differences, I would say, from what we presented at the cash flow seminar. The basic economics remain the same and the dynamics that we talked about then.

I mean, I'm going to add there that whilst they remain, they've probably slightly improved. So the revenue side has improved through the reimagined technologies. And we're being more successful at bringing down the engineering and the cloud costs. So the margin on existing live modules is now, we would now forecast that in 26, 27 to be stronger than we were forecasting that before. So you're right, the number of modules we would be forecasting is probably less. We still expect to get there. It's just taking a bit longer because of what's happened in the market. But the margins that we can make on the product is better. And that is as a result of the work that we've done investing into the product is having strong results.

Excellent.

Thank you.

Thanks. It's Giles Thorne from Jefferies. I have three questions, too. The first one's for Tim, and it's a bit more of a mid-level, high-level question. But you've obviously been calling out two headwinds to partner behavior, both existing and potential partners, being the food inflation and then obviously extracting the best performance out of the kit. Which would you say has been the bigger influence on partner behavior over the recent past, and then how will that reverse or evolve as we move forward? recognizing that obviously one's not really in your control and the other one's something only you can influence. And building on that second question, it's a bit more specific, but it might be a segue, which is we've seen E.ON come out with the third CFC, which is little more than 12 months or maybe 15 months since Green Beans went live. So is there anything in the early stages of Green Beans and, you know, economics and efficiency and performance that accelerated that third CFC decision? Or was this just, you know, this had been in the hopper a very, very long time? And then lastly, and it's one for Stephen, well, maybe for Tim too, Just the improving retail performance, how is this influencing your negotiations and your negotiating leverage with M&S?

I'm going to go in reverse order, and I might ask you to remind me what you've asked me. We're not commenting beyond the fact that we've got ongoing conversations, so I think we said that today. We've had very constructive conversations, got a great working relationship, had an all-day strategy meeting on ORL recently, so our relationship with M&S is very strong. I think as Archie put it very nicely in joint ventures, sometimes you have a few difficult moments. That is a difficult moment we're working through together in a very mature way. And those conversations are confidential and ongoing. In terms of your question, I'm remembering your first question now was about, do I think the influence on growth is more around... the market and those kind of sectoral trends or more around us working more closely with our partners and how they're improving their operations and stuff. And I think it's a bit of both, right? If you don't think that the online market is growing or if it went through a couple of years of decline, even if your absolute volume suggests that you should move towards automation, you probably hold back on making that decision whilst you see where it stabilizes. And is it going to continue growing? And so I think now we've seen the, I mean, I think one of the interesting numbers in there was the basket size starting to actually be slightly stronger in ORL than it was before. Basket size falling was actually the bigger driver of the loss of volume at ORL. And you saw that very strong trajectory of each's growth there. So I think in a more stable economic environment with more stable pricing and wages and input prices like energy and stuff like that, it's easier for businesses to make investment decisions. And with underlying growth in the sector, in terms of the online channel shift, that's positive. And what's definitely also an influence and true is that it's not as simple as being a phenomenal bricks and mortar retailer and buying some great software and robots from Ocado and just putting them together. That's like bringing the ingredients and bringing the oven and expecting to have a fantastic product at the end. You also need to be a chef in the middle. And so I think we're working together with our partners to help them as to how to leverage the brilliant things they bring, how to leverage what we bring to the party as well, and how together to get the most out of that, to deliver the most outstanding offer to their shoppers, and by doing so, how to grow efficiently and how to grow in an economically sustainable and ultimately profitable way. And that's what we're doing. And we believe it will deliver strong results in the future. There was a middle part of the question.

Is there any evidence of that in green beans?

Look, again, it's not for me to talk about an individual client's economics or evidence or growth. Obviously, they decided to commit to a warehouse. It's a very large warehouse. Tokyo is a very big metropolitan area, extremely big metropolitan area. and they've got big ambitions for it and obviously that they're happy with the progress that they've made to date to make that decision.

Thanks. Charles Allen from Bloomberg Intelligence. Three sort of technical questions. First, how much of the improvement in technology profits could be attributed to basically the empty modules, the ones you're charging 33 million for but presumably have little cost anymore? Secondly, on your cash flow forecast, is that assuming there will be a retail EBITDA contribution in that cash flow forecast, even if it were deconsolidated by then?

Hold on, can we just take this one at a time, if that's OK? So the first one is the Hatfield fees at 33 million a year is about... The cost structure there is not dissimilar to any other warehouses. There's about 10 million of that or so would have been operating costs if they were live. I imagine there's still a bit of that going through because we've got an unwind in there and we've got to take that... dismantle that facility etc so i'm not sure exactly how much of that transfers in but there's some elements of that in there it's not a huge amount of it the next question was the next question is you've your forecast for positive cash flow by 2026 does it include a retail ebitda contribution yes it does

It's a pretty modest number. I won't give you the precise number. As you know, Ocado Retail is not cash-generative today. It generates a pretty modest EBITDA, insufficient to cover its lease costs and its capital recharges to Ocado Group. But it does, to short answer your question, but only a modest amount.

And thirdly, are you considering asset-backed financing solutions to refinance the debt in the next few months?

I mean, I think it's definitely something on the horizon for future sites. It's more of interest to us than the historical. It is an option on historical, but it's definitely something that people are talking to us about, future sites with existing and potentially new clients. Can we get the mic all the way over this side? We've got another one this side. How about that? Yeah.

Thank you. Sridhar Mahankali from UBS. Really, a couple of questions, please. I'll just go one by one, I think. I think the first one is you've talked about not having structural impediments to reaching profitability at your clients. I guess maybe you could talk about a couple of key issues that Partnership Success is trying to solve to get them there. And is there any particular site or a client that is really getting very close to those targeted economics? Obviously, you don't need to go into any detail, but that will be very helpful.

Look, part of partner success is highlighting where certain decisions that you could fall either side. I mean, I showed one on there earlier about range. So some people just intuitively think, well, I'll put the range in that's in the average of my supermarkets in that area, as opposed to I'll put in a differentiated and larger range. Not only will I put in that every product is available in any supermarket in that area. But I'll then go and find things that have wider appeal, whether that's, you know, specific ethnic products or wider ranges of organic or more premium products or whatever it might be, which Ocado Retail has done very well over the years. Right. And just being able to show people data that shows the impact of that like that, it may be nothing more than a partner success trying to persuade that retailer, their supply chain, their retail teams, that you should go and do that. And then those things can drive basket size. They can improve acquisition. They can improve retention. They can improve frequency. They can improve media. Media is another area where a number of our clients don't have existing businesses that create that media opportunity. So the Ocado Retail's media take this year is over £100 million. It's at a percentage of sales that is probably a global average for profitability. And some of our partners haven't yet gone out after that money. And so it's kind of helping them to understand what that's for, how you use the tools, how you sell the tools, who you should go and talk to. can be around their supply chain to minimise waste whilst at the same time keeping extremely high levels of availability. It can be around just discussing pricing and promotional strategies to try and avoid promoting slabs of cheap water that fill up vans or try and persuade people not to sell loose bananas, as well as labour planning and... Just trying to help people get the most out of the machinery and utilize it most efficiently. Route geographies and routes and the trade-off between availability and between releasing vans and between labor planning. And it's just a lot of areas. Yeah, we're seeing some very positive results in a number of places and expect to see more. We've got more teams deployed locally. We've got tens of people deployed helping in the US. We've got people moving all over the world. And wherever we're helping and wherever, you know, together with our partners, people are making changes, they're seeing positive impacts. And those can be improving productivity in the warehouse. They can be improving drops per eight hour route in the vans. They can be improving labor utilization. They can be improving a supply chain availability. They can be improving waste. They can be improving acquisition, retention, literally across the board where we work together. critical that we work together with our partners and we can see, you know, material improvements that are helping to drive the behavior and the outcomes that we want to see and our partners want to see.

Thank you. Second one, just in terms of the technology solutions, midterm EBITDA margins, we could just talk through how much improvements are going to be driven by the contribution versus the direct operating costs over time. Thank you.

Well, how they're going to evolve, you mean the margins? Gosh, I mean... We are hiding for a very decent EBITDA margin for technology solutions. I think the operational leverage is strong. It will, of course, depend on the number of live modules. If we go for that sort of 150 live modules, contribution of around £3 million or so per module, and then you can take off your technology cost of probably sub-£100 million, P&L, that sort of number, and similarly support costs as well, down to around sort of £130 million, £140 million, you can see the operational leverage in that business. It will be a mid-high single teens, sorry, single tens of percent of EBITDA margin.

Look, I think what you've seen, I think 71 was the number we put today percent in terms of marginal contribution. We expect with the re-imagine for the revenue to increase. And despite the incremental automation that will roll out, we still expect the operating costs to come down quite away from where we are today. So expect that margin to move up from 71. And then as you're pointing out, then it's about... how much of it we're doing versus the central fixed. I think we're done. Thank you very much, everybody. Thank you for coming.