Ocado Group PLC

LSE:OCDO

Earnings Call Transcript 2020-Q2

[Stuart Alan Ransom Rose, Non]

Good morning, everyone, and thank you for joining us today. We're now some 4 months into the COVID crisis, a crisis, which has affected us all and one which we must bring under control. Everyone working in the front line deserves our gratitude, and this, of course, includes our own colleagues who have responded magnificently to the challenges that they have faced. Today, we're reporting our results for the first 6 months of the year. These results reflect the changing dynamics of the grocery retailing sector in the U.K. as well as the historic milestones that we have achieved for Ocado in the Solutions business, notably, the opening of our first new international CFCs for Groupe Casino in Paris and for Sobeys in Toronto. Now that we are hopefully entering a post-COVID world, we're also going to take the opportunity to talk a little bit about the future. Online channel shift has been growing significantly in the recent years, and we have been the vanguard of this movement. We believe the massive increase in demand for online grocery services that we saw during the crisis will lead to even more permanent changes in the retail grocery landscape going forward. At Ocado, our role is to enable our partners to grow faster, delivering to all customers an outstanding experience through a flexible ecosystem of CFCs, micro-fulfillment centers and Store Pick in the most economic and environmentally sustainable way. We want to tell you about how the market opportunity has grown. And what we are doing to help our partners grow faster online. To talk a little bit about how we're building an even stronger and more resilient supply chain and how we are innovating to stay ahead of the game, you will hear from Tim and Duncan, but you will also hear from other colleagues and our partners talking about the scale and scope of the opportunity that faces us. With that, let me now hand over to Duncan, who will take you through the first half numbers.

[Duncan Tatton-Brown]

Thanks, Stuart, and good morning, everyone. In these times of significant change for the grocery industry, I'll be brief this morning so we can spend more time with Tim talking about the future and what's going to happen rather than the results over the last 6 months. In summary, we performed well. Group revenue's been up 23%. EBITDA has been down and losses has been up as we continue to invest and we've launched our first international CFCs. We've reported exceptional income in the first on the back of the exceptional losses last year. These primarily relate to Andover, with last year being the asset write-off and this year being exceptional income, which we will continue to recognize throughout the balance of this year and into next. Remember, we comprehensively covered for the fire in Andover. So now on the results by segment. In International Solutions, we opened our first 2 CFCs in Paris and Toronto for Groupe Casino and Sobeys. We started to recognize revenue, but due to the detailed contractual terms, we actually only recognized revenue in 1 of the 2 CFCs right towards the end of the half. More importantly, cash fees from our clients were up strongly, up 58% on the year. And now, we have a cumulative unrecognized cash fees of GBP 219 million. EBITDA was down as we continue to invest in the business, and we have the first operational costs for the live CFCs. The UK Solutions & Logistics revenue grew slower, up 9%. Revenue includes 2 things: cash fees, we -- sorry, the costs we incur on behalf of our clients, Morrisons and Ocado Retail, who both grew strongly and the fees that we [ owe ]. Now costs grew slower than revenue because of our continued improvement in efficiencies, and cash fees were down year-on-year. Again, as a result of the fire in Andover, we reached an agreement with Morrisons for a temporary holiday from Erith alongside some discounted Store Pick fees, remembering, again, that the costs of that deal are covered by insurance. EBITDA was down as a result of that decline in revenue or lower growth in revenue and because we continue to invest. We need to get both of our U.K. clients onto our international software platform as soon as possible to realize all the benefits. Importantly, note that engineering cost per order in Erith fell again by around 20%. And our new generation robot is now in production. So on to the Retail segment, talking about the U.K., although I think the characteristics of what's happened to the U.K. is probably similar to grocers all over the world. There's been a significant change in customer behavior with an excess of demand, not all of which we'll be able to meet. As a result, our customers had fewer orders than normal but much bigger basket size. And basket size in the half grew by 28% with orders flat. So post-COVID, the growth in basket size was bigger with a small decline in orders. We remained price competitive, and our margins were actually slightly down. Now the effect of this change in customer behavior on the operations was that we, because of the increased demand, had a smooth week that means to say we were doing the same amount of business every day of the week, which means that we can produce -- we can pick more groceries across the week. And the combination of that and the bigger basket size, which means that customers often buy more of the same item, meant that we were able to pick more and be more efficient. And so our efficiency metric, our UPH metric grew by 7%, inclusive of Erith, and Erith's underlying was improving, and Erith's UPH is now ahead of the average. It's not the only thing on CFC costs because a lot of the CFC costs -- a good proportion of the CFC costs are fixed. And given the extra sales through those facilities, extra volume through those facilities, fixed costs as a percent of sales fell. And so despite some additional costs, like the 10% bonus to frontline staff, overall CFC costs as a percent of sales fell by 80 basis points. On delivery, our metric for delivery, drops per van per week, actually fell by 9% because of the much bigger basket size. But the amount of items, we call eaches, processed by each van grew by 40% during the COVID period.Now the fees charged to Ocado Retail also fell as a percent of sales, again, because they're largely fixed, and so in -- cost as a percent of sales fell. A point of detail to note. Under IFRS 16, some of the cash fees that we charge, which really reflect historic capital spend, are now treated as lease charges, and therefore, accounted in depreciation and interest, leading to lower fees as a percent of sales. Overall, Ocado Retail generated an EBITDA margin of 4.5%. Now not all the measures we look at are financial. People are integral to our success, and we continue to work hard to look after them, wherever they were in the world, whatever role. Examples in the first half are the temperature monitoring, the antibody testing we provided to frontline staff. We ran mental health sessions for those of our staff who wanted it. And overall, in the first half, our mental well-being for our staff was up, importantly. We continue to hire. We continue to induct and COVID had no impact on our ability to continue to innovate and to grow our business. I would like to take another opportunity now to thank all of our staff for all of their amazing work, particularly in the first half of this year. Now back to a financial measure, cash. We generated strong cash flows in the first half of the year, partly because of the fees that we've earned from our clients not recognized as revenue. And given our financing at the beginning of the year and our financing in the last few weeks, we now have GBP 2.3 billion worth of cash, putting us in a strong position and giving us the opportunity to make the full of all the opportunities ahead of us, the full opportunity set ahead of us. So outlook remains positive, but no material change. I'd now like to hand you over to Tim. Thank you.

[Timothy Steiner, CEO & Executive Director]

Thank you very much, Duncan. Good morning, everyone. I'd like to talk about Ocado Group in a post-COVID-19 world. We've seen significant and permanent global channel shift. The world has truly changed. Here in the U.K., we've seen the market double in a few months. That's 15 years of growth in less than 15 weeks, and it's a global trend. In the U.S., the market year-on-year was up 6x. In China, we saw triple-digit sales growth. The channel shift has been achieved despite constrained supply chains, despite capacity being difficult to bring online so quickly. There is more demand behind it. And we've also seen that this demand is going to stay. There's a strong expectation that it's sustainable. There are stats from the U.K., from the U.S. and from China that all support that people who've tried the channel will continue shopping in the channel. We know historically that customers that have done 3 to 5 shops online in a 3- to 4-month time frame historically stay with the channel. That is the hurdle you get over to make it more of your kind of common practice, to make it more of a habit. What we also know is that with the superior service and proposition that Ocado.com has and the OSP platform allows our clients to offer, we can acquire customers from other online retailers more easily than any other place. And so we know that historically, tescos.com (sic) [ tesco.com ] and sainsburys.com (sic) [ sainsburys.co.uk ] have been our best channels for acquisition, and they're the ones that we most overindex in. What we know today is that consumers have got a much better understanding of viruses and the transmission of disease. And everybody is taking less risk than they did historically. Some massively less risk, but everybody, some less risk, and it's about deciding where and when you want to take your risk. Do I take it to visit my family and friends for entertainment, for fun? Or do I take it to go to Tesco's and walk the aisles to get my groceries when Ocado would deliver them to the door and for a token incremental cost?And it's the token incremental cost I hear that I think is key. Because if you compare this to other markets where we're seeing very significant movements, like in transportation, where globally, public transportation levels remain on the floor, but private-car travel is back at pre-COVID levels or above in many markets. Private-car travel costing a lot more than public transport. Getting your groceries delivered to your doorstep does not cost a lot more than going to the hypermarket. In fact, if you include the cost of your own travel to the hypermarket, we believe it's cheaper. Business executives are going back to using private jets over flying commercially at 10 to 20x the cost. Again, massive cost increase, but you feel safe. Ocado can do that, and our clients can do that using OSP for customers and they're safe at a marginal token incremental fee. And our facilities also lend themselves naturally to social distancing, the automation, allowing all of our people to work more than 2 meters apart safely. The opportunity here is huge. As you know, we have existing commitments for 54 CFCs around the world, but there is enormous runway for growth. Our partners today have GBP 210 billion of annualized sales and operate in some of what we call key markets, those with more than $25,000 of average per capita GDP as well as more than 5 million people, where the grocery market is GBP 2.8 trillion annually. Now what we've seen so far is the market go from single-digit percentages to teens. But we can see demand for that to double again. When we see the market going to 20% to 30%, there's a big opportunity here for our clients on OSP to expand profitably and sustainably. And as they expand profitably and sustainably, the economics of the big bricks-and-mortar stores will start to be challenged.And over an extended time period, we see less opportunity for them going forwards. Our clients can grow and take up and win market share against their competitors. The opportunity here is potentially as large as 75% to 80% of the market going online and that online share to be significantly consolidated in terms of who holds it. When we see the market get to 75% or 80% online, as with all other online markets, it will have consolidated into fewer players, those with world-beating and -leading technology behind them, such as the Ocado Smart Platform. We really believe we are now seeing a redrawing of the global landscape in online grocery and an enormous opportunity for Ocado and its clients worldwide to dominate their markets. Now I'd like to hand over to Mel Smith, who's been the CEO of Ocado Retail since August last year. Over to you Mel.

[Melanie Smith, Chief Executive Officer]

Hi, I'm Mel. I'm the CEO of Ocado Retail. We've had an extraordinary few months. The online grocery channel has doubled in size, and Ocado.com has been the fastest-growing grocer in the U.K. market. We've always had an amazing customer proposition. We have the widest range in the market, twice bigger than our competitors. We have the best service and delivery scores. We have great value for money on our products, and we're the most sustainable grocer in the U.K. And that's why our customer service scores are better than anyone else's in the market. And we are well set up for growth. We are going to add 50% more capacity next year in our new facilities. We are super excited about M&S coming on the platform in September. I cannot wait to get Percy Pigs in my weekly delivery. And we're really, really excited about growing our Zoom proposition across London, which is our immediacy offer. And we have 1.1 million customers signed up, ready to get deliveries from us as soon as we have capacity. And finally, I'd really like to thank all of our colleagues on the front line, our frame loaders, our personal shoppers, our marshals, our drivers. They have done a tremendous job in the last 6 months looking after our customers, including 350,000 vulnerable households who we've been able to serve. And without you, we wouldn't have been able to deliver and look after our customers. So thank you very much to our frontline colleagues.

[Timothy Steiner, CEO & Executive Director]

Thank you very much, Mel. To wrap up, Ocado Retail is gearing itself up to grow faster here in the U.K. We've seen 27.5% (sic) [ 27.2% ] on the first half sales growth, with sustained increased demand. So we'll expect that to carry on going at a fast rate. In September, we're going to see a significant improvement to our customer offer with the range switchover to Marks & Spencer's bringing in over 6,000 new food lines, 1,000 of which have been developed specifically for the offering by our friends at Marks & Spencer's; and over 1,000 Clothing & Home lines, significantly improving that offer on ocado.com. We're also looking to grow the nearer-to-customer missions with Zoom and soon launch the second site in North London. We've also got big pools of potential new customers, not only those on the list that Mel spoke about, the 12 million weekly customers in Marks & Spencer's but also the population that now understand that this is the safest way to shop, not only the easiest, the freshest food, the biggest ranges and the least effort but the safest way to shop as well. And we're adding significant new capacity over the next 18 months. Over 30,000 orders a week of capacity in Bristol coming onstream early in 2021, and then later in 2021, the rebuild of Andover and Purfleet coming onstream with around 150,000 orders of capacity between them. So we've got a significant growth opportunity and Ocado Retail is ready to seize that opportunity here in the United Kingdom. We've been talking a lot about the U.K. but what's happening in the U.K. is being mirrored all around the world, in all of the markets we're present in and a number where we have potential clients. To talk more about this, I'm going to hand over to Luke Jensen, the CEO of Ocado Solutions. Over to you, Luke.

[Luke Giles William Jensen, CEO of Ocado Solutions & Executive Director]

I'm Luke Jensen. I'm the CEO of Ocado Solutions, and I'm delighted to be talking to you today. Ocado and our partners are doing something completely new in building these facilities together. The kind of partnerships we have developed are unique both in the grocery sector and the tech sector. We build these CFCs together, and we operate them hand-in-hand. They bring together amazing tech talent and grocery expertise under one roof. So these launches have really broken new ground for us, our partners and the grocery sector globally. The go-lives are more than just symbolic for the Ocado story. They're not just pins to drop into a world map. They're also core to bringing an unbeaten level of customer service to shoppers in these countries for the first time. I'd like to introduce Jean-Paul Mochet, President of Monoprix, Franprix and one of the leaders of Casino Group, who's going to talk about our partnership.

[Jean-Paul Mochet]

[Foreign Language]

[Luke Giles William Jensen, CEO of Ocado Solutions & Executive Director]

Thank you, Jean-Paul. That was great, and we look forward to many years of working together. We're a company that operates now in 9 global markets. So of course, travel restrictions represented a particular challenge for us. We always expected to have more people on the ground temporarily in the CFCs to guide them through testing, ironing out any kinks and to troubleshoot. But with borders closed, we've had to look for new solutions, such as guiding via video link where possible, and of course, relying more on the dedication of our newly trained engineers on sites in countries, who have been superb. Another ongoing area around training engineers for our upcoming CFCs is that normally, we would have planned for recruits from around the world to spend time at our sites in the U.K. to get to grips with the tech in fully operating sites. Obviously, now, we've been looking at ways around this, with remote training more and more in the spotlight. Just taking the U.S. as an example, we're looking at training with live video, recorded video, other forms of virtual and online training, pop-up training facilities in-country and potentially in the future, virtual reality and augmented reality. So what's great is that our people have found very smart ways of getting around the constraints and the restrictions to be able to continue with our projects and go live with our sites. There's no doubt that COVID-19 has driven the online grocery market to a new inflection point. Of course, demand for online skyrocketed everywhere where lockdowns were imposed. The question of structural change in the market comes down to how much of this is going to stick after the crisis as we start to ease in different markets. One core lesson that we've learned from being at the forefront of online grocery for 20 years is that once people have shopped online for the first time, they're likely to come back for more, particularly where the service is great and the cost is relatively low. COVID-19 has pushed many people into shopping for online grocery for the first time, and we would expect much of this to stick. Recent evidence bears this out. If you look at the Kantar data in the U.K. showed that in June, even though lockdown was easing, grocery online sales continued to accelerate. Our partners, Casino and Sobeys, have brought forward their customer launches in the CFCs to accommodate higher demand than anticipated. And consumer surveys are beginning to show, particularly in the U.S. and Europe, that a large proportion of those that have shopped online for groceries during the crisis will continue to do so. And I'll expect to see similar patterns in other market as more research comes out. And this bears out long-standing trends in our own data for the U.K. that shows that as new cohorts come to online grocery, they tend to stick. As e-commerce penetration rises around the world, we're looking at a structural change. And for Ocado Solutions, this means that we're having more conversations with our current partners about how we can help them go further and faster. And in terms of new prospects on the horizon, all I'll say is that it's very clear that building sustainable, scalable offerings for grocery online is now absolutely top of the agenda in grocery ballrooms worldwide. And Ocado remains the only player that has shown that it can help those grocers achieve that profitably at scale. Thank you.

[Timothy Steiner, CEO & Executive Director]

Thank you, Luke. So as Luke outlined, our partners are seeing significant demand today and expecting significant demand growth in the medium to long term. So looking at the actual crisis, if we look at our partners who are already online on the OSP platform but using it for our -- its Store-Pick capabilities, we saw massive growth, and we enabled that growth for them. And you can see from March to May, we grew by 400%. Erith volumes, at the same time, also grew significantly, growing 50% from March to May to help Ocado achieve its growth. We're also seeing a faster launch from our partner CFCs. So you can see statements there from both Sobeys and Groupe Casino, where we helped them in the middle of lockdown to launch their new facilities and are currently helping them to scale those facilities in Toronto and Paris at faster rates than were previously planned pre-COVID. We've successfully accelerated for our partners and continue to evaluate how we can work together to go even faster. We can serve our customers' needs flexibly with our OSP ecosystem. It's a multi-format platform that offers long-term leadership across all missions with tactical flexibility. So you can look at Store Pick, micros, minis and standard warehouses, all available on Ocado Smart Platform. And they all have different benefits and different drawbacks, no one model serves all cases, but the flexibility that we bring means that our partners can operate whatever they want on our platform. Look at shopping missions, so big baskets versus immediacy, that's standard in mini-warehouses versus micros. Customers may pay more for a reduced offer in exchange for speed and a micro, whilst the most efficient of its class, is more expensive to run than a standard. You can look at market-specific overlays like density, wage rates, land and fuel costs, consumer preferences. Notably, in some of the markets that we're operating in today with our clients, enormous geography; and outside of the densely populated urban areas, Store Pick might be most appropriate. In smaller cities, minis might be most appropriate; in bigger cities, standards. And again, for immediately, micros. And then you've got tactical. What's best for now isn't always best for the long term. So right now, the way to capture market share immediately is to expand Store Pick if you have a store network, and that's exactly what we've helped some of our partners to do. But they can later transfer those customers seamlessly to being served by a standard warehouse to create a profitable, sustainable and long-term and actually superior customer proposition for those customers as well, but they can pick them up now and start serving them on Ocado Smart Platform. This flexibility across missions, markets and time horizons is unique to the OSP model. In a post-COVID world, serving our partners' needs is both more challenging because we need to go faster, but also more challenging because of travel restrictions. But we have been working for the last year on building a scalable and global supply chain. We strengthened our supply chain to support future significantly faster growth. And over the last year, we've created new partnerships with scaled global leaders in contract manufacturing, such as Flex and Jabil, whilst continuing to work with our trusted long-standing manufacturers here in Europe. This has enabled us to scale MHE production for our partners faster, avoiding a hardware bottleneck that could be a risk to global growth. So it's giving us speed for rapid scaling to meet demand, flexibility to serve future growth in our OSP club and proximity to reduce price and lead times as regional manufacturing is available if required, which will, long term, improve our total cost of ownership as our capital costs of our assets reduce with scale over time. And we're benefiting from the scaled manufacturing operations of our partners and the larger order sizes we expect to place with this increased growth. We'll start to see the benefits of these new partnerships from the end of 2020. To best the serve our partners' needs, we're investing now for this acceleration. We've decided to bring on board more people into our tech team so they can go even faster. We ended 2019 with 1,700. We added 300 people in the first half of this year and are targeting an increase of over 300 people in the second half to take that team to well over 2,000 heads. This is to improve our platform capabilities and go faster in e-commerce, in logistics and supply chain, in fulfillment-handling robotics and in retrieval robotics. We want our platform to be secure, reliable, scalable, maintainable and have the functionality that it needs to allow our clients to dominate their markets and be the lead player in e-commerce in every market they operate in. The investment will also cater for the implications of ramping more CFCs faster. We've previously guided to an average of around GBP 30 million of net cumulative peak cash outflow for a "standard" CFC in its year of launch. But actually, if the client is going to ramp it faster in, say, 6 months rather than 24 to 36 months, the peak net outflow will be slightly higher. It will also mean that the NPV of the CFC for us will also ultimately be higher, too. Our strong cash position today of around GBP 2.3 billion enables us to capitalize on the full opportunity set over the medium term, not just to ramp up the existing CFCs faster, but to deliver more for our existing clients as well as to sign on new clients. We have the capital, human and financial, to go faster. Our clients not only benefit from the only proven platform to serve the online market profitably but also from being a member of the OSP leadership club. Scale gives us more resources to realize our partners' ambitions. The club is large and growing. Today, globally, it represents the #2 in volume of global grocery sales. That scale justifies the resources that we bring on board and the rationale to invest in faster innovation. Nine partners today gives us a strong multiplier effect. The reach adds global insight to local strength. The club benefits from a global view of consumer trends. We're #1 in terms of geographic reach in global grocery, whilst making the most of our partner strengths in local markets. 75% of the top grocery markets globally are dominated by local players. So what we have always understood was partnering up a global technology brand with dominant local players creates the opportunity for the partnership to be the #1 in their space. The collaboration compounds the benefits of scale and reach. Our geographic diversity facilitates shared learnings between partners and we now have quarterly executive leadership meetings where the senior management teams of our partners and ourselves come together to discuss what's going on and to learn from each other, to bring diverse experiences to the topics of shared importance, whether that's online operations, whether that's crisis management at the moment, whether that's future-proofing their propositions. As the club grows, the benefits are magnified for all. Innovation has always been at the heart of Ocado and what we do. So now I'm going to hand over to James Matthews, the CEO of Ocado Technology.

[James Matthews, CEO of Ocado Technology]

Hello. I'm James Matthews, the CEO of Ocado Technology. Ocado Group has been building technology for itself for nearly 20 years. In recent times, we've transitioned to building a globally scaled technology platform for some of the world's biggest retailers. This is Ocado Smart Platform. Today, we're a 2,000-strong team of technologists, covering disciplines from materials, mechanical, electronics and software engineering in 7 locations across Europe, and we're growing rapidly, but we need to grow faster. The world has changed for online grocery, which means more demand for our platform in all its forms. We're investing aggressively in hiring top talent to support the development of OSP. We're looking to hire nearly 500 additional technologists in the year ahead. Growing faster isn't enough. We need to do more with excellent execution, both in launching our partners' operations in OSP and supporting their growth in online grocery by giving them the best tools to lead for the long term. Launching our first 2 international automated warehouses ahead of schedule, despite the challenges of COVID-19, was a huge milestone for us. Now we will focus on helping Groupe Casino and Sobeys to run and scale their businesses in a cost-effective fashion with our end-to-end platform as we have done for both Ocado.com and Morrisons here in the U.K. We've already taken important steps towards achieving this goal. We're in the process of launching our third-generation fleet of bots, which have been in development for the past 3 years. This bot, designed entirely by Ocado Technology, will set new benchmarks for performance and reliability. We continue to relentlessly push the boundaries of cutting-edge technology across our platform, from our user interfaces through supply-chain and fulfillment software to our last-mile systems. A significant recent development is bringing into production our robotic suction cup pick technology. This technology combines powerful machine learning and computer vision to enable the picking and packing of products without any prior knowledge of what they are. This is especially important when you have tens of thousands of different items with varying shapes, sizes and weights. We started with 1 robotic arm live at the beginning of this year and now have 3 in production, with another 2 being added in the near future. Each station is picking at double the speed it was capable of 6 months ago and is constantly improving, packing groceries for Ocado.com customers. These are just a few highlights to give you a glimpse of the exciting work that we do. Continuous innovation is business as usual for us. It's embedded in all of our engineering teams and it's central to our unique culture.

[Timothy Steiner, CEO & Executive Director]

Thank you, James. To just build on what James was saying, at Ocado, we've been innovating for 20 years, and I can't think of a single innovation that didn't benefit us in some way and help us to get to where we are today to be a leader in this space. What we want to do is to continue innovating, but at an accelerated pace, both within the core and outside. Inside the core, James spoke about things like robotic picking arms. It's important to understand that 50% of the human capital inside a warehouse is picking. Transforming that to robotic pick would enable everybody working inside those warehouses to be twice as productive. And then we could continue to work on automating further processes. This has a bigger benefit to us in a post-COVID world than it did before for a few reasons. Firstly, as the platform grows faster because our clients grow faster, the benefits of robotics that involve the upfront investment to create them but a saving on using them is obviously a much bigger opportunity. But robotics also benefits in a COVID world in terms of having less people in the workplace, less people through the doors, the changing rooms, the restaurant facilities and also to help our clients' customers to understand that their food hasn't even been touched by a human. So this is an area we want to expand the investments that we've been making over the last 6 or 7 years so that we can soon enable our clients to roll this out faster. We're also innovating outside of Ocado Technology, James' area. We have 10 [ ex ] teams that are predominantly focused on significant moon-shot-type changes that we would implement in our business 5 years out if they can pull them off, really significant and fun innovations. We also have ventures, both minority and majority stakes in businesses that we think are adjacent to us and are very linked in what they're trying to achieve. In vertical farming, for instance, whereas pre-COVID, that it started to come into its own because of the declining cost and the improving efficiency of LED lighting and where it was already more efficient to grow things like leafy greens close to where they're being used, rather than involving the air miles and the carbon that is wasted in transporting them, it also has an increased demand for vertical farms in a post-COVID world as people try and rationalize supply chains and think about food security. So these are areas that we're very excited to be involved in. Our other investments in 3D printing and in robotic food prep are also areas that we think will benefit in a pre-COVID world. In conclusion, the world has changed in grocery retailing for good. Ocado Retail has shown that even in times of crisis, it can deliver market-leading customer experience with strong and sustainable economics. Ocado Group has prepared for greater velocity of execution and innovation. We are ready for the future. Ocado has the financial capital to capitalize on the full opportunity set in the medium term. Thank you for joining us this morning. Now we're going to take questions. We're going to give you all a few minutes to register online to start asking questions, so please bear with us until that starts and stay online. Thank you.

[Timothy Steiner, CEO & Executive Director]

All right. We've got our first question from Andrew Gwynn at Exane BNP Paribas. He's actually going to ask 3 questions on capacity. Number one, what are the specific challenges on ramping up capacity in CFC4? Number two, if I was a new partner coming to Ocado, what's the earliest you could build me a new CFC? Number three, in light of the above, what are the plans to improve capacity for new partners?Andrew, so the first one is CFC4 obviously is kind of on the frontier in terms of scaling up this new type of robotic infrastructure. So it's doing over 100,000 orders a week at the moment. That's bigger than any of the warehouses that we're building for our clients at the moment. So any kind of technical challenges that we have in scaling it, that's one part of the challenge, and we've worked our way through loads of those. And the other part is just to do with the plan that we originally had for it and the production of both peripherals and new robots that we can add to the grid there. So we have scaled that up and added some faster, but some of the robotic components in those robots have lead times on them. And so it's just about accelerating the build time of actual robots. So we've added more on. We'll continue adding more on. We'll continue working through any challenges that we see in CFC4. It's significantly ahead of its plan at the moment as a result of our efforts. The second question, being the new partners. CFC normally takes between 18 and 24 months from someone actually filing a greenfield site with planning, so in terms of both initial design and then construction of the physical building. Then we come on-site and install our grids, add our robots and peripherals. And then together, we do kind of combined commissioning and testing and then go live to clients. So 18 to 24 months is usual. We are working on getting that down, and there are ways to make it faster if, for example, an existing design works, that can take a few months out. And there are other things that we're working on to try and bring our own part of the build process down. If a client has an existing warehouse, then it's possible to do all this faster, and that's frankly what we were doing in Bristol, was going into a building that was almost finished when we announced that project. It was built on -- as a spec building. So that's another way it can help. In light of the above, what are the plans to improve capacity for new partners? Look, I think the key is that our partners are asking us to bring on live their incremental capacity faster and to use the capacity they get asked for faster. And those are both things that we're working on with them as well as to start planning for new warehouses and to build those as quickly as possible as well as some of our partners are looking at adding or increasing the scale of the Store Pick software that we add to them as well so that they can help their customers right now. And then later, as we said in the presentation, later convert those into new warehouses that will give them the long-term sustainable economics. The next question is from Xavier at Bank of America, and it's has the pipeline of international CFC openings changed over the recent months. Are you going to open CFCs sooner? What should we expect in 2021 and 2022? Look, I think the projects that are already live are in construction. We've had a few small COVID-related delays, but they've normally been only a week here or a week there. So we're working very hard to keep those down at low levels. We are working on expanding the production of things like robots so that clients that wanted to go live with, say, 2 modules in a 6-module site and not -- and thought that would last them for 12 months are now saying, "Can I have the third one as soon as possible, the fourth one as soon as possible after that?" So we're working at accelerating the speed with which we can give them greater capacity. Obviously, in terms of warehouses going live, I wouldn't expect any additional ones in 2021. We might see somebody add something that would be capable of being built in 2022. I think we'll see more warehouses going live in 2023 than we did before. So it's an acceleration of the pipeline. The next question is from Sreedhar at UBS. Question number one, can you please give us an update on your thoughts on what Zoom footprint could look like in the U.K. and how different the second facility could look versus the first?So I'll take that part of the question first. Look, there's clearly demand for a dozen or 2 dozen Zooms just in London alone. And so we're excited to build our second one and then look for -- and we're looking for locations for further sites. The second one is going to look quite different from the first one in that it can both be built on a much smaller footprint and will be fully automated in both the chilled and -- in the ambient as well as in the chilled side using our own OSP robotics. So it will have significantly higher throughput per square foot and also significantly higher productivity. And so we're excited to get that second one open. Number two is on robotic picking. Can you give us the actual efficiency, what the actual efficiency is versus human picking? Do you have any thoughts on the further rollout of these? Our target for this year-end is for a robotic pick cell to have equal or greater efficiency than the human picking. So we think we can match the human pick speed in a robotic cell. That robotic cell may actually have 2 robots in it but will be the same footprint and the same pick station that used to be operated by 1 human. And so we're, again, excited about the rollout as we learn how to handle more of the products. Our next question is coming from Victoria at Credit Suisse. Do you think your partners can use Ocado CFCs alongside of alternative micro-fulfillment-center solutions from third parties? It's physically possible, Victoria. Our clients can also buy our own micro-fulfillment solutions, and that's something that a number of them are talking to us about. Our software is very flexible. And so a lot of things are possible. If a client already had an alternative fulfillment solution, then we can work with them to work out how to integrate that to the solution. Frankly, that is what we're doing at Ocado Retail. So Ocado Retail's first fulfillment centers in Hatfield, in Dordon, were not built using our own new robotics. And yet they will obviously talk to the Ocado software and operate in the Ocado ecosystem. Number two, if you had no capacity constraint, what was your first half retail sales growth would have been?It's really hard to say it, Victoria, because obviously you can see what looks like an unlimited wall of demand. And you can see as many customers in a single hour as you had in an entire half of the previous year come to you. It's difficult to know exactly what that would have translated to, whether it was 2x, 5x or 10x what we were running at. But it's certainly somewhere between 2 and 10x our last year's volumes, so I think -- we think would have been quite easily achievable. In number -- third question from Victoria. Tesco's had [ 1.3 billion ] online sales in Q1. Do you think they were taking market share from us?Look, if somebody expands their sales at a faster rate than we do, then they're ultimately taking market share from us. That's not something we're concerned about. In fact, the more that the Tesco's and the Sainsbury's and others grow in this channel, the happier we are. We have always managed to -- or found it easier to acquire the customers of Tesco's online or Sainsbury's online businesses than their bricks-and-mortar businesses. The way we look at it is that they have helped us to persuade that client to shop online. Now we only need to show them that we're better online than Tesco's and Sainsbury's, which we always have been. And so we -- those are the 2 groups that we've most overindexed in customer acquisition. So we're happy if they grow. How does the Tesco's urban fulfillment solution differ from Ocado CFC? And why does it cost them so much less? Look, I think that people have always said that they've built warehouses for less than ours. I think the key point is to do with the -- 2 things. One is the volume of sales that they achieve out of them. So there was a lot of bragging from another one of our competitors at the time that we built Dordon that the solutions that they built at that time would be doing GBP 10 million of sales, and Dordon, for between us and Morrisons is doing significantly, over GBP 1 billion of sales. So you can't really compare them very well. I don't know the scale of Tesco's urban fulfillment solutions, but it could be GBP 4 million just for the kit that goes in them. That might be not significantly different to the kit that goes in a micro-fulfillment center, and I would expect them to have slightly less sales than one of our own micro-fulfillment centers. So I still expect us to have a better capital-to-sales ratio. It's also important that you compare apples with apples and apples with pears and don't just compare the cost of kit in 1 building with the total cost of another building. Moving on, I -- well, sorry, before we move on, the other thing I think I'd say is that Tesco's and Sainsbury's have both been clear in their results announcements about the fact that they're not expecting to see significantly increased profitability this year despite overall higher sales, despite significant demand and despite the very significant increase in their online businesses -- well, in fact, because of the increase in their online businesses and despite the hundreds and hundreds of millions of pounds of rates that they've been effectively given back or not asked to pay by the government this year. Those -- that rate savings is not translating into profitability because of the increased costs of operating their e-commerce businesses. So what you've seen in our results is how increased sales in the online business translates into increased profitability. What you're seeing in their businesses is increased sales resulting in increased losses in the online space, eating into the profits of the store-based business. The next question coming from Numis. Do you expect more of your partners' facilities to open at or closer to full potential capacity? What are the implications for build times in this scenario? And I think the point there is that we do expect our clients to ramp their facilities much faster. So clients that were previously looking at 2 to 3 years to full capacity or 3 to 4 years to full capacity, I would fully expect them to kind of ramp them in kind of 12 to 18 months on average. Obviously, it'll depend in what geography they open them in; whether they've had existing bricks and mortar; where they've got an existing bricks and mortar business there, whether they've got an existing e-commerce business there; but overall, I think the expectation both for the Ocado Retail sites here in the U.K. and for our clients' sites is for much faster ramp-ups. And in terms of build times, it doesn't -- that itself doesn't change the amount of time it takes to build a new facility. It does, of course, change when you should order that new facility, and it significantly accelerates that. The next question is coming in from Robert, Goldman. Did Tim say that you're now able to build a GBP 350 million run rate CFC to full capacity in 6 months if the client wants to accelerate and ramp up? How many clients asked for this to date?It -- Rob, it's crucial that a client doesn't ask us for that a week after it opens because obviously in order to ramp it to full in 6 months, we need to have the robots on the grid and robots have got a lead time and clients are asked in front of launch, significantly in front of launch, what do you plan for in your first year, and that's what we're building. And so yes, we're trying to add to it versus what they -- their expectations and their requests were at the time. But it's not necessarily at this stage in the development of the business. We don't have several thousand robots sitting in a warehouse that we can just ship to somebody. And particularly because we're just about to manage the launch of the next generation of robot, we didn't want a whole load of the previous generation of robot sitting on a shelf somewhere. So it's not as easy as we'd love it to be. And in the future, it will be much easier for clients to change their minds and just ask us to add robots on much more quickly. The next question is, in the U.K., what was the average delivery fee charged in 2Q '20? How does this compare to precrisis levels? And do you think delivery fees are likely to increase if demand for online grocery remains elevated? Duncan, do you know what the number is because I don't have it on hand.

[Duncan Tatton-Brown]

Yes, Tim, I don't have the number, but our pricing has remained the same. So Smart Pass customers obviously don't pay and about half of our sales are on Smart Pass or more than half the sales are on Smart Pass. Our prices have remained between 0 and GBP 6.99. I think obviously because of the peakness of the demand, the intenseness of the demand, there was probably a little bit more at the higher ends, so underlying, probably a little increase but our policies have remained the same through the whole period.

[Timothy Steiner, CEO & Executive Director]

I think in terms of, Rob, what do I think is going to happen going forwards is that I think that there's been quite a lot of discounting of delivery fees in promotional activity in trying to acquire customers, whether that was us or our competitors. And I think that level is going to decrease. I don't think that clients that are past the acquisition stage are going to find the delivery fees are going to be consistently any higher. I just think there are going to be less short-term discounts given to people, given the elevated demands, which are likely to remain. The third part of the question is what do you estimate online grocer penetration is now in the U.S. than where it was precrisis. So look, we've seen stats indicating many hundreds of percent, 5 or 6x kind of times our volumes in the U.S. And so -- but we know -- I think it's early days, and let's wait and see how much of that is sustainable and what kind of levels of penetration actually means in terms of market share. So I think we've gone from low single digits to either high single digits or early teens in the U.S. But I think that a lot of it will be sustained, and we should just carry on watching. The next question is coming in from Thomas of Berenberg. Four questions. I think we really should set 4 as the limit so no one goes for 5. The first one is in terms of Ocado Solutions fees that were charged to Ocado Retail as a percentage of sales that fell, is this because the installed capacity figure that the fee was charged on was much lower than the amount of sales you eventually managed to squeeze out of the U.K. facilities?Kind of, yes. So I think, Tom, the flattening in the U.K. and the charge rate is predominantly about the maximum, the peak hourly throughput and so obviously it's much more efficient to run a business with equal sales 7 days a week than it is with a peak on a Friday or a Saturday. And therefore, the fees as a percentage of sales fall and I think that there was also some increase in average item prices based on larger pack sizes being bought and stuff like that, that would have slightly helped. But mostly, it's to do with flattening and also to do with more people staying in the U.K. over holiday periods and stuff like that. The second question is there was a new contract with [ fastest ] as well as the 2 new partners. Can you quantify the saving benefits on MHE from these new contracts? I can't quantify the savings benefits, but obviously as we keep scaling and as you start manufacturing at greater scale, we expect benefits us in 2 ways. One, because increased purchasing power and increased efficiency in production, but also because we work to kind of combine components into a single component. So knowing that you're going to build something at great scale allows you to spend more money on the R&D to create a component that's actually cheaper and sustainably cheaper. So for example, in chips and things like that, where you can combine stuff and then turn what's a GBP 1,000 part into GBP 150 part, but you might have GBP 10 million of R&D to do that, it's worth doing that when you know you're going to be producing tens of thousands or hundreds of thousands of robots. Can this help offset the cash flow from ramping facilities faster? Obviously, a cheaper kit means less total outlay and therefore, less cash burn, irrespective of how fast a facility ramps. Third question, regarding the Ocado Zoom facilities, which is only circa 10,000 square foot, to accelerate their rollout, are you able to place any facilities within your existing spokes in the U.K.? Could you also use any of M&S' overspace store estate? So look, our property teams are obviously working and looking for space. Most of our spoke sites, as you can imagine, are fairly full, given the significant increase in volumes. One of the sites that we've taken as a spoke site was always intended to be a Zoom site and that we hope will be our second Zoom site, but we are obviously working on more. In terms of -- and if anyone's space in stores or any old stores -- so look, if the space is right in terms of its kind of floor loading and ceiling heights and stuff and locations, then we could look at using it. It's obviously an ongoing project within the business. Question number four, your partner, Kroger, has announced a smaller mini-CFC in the Great Lakes. Can you clarify that, because the facility is smaller, does that mean more facilities will be constructed than the 20 sites, so the number of facilities that are equivalent to 20 Andover-sized CFCs?Yes, the contract, Tom, is for 20 Andover-sized CFCs. And sorry, and if they were all half the size, then the contract would be for 40. So that's correct. I don't know, Duncan, if you want to add something?

[Duncan Tatton-Brown]

No, I think that's right. But one thing I will add is go back to the one around offsetting the cash burn from ramping the facilities faster. Just as a reminder, the capital raise we did was to ensure that we weren't constrained, and we have the capital to help our partners go faster. I think the long-term lowering of the cost of the MHE is a long-term project. We're making progress, third-generation robot coming onstream now. And I did mention that our engineering costs in Erith fell during the period by 20%. So that's the other big cost of these facilities, the long-term maintenance costs. Both of those are declining. But no, we have the capital to cope even without those declines.

[Timothy Steiner, CEO & Executive Director]

So the next question is from James of Barclays. He's saying there is no doubt an obvious answer that I'm missing, but can you explain why active customer numbers are down by 14%? On the face of it, seems disappointing your customer base has shrunk rather than grown during the pandemic. James, I think if we'd left the site and allowed ourselves to acquire new customers, we would have had an extremely huge increase in active customers because we might have everybody that came into the site do 1 order over the period. And the result of that obviously would be that our existing loyal customers would have been very disappointed. And so what happened was we saw a significant increase in demand from everybody. And that's for the obvious reasons where everybody is staying at home and eating at home rather than at work or at school or eating out for entertainment purposes as well as significant increase in demand, so wash your clothes more often, clean your house more often. And so massive basket size demand and massive customer increase in demand. So when we reported at the last half, we were saying we had something like 850-odd, I can't remember the exact number, active customers, and that we were doing 350-odd or so thousand kind of orders in a peak week, so what you could see then is an active customer on average was shopping once every, say, 2.5 weeks. What we found ourselves in, in the beginning of March was at a period of time where every one of those customers wanted to do more than 1 order a week, and we had demand for over 1 million orders just from the existing customer base, just from the existing customer base that had been active in the last half. We probably had an equal amount of demand from customers that hadn't been active in the last half that had existing registrations on our site. And each of one of them wanted to buy a basket that was between 50% and 100% larger than before. And obviously, we couldn't do 3, 4x demand even before we allowed any new customers on to the site. And so we had to try and work out a system to help prioritize. And we went through our own database to look for evidence of certain types of customers, and we worked with the government and the list that they provided to us and all the other retailers of the people on the shielded vulnerable list. And so we gave them priority and our kind of active customers, those that historically were shopping every week or every -- at least every other week, and we prioritized their access to the shop. So it meant that some customers who might have shopped 1 or 2 times in the previous half and been included in our active customer numbers, we actually, unfortunately had to deny them access to the shop in order to protect the experience for everybody. And it's an indication of the amount of demand that there is out there as opposed to something negative. What we didn't do was push all our existing loyal customers out the shop and sit there with a list just so we could take names and bring some new people to the store. The whole market has brought new people to the industry. And as I said before, we've always historically done extremely well at persuading customers who've moved online to realize that we have the best proposition and the best service at no greater cost than shopping at the biggest grocer in the country and got them to move to us. So yes, it would have been great to have grown customer numbers, but we'd have to be able to grow sales because the existing customers were -- had a much significantly increased spend in the period. Duncan, do you want to add anything?

[Duncan Tatton-Brown]

No, I think you covered it there. The one thing I would say is we have, in the metric active customer numbers, dropped customer numbers. We don't think, long term, we've lost customers. We think there's a massive pent-up demand that we'll be able to serve.

[Timothy Steiner, CEO & Executive Director]

The next question is from James at Peel Hunt. It says on our desire to hire 500 new technologists over the next 12 months, can you give us an idea of your key recruitment channels, graduates, from competitors, et cetera? And where will their focus be once on board?So James, look, we do hire kind of across the experience spectrum, so you can't just bring in graduates because you kind of then overweight that kind of new versus kind of management and people with more experience. Remember that we're not hiring 500 people in Hatfield. We're hiring 500 people over a number of sites: 2 in Poland, 1 in Bulgaria, 1 in Spain as well as both in Hatfield and in Central London. So it's spread amongst a multiple number of sites. And we are growing at a time that many are not. And so actually, the kind of -- the applications coming in for our jobs at the moment is very strong. Where will that focus be once on board? I mean it's innovation across the board, particular focus on front end on web and mobile functionality for OSP, but also focused in other areas such as robotic pick in a small way, as James mentioned, such as on our core robots, so warehouse related, but really across the board, just to try and drive faster change. The second question is, at a time -- in the medium term, when you start to really use robotic picking as well as their efficiency, could you fit a larger grid in a smaller-size CFC without as many people, making it even more efficient? So it will become more efficient. Possibly, the actual active production area may not change in size, but obviously if you have less people in the building, you need less -- the costs come down because you need less parking outside for those people to park while they're at work. You need less food facilities, less changing facilities, you have less management, you need less management facilities. And so there are other facilities. It's not clear to me at this point that the robots themselves -- well, the grid will shrink as a result of using the robots. But we're always working anyway on optimizations that can help us to drive greater efficiency through the size. The third question is, can you give a bit more color around your 3D-printing investment and whether that is there to mean you could insource some level of manufacturing? Or are there more left-field uses? Both. I can't get into too much detail here, but the particular company we've invested in has some unique and patented technology to make very accurate 3D prints with multiple components that could help us to build unusual grippers, for example, to use in robotic pick as an area we're just talking about, but could also help to build components for other parts of our facilities. And to the extent that you might, in the future, [ a way ] of reducing our spares that you have to carry, for example, if you could print them on-site, stuff like that. So there's a whole host of very exciting future uses of advanced 3D printing. Moving on. Nick from Citi. And number one, you talked to your partners wanting to move faster. When might we expect that to come through in your CapEx projections? And specifically for Bristol, Andover and Purfleet, will they come onstream at higher capacity? So I'll do the second part of that question, then I'll hand the first one to Duncan. I mean yes, we're working on ensuring that we can ramp Bristol, Andover and Purfleet faster than we were planning to do so initially. So yes, I would expect them all to ramp faster than their original plans to come onstream at higher capacity and to go harder. Duncan, perhaps you can talk about the CapEx.

[Duncan Tatton-Brown]

Yes, so we are, as Tim mentioned, working with our existing partners internationally about their accelerating the pace of opening. That capital will be modest in this year's overall spend. And therefore, we haven't changed the projection for a modest change. So I think still within the noise, GBP 600 million is about the right figure for this year. And obviously, we've got the capital to enable that in future years.

[Timothy Steiner, CEO & Executive Director]

Number two, has media income declined in percentage terms? Please, could you provide some color?So media income did decline in percentage terms. And really the color behind that just is we delisted some categories like water, for example, in order to be able to deliver to 6,000 more households. Obviously, any media income have been booked in from the branded water suppliers, we obviously couldn't take. There was a lot of canceled promotional activity because the suppliers couldn't supply the products. Their cleaning stuff that was flying off the shelves, you couldn't start launching 2-for-1 offers when you didn't actually have any -- enough product to sell. And so there was a bunch of media activity that had to effectively get canceled and we also have, for a period of time, taken the mobile app off-stream and that also has had some impact there. I'd expect all of that to come back in the medium term. The next one looks like it's an accounting question. So I'm definitely going to hand that one. Much too difficult for me. I'm going to hand that to Duncan.

[Duncan Tatton-Brown]

Well, Nick, I'm pleased that you're asking. Your question was maybe one for next year's accounting seminar, please, could you provide some illustrative figures for the movement in UK Solutions & Logistics? So Nick, I'm pleased you would like another accounting seminar. I'm sure we can help on that. Yes, I mean, I think the fundamental point to make here is the consequence of the Andover fire and our temporary arrangement with Morrisons meant that we forego some of the capacity -- sorry, they forgo some of the capacity in Erith, and we gave them a discount on their Store Pick, which, by the way, we've enabled them to grow very substantially. So the combination of those meant that, actually, fees in Ocado UK Solutions & Logistics were down. The revenue we get from costs grew slower than sales because we're being more efficient. And we had some additional costs. It's quite important for us to complete the works that we need to enable to move our U.K. partners onto our international platform. So the combination of those lowered UK Solutions & Logistics this year. Remember, of course, we'll continue to get exceptional income from the Andover fire for the balance of this year and into next year. So there's a lot more to come there, but that won't get reported in with that.

[Timothy Steiner, CEO & Executive Director]

Thanks, Duncan. Next question comes from [ Christopher ]. Tim, great presentation.Thank you. Ocado Solutions sits at the apex of various structural shifts. It is the greatest risk to the success of Ocado Solutions' execution risks. What are the other main risks to Ocado Solutions? Obviously, there's always execution risk in the business. We work to minimize it constantly. We work to reduce the risks overall that our business has. The addition of the GBP 1 billion of capital was a risk-reduction exercise as well as giving us more capacity to grow faster without having to take more financing risks. There are challenges around travel and executing in an era where it's difficult to move between sites. There are challenges everywhere in the business. But I don't think there is any main risk. We want to see that our clients are successful in their markets with our facilities. We want to build and roll out more facilities. And we believe that there is long-term accelerated global channel shift that will mean that our current and future clients will need more facilities. And so execution clearly is one of the risks, but we think we've got it very much under control. The next question is from Geoff of Morgan Stanley. When do you anticipate Ocado Retail ceasing to be capacity constrained? Will marketing expenditure be reduced to de minimis levels until then?So Geoff, marketing levels have been reduced. And yes, most of the communication with customers of Ocado Retail is kind of e-mail-type communication that is cheap. It's not a "Here's a voucher. Would you come and try us for the first time, please?" at the moment. I'm not sure when that will go back to being something that's kind of required to get people to go over that initial hurdle or sway them to move from their existing supplier to us. When will it be capacity -- cease to be capacity constrained? It's impossible to know because it's just impossible to accurately predict exactly how much demand there's going to be. As we mentioned earlier, we're adding 40% to 50% of incremental capacity onto the platform next year over the course of the year. So there's a lot of theoretical capacity being added, and we will try and ramp it up as quickly as possible and serve as much of the increased demand that we can. And we are actively looking for a number of locations to increase the capacity on a kind of 2- to 5-year time horizon. And we'll keep you updated as we do. The next question is coming in, and I can't see who it is from, but it's asking how will the reduction in VAT affect working capital for the remainder of the year. That's a Duncan question. Will another capital raise be needed in 12 to 18 months to meet CapEx needs, that's a Duncan question. When do you expect the website to open up again? I'm just going to take the last part on the website. We only turned the website off for 2 days. It was a point in time where we were fully booked out for the next 3 weeks. And in fact, the customer baskets had risen so quickly in the preceding 48 hours, that not only had we sold all the slots, but that we needed to actually constrain and say, "You've added enough to your baskets." Otherwise, stock won't be there. We physically can't pick it. So we just said to the customers, the baskets are frozen. And we had 3.5-hour cues to get on to the site to see that you couldn't take -- you couldn't get a delivery slot, they were fully booked. And so we shut the site for 2 days, made some changes to it, reopened it. That was back in March. I think maybe we're referring there to the app. The app has started to go live for some customers at the moment, and we'll continue to roll it out to more as and when we can. And then the 2 questions on VAT and capital, I'll hand over to Duncan.

[Duncan Tatton-Brown]

Yes. So the reduction on VAT will obviously only apply to those items that currently incur a VAT because food is non-VAT-related. Our overall VAT rate is relatively low, call it about 5% of sales. And therefore, any reduction in the headline rate of VAT will be proportionate on that. I think the other thing I would say is that we are a price follower in the industry. So if the reductions in VAT result in lower prices, of course, we will be lowering our prices, and we'll remain price competitive. I don't think there will be any material working capital impact on VAT. And it's worth just saying here, we didn't take the government's offer to delay VAT payments this year. On the capital raise, I think given we have GBP 2.3 billion of the capital, given that we talk about around a GBP 30 million per standard-size CFC opening over a 3-year period, but for new CFCs, it's going to be quite a rate of commitments that we're going to need to see before we need another capital raise. So as I've said, I think, many times before, if we do need another capital raise, it's because the demand for our platform has continued to grow and grow and grow. So it would be good news. But I think materially, we should be fine for quite some time. Tim, back to you.

[Timothy Steiner, CEO & Executive Director]

Okay, so we've got 1 more question that's come in from Dimitrios at Schroder (sic) [ Schroders ]. Can you provide some color on trend performance for the first few weeks of H2? Look, the first few weeks of H2 have looked like the last few weeks of H1. So very strong demand, significantly increased sales based on a flat week and significant demand. I don't know, Duncan, if you want to add anything to that.

[Duncan Tatton-Brown]

No. I think that's right. I think the one thing I would say is obviously this -- the year-on-year sales growth is more driven by the particular week the prior year. So if we go into particularly low periods in the prior year, for example, high periods of holiday, you can get slightly higher percentage growths. But overall, our sales level in absolute terms is very consistent.

[Timothy Steiner, CEO & Executive Director]

Do you expect the improvement in profitability for Retail to continue improving with scale? As scale continues to improve, we'd expect the profitability to improve. And there's 2 different parts to the profitability improvement. One is the increased scale going through the same facilities. So that continues irrespective of the other changes. And then there's the COVID-related changes, which have to do with the flattening of the week and to do with larger basket sizes. I think we expect both of those to continue, although not at the most exaggerated levels that we have seen in the last half. But as scale grows, as we've highlighted over the last couple of years, as we first turned on Andover and then turned on Erith, we were paying for the fixed costs of 2 sites and not yet enjoying the volumes that they could give us. And so as Retail scales, we would expect to see an improvement in profitability. Duncan, do you want to add anything to that or?

[Duncan Tatton-Brown]

Yes. I mean I think the only thing to say is that you'll see a slight sawtooth in our profitability as we add new capacity and coming back to some of the earlier conversations around fixed cost and fees as a percent of sales. As we put in place new capacity, in the very early period when the capacity is not fully utilized, there's a slight reduction as a percent of sales. And then as we utilize the capacity, it gets better. So at periods where we add lots of capacity, it might temporarily drop.

[Timothy Steiner, CEO & Executive Director]

Another question for you, Duncan, what are your goals or aspirations for credit ratings in the medium to long term?

[Duncan Tatton-Brown]

I mean I think we're comfortable with the credit rating where it is today. What I do think is this business, over time, will become a business that is funded much more through debt because the very nature of our relationships at our International Solutions clients are around providing capital assets to support a long-term underlying contract with some stable cash flows. I think it's a perfect debt-financeable business proposition, but not one it's best to debt-finance in the early years of startup. And for our International Solutions business, effectively, that's where we are, in the early years of start-up. So I think credit rating is important for us now. Over time, it will become more important for us, and I wouldn't expect it to move any lower than it is today.

[Timothy Steiner, CEO & Executive Director]

I think we've got time just for 1 last question before we need to wrap up. Not sure who it's coming from, but the question is you've spoken before about an ambition to organize the business be able to build tens of CFCs in any given year. Will you be there for 2023? I think -- are we working on being capable of doing it? Yes. Will we do it in 2023? It's too early to know because we've still got time to take commitments for 2023. And so we'll just have to keep an eye out on expectations. So I'm not sure I can exactly answer that question. I'd just like to say thank you to everybody for joining us this morning. Obviously, because of COVID, this is an unusual way of presenting our results. Hopefully, everybody has appreciated that it's been a very busy period for the business. In all of our divisions, particularly proud of our achievements. And I'd -- so just like to, again, to take the opportunity on behalf of management and all of the stakeholders, to thank all of the teams at Ocado who have been working incredibly hard in very, very challenging circumstances over the last 6 months. And thank you all for your time. Thanks.