Hello, and welcome to today's Ocado Fourth Quarter Trading Update. Throughout this, all participants will be in listen-only mode, and afterwards, of course, there'll be a question-and-answer session. I'll now hand you over to Tim Steiner, David Shriver, and Melanie Smith, who will take you through today's presentation. Gentlemen, please begin.
Thank you, Ian. Good morning, everybody. This is David Shriver, Communications Director with Ocado Group. Welcome to the fourth quarter trading update for Ocado Retail, which, as you all know, is a 50/50 joint venture between Ocado Group and M&S. I'm joined today by Tim Steiner, Chief Executive Officer of Ocado Group and Chairman of Ocado Retail, and Niall McBride, the Chief Financial Officer of Ocado Retail. Tim will begin by reviewing the quarter and the year as a whole. Niall will fill in the details, and then we'll go to questions. Tim, over to you.
Thank you, David. This has been an exceptionally difficult year for everyone in the country, and indeed around the world. Throughout the COVID crisis, we have endeavored to bring the best and most staff and those working in our customer fulfillment centers. They have all worked tirelessly to help feed the nation. Q4 has continued to show strong momentum driven by a combination of three factors: a positive customer response to the M&S offering, secondly, a smooth trading week compared to the peaks and troughs that reflected normal shopping habits pre-COVID, and lastly, strength in basket sizes compared to pre-COVID levels. Customer behavior has continued to normalize in this regard.
The sales growth we have recorded in the period, together with the operational leverage in the business, means that we are again increasing our guidance for full year 2020 Ocado Group EBITDA from over GBP 60 million to over GBP 70 million. To review the performance of the quarter in more detail, I'm going to hand over to Niall.
Thanks, Tim. First, I would like to echo Tim's tribute to our frontline colleagues in Ocado Group and also thank our colleagues in Ocado Retail who have gone above and beyond to deliver for our customers this year. In respect of Q4, Ocado Retail has had another strong quarter up growth. Sales in Q4 grew 35%. This slower rate of growth compared to Q3 reflects the seasonality of the period. The bottom line, however, is that we've continued to operate at capacity over the quarter. Here are some things to think about when examining our Q4 numbers. First, our orders per week continue to grow thanks to the continued strong demand for online grocery in the U.K. and normalization in customer shopping behavior with respect to basket size, creating opportunities for us to offer more slots.
Our ERF CFC is now running at over 130,000 orders per week, as calculated on a pre-COVID basket size, with growth in the quarter weighted towards the back end. Next, our average order size was GBP 133, elevated versus pre-COVID levels, but a further reduction versus the level seen in the last two quarters. During the period, basket sizes increased in response to back-to-school, the M&S switchover, and again, then in response to the additional lockdown measures announced in early November. However, these impacts were less extreme and more short-lived than those seen earlier in the year. The underlying trend remains one of steady normalization. Finally, the year-on-year comparables highlight both the demand we are seeing and the impact of seasonality. Demand for online grocery remains incredibly strong, and as a result, we continue to see a smooth week across our operations.
In this quarter, we compare against seasonally stronger growth in Q4 2019, resulting in lower sales growth. Overall, it's been a strong final quarter to round off the year for Ocado Retail. In addition to managing the challenges of the pandemic, we completed the successful switchover to M&S in September. At the end of our first full quarter with M&S, we are very pleased with the progress. The customer response has surpassed all of our high expectations, and behind the scenes, the level of collaboration is very encouraging. I think if you look at Ocado Retail today, you see the best technology, service, range, and with M&S, the best products. When you bring it all together, we have a highly differentiated proposition for customers, and despite the uncertainties of the world as we head into next year, this gives us great confidence in our prospects.
With that, I'll hand back to Tim.
Thanks, Niall. Before we take questions, I would just like to say that I'm incredibly proud of the agility the business has demonstrated through this challenging time. The platform for growth provided by Ocado Group to currently nine of the world's leading retailers, of whom Ocado Retail is one, has proven up to the challenge. Our flexible and adaptable platform, based on the unique and proprietary technology we have developed over 20 years, has shown that it enables our partners to grow faster and in a sustainably profitable way. As a result, we enter the holiday period with confidence and look ahead to better times for all in 2021. Operator, we'll now take questions.
Thank you. Ladies and gentlemen, if you wish to ask a question, please press Star and then Two on your phone keypad now to enter the queue. If you wish to retract that question, just press Star. There will be a brief pause while the questions are being registered. Our first question is from the line of Andrew Gwynne at Exane BNP Paribas. Please go ahead, sir. Your line is open.
Yeah, good morning, Tim. Hope we are all well. First question, actually, just thinking about trading as we head into next year. Obviously, this period was very heavily capacity-constrained. Sales were actually down sequentially versus Q3. As we think about modeling next year, how should we think about sales development? Is it the most sensible way of just thinking about, Q1, again, going to be quite capacity-constrained, and then hopefully that capacity comes through a bit more fully as the year progresses? Any help there would be greatly received. The second question, and it's good to have you on the call, Tim. Second question is for you, Tim. Obviously, an extraordinary year for retail generally, and obviously groceries in particular. We would expect to see, hopefully, a good number of retailers entering the funnel to become OSP customers.
I'm just wondering if you can talk about how the interest levels have built during the year, and I suppose essentially are any new partnerships to be announced imminently. Thank you.
Niall, do you want to take the first question, and I'll take the second?
Sure. Seal's optimistic in terms of our ability to add capacity to serve all the way through FY2021. We are growing the existing sites, and obviously, as you know, we've got new sites coming on. How exactly the Seals will develop is dependent on the sort of wider environment as well. We all know the vaccine is coming. We don't know when, and we don't quite know what that will mean for more lockdown measures or social distancing. What we strongly believe is that the long-term effects from FY2020 is that the channel shift to online is permanent. We see that continuing into FY2021, but the sort of range of outcomes is hard to predict at this point. Tim?
Are you able just to clarify when the Bristol CFC will open? Would that be an impact in Q2, or is it going to be more like Q3 before we see the?
Bristol should come, it'll be right at the end of Q1, start of Q2. It will be more of a Q2 impact.
Okay. Q1 sales probably similar to Q3, Q4 this year, and then hopefully progresses through the year. Okay. Thank you.
Our next question is, sorry, Tim, do you have any further words to add, or should I just pass it to? Oh, okay. It looks like we've got Tim back.
Sorry, I'm back. I'm back. Sorry, I got bumped off the call.
Tim, back to you.
Sorry. Thank you. Andrew, yeah, we have got increasing interest from around the world in the platform. I know this isn't actually the call for that, but we do go into next year with a lot of interested retailers, as well as a lot of interest in our some existing clients in growing further and faster with us. Obviously, it's been somewhat constrained by the ability to travel, which is starting to open up now. We look forward to next year with some confidence.
Okay. Thanks, guys. Have a good Christmas break.
Thank you.
Thank you, Andrew.
Okay. The next question is over to the line of James Anstead at Barclays. Please go ahead, James. Your line is now open.
Yeah. Good morning. Two questions may slightly kind of follow what Andrew was asking, but zeroing in a bit more detail. The first one is, and I fully understand the point about why percentage growth rates have surged in the third quarter, but I was a bit surprised that absolute sales in this quarter are lower than the previous two, especially given you mentioned increases in capacity tariffs. I wonder if you can explain what I'm missing there. Secondly, the 40% capacity increase you're talking about by the end of next year is perhaps a slightly different way of asking what Andrew was, but can you give a better idea of what that would mean for year-over-year capacity growth for next year as a whole? Can you just confirm that that also takes into account the capacity you're giving back tomorrow for next year?
That 40% is presumably a net number, not a growth number. Thank you.
Niall, let me start, and then you jump in and add to what I'm going to say. I just want to remind you that the very start of this period was the switchover to M&S, which was a very substantial switchover of a large part of our range that accounted for a quarter of our sales. To achieve that successfully, we took volumes down slightly for that first few weeks to deal with all the operational differences, including the significant changes in the rate of sales through different products and where they were in our traditional warehouses. In the Hatfield and Dordon warehouses, they're very, very, very finely tuned with the knowledge that we have. When we moved the ranges over, we saw massive increases in some categories. For example, we saw changes in the mix.
Therefore, to successfully operate, we had to take volumes down very slightly. That accounts for some of the difference in Q4 versus Q3 sales. I'll just take the last part of the question, and I'll let Niall talk about both as well. Yes, the 40% is excluding the Morrisons' growth. The 40% is the Ocado Retail capacity that takes into account the fact that Morrisons is coming back into Erith. The overall growth going through the platform is obviously larger than that. Niall, over to you.
Yeah. Just to tack on to the end of, in terms of the ramp as well in Erith, it was obviously back-weighted because at the start of the quarter, we focused very heavily on making sure that the switchover was successful, which was absolutely the right thing to do. In terms of the 40% capacity, essentially that's a net number, and that's the ultimate capacity. The exact shape of what's going to happen through next year is dependent on how quickly we ramp and demand at the time and the rest of it. It's hard to sort of be precise. Bristol, as you say, is coming on end of Q1, which is about 30,000 orders per week, and then Andover coming back and Purley coming on just at the end of the summer.
It is back-weighted towards the end of the year for those bigger lumps of capacity.
As Niall said, the 40% is what those sites represent when fully operational. I would obviously assume in your modeling that you give them some number of quarters to become fully operational so that you spread that volume out over a period of time as they ramp up and then look for Ocado Retail to announce future sites going forward. Also, as Niall's saying, just make sure that you think about the shape of the week and the basket size and the impact those have as we move towards normalization, which none of us know what the new normal will be, and none of us know when we're going to achieve it. There is that level of uncertainty, of course.
That's very helpful. Thank you.
Our next question is over to the line of Nick Coulter at Citi. Please go ahead, Nick. Your line is now open.
Good morning. Season's greetings to all. Maybe I could come back to this statement. My apologies for flogging a dead horse, but what will be the weighted capacity increase based on a normal basket of 45 inches for next year, just to cut through all of the timing and capacity ramps? What is kind of your best guess or range of expectations for the weighted capacity coming through next year? Thank you.
Nick, I think if you looked at it in a slightly different way and said, "What is your peak day capacity?" Our peak day capacity when we, A, continue growing Erith, B, hand volume back to Morrisons, and C, add the new warehouses in, and D, ramp those new warehouses to full capacity, is 40%. Whether that results in a 40% increase in throughput will depend on the shape of the week that we trade in, but it will result in a 40% bigger peak day once all those things have happened.
Yes. No, I understand that. I'm just asking, once you build in the ramp and say, assuming that you have, I don't know, three normalized quarters, just to get a rough idea to help us. Otherwise, I guess we can make our own assumptions, but when you say end of summer for Purley and Andover, do you mean they're beginning to ramp in the fourth quarter effectively?
Yeah.
Yeah. Okay. Bye.
The way I would probably think about it is if it takes circa a year to ramp the facilities up, you've got they're opening effectively to almost a year from now, over the next two years to scale that growth into our model.
Great. Okay. That is about.
Something along those lines. I'm not saying exactly, but we're not going to be 40% up this time next year because those facilities will only have been open for a matter of weeks.
Got it. No, I understood. Obviously, it's kind of 20 questions going on around this topic. Moving on, if I may, if I could ask about the robotic picking arms at Erith and if they're on track to pick as fast as people or perhaps faster by now. I had a follow-up question on the OSP rollout to Kroger, if I may. Thank you.
The robotic pick arms do continue to make progress and do continue to increase in their operating speeds towards that of the human operators. We do expect that to continue. We expect to make very, very significant progress in the next 24 months, but that will not go in that there is not kind of straight line progress in that sense because there are new technologies to integrate from the acquisitions that we have made. I expect we will see some kind of more lumpy changes in both the amount of the range that we can pick and the speed at which we can pick it.
I just expect at some point in the probably two- to three-year journey to suddenly we'll go from picking a very small amount at a number of robotic pick arms that represent a small amount of volume today to quite a significant amount in a two- to three-year time horizon.
Got it. That's mostly Kindred's expertise coming into play, is it?
It's a combination of some of the software and hardware skills and knowledge and IT of Kindred together with the existing Ocado Robotics team. Because whilst Kindred has, I think it's end of the year with almost 180 live robotic pick cells at clients, they're not picking groceries. We can't just lift them up and fit them into Erith and expect them to work. We've done an enormous amount of work on dealing with the handling characteristics of groceries and the complexity of the more complex part than picking them is placing them. Whereas in most general merchandise, the client's only buying one or two items, and they go together quite easily.
There is a lot in Kindred, and it needs to be. There is a lot of work in terms of merging the best of both effectively to move the platform forward and to be able to pick a wide range and the majority of the throughput of groceries in a two- to three-year time horizon.
Got it. Super. Thank you. Just on Kroger, if I may, and the ISF rollout, very broadly and hypothetically, how should we think about the economics? Is it an annual license fee? Is it volume related? What are the elements of the service?
We would never speak about a specific client contract like that, but the majority of ISF contracts are volume related. They are some form of charge based on the volume that is going through.
Got it. What does that mean? Because obviously, you're bringing the greater proportion of Kroger sales effectively onto the first rung of the Ocado ecosystem at a much faster rate. What advantages does that give you going forward?
It sounds, as you said it, it sounded pretty obvious. We're bringing a larger part of Kroger's volumes onto the platform at a much earlier stage, which, A, has significant revenue opportunities. B, gets our software into the Kroger ecosystem faster and more broadly, and is a very good path for a very significant business with Kroger, both in terms of ISF and in terms of for the warehouses. It is a deepening of our relationship.
Presumably, that improves your visibility in terms of speed to rollout and progressing them through that ecosystem.
I think it will be enormously helpful for both parties in a wide variety of ways, which is why we're both progressing it.
Okay. Super. Thank you so much.
Thank you.
We now go to the line of Maria Laura Derner of Morgan Stanley. Please go ahead. Your line is now open.
Thank you very much for taking my questions. The first one that I wanted to ask is, with respect to Christmas, are there already any type of comments that you can actually make in terms of trading patterns? The second question, with respect to the capacity expansion that you announced this morning for 2021, one of the comments earlier made was, as seen, that it takes a number of quarters to become fully operational. Just wondering, from that standpoint, on average, how many quarters in the past, whenever you've opened a CFC, has this taken? The only reason why I'm asking this is because I'm trying to understand how diluted this could actually be for the profitability mix into 2021. Thank you.
I'm very sorry. Can you repeat the first question?
The first question was, any type of comments with respect to customers' trading patterns into Christmas that you've observed so far?
Okay. Thank you. Niall, if you want to take the first one, yeah.
Yeah. Sure. The Christmas basket is bigger than the normal basket, and that's true in any year. This year, it's equally true. The step up from sort of, in inverted commas, regular week to normal week, Christmas week, isn't disproportionate to what we've seen in previous years. What we have seen is, obviously, M&S being in the basket is a big change this year to last year. Ten out of the top 20 items in the basket are M&S items. The other trend that we're seeing is that people are searching for and putting more non-turkey centerpieces into the basket. Salmon and beef are a greater weight of the centerpiece basket this year, but the turkeys are still number one. Other than that, there's not much I wouldn't call out anything otherwise in the trends for Christmas this year.
I am going to talk about warehouse ramp-ups because they are relevant to Ocado Retail. I do want to try and get this call back focused on Ocado Retail trading as that is the purpose as opposed to a group update. Historically, if we looked at, for example, Erith, when we turned it on, it achieved as much in the first 12 weeks as Andover had done in 12 months. Looking at the historical progress of new robotic sites ramp-up is not the right guide to predict future ones because those first sites were groundbreaking. Erith already today at 130,000 orders that Niall explained has reached is bigger than any of the sites that Niall is turning on next year for Ocado Retail.
The fact that we've achieved that is very significant and means that the infrastructure, the software, etc., are all tested and all operating very well at that scale. Therefore, the reality is that as fast as we plan to grow them in terms of physically adding the infrastructure into the building and hiring people capable of working in the building, we can scale it. We can scale them dramatically faster than anything that we've scaled before. Obviously, you have to plan to do that, and you can't if one of our clients took a facility and said, "My plan is to grow to 10,000 orders in the first six months," you can't suddenly grow to 50,000 orders because you don't have the infrastructure on site to do it.
If you said, "I've got a site here that's capable of doing, say, 35,000 orders, and I want to be able to achieve that in 6 months, 9 months, or 12 months," and the kit's there on the first day, there's no reason that site shouldn't be able to ramp to that volume in any one of those time frames.
Okay. That's helpful. Thank you.
We now go to the line of Sam Bowler at Deutsche Numis. Please go ahead. Your line is now open.
Thank you and good morning. Just a couple on actually the Ocado Retail profitability. It looks like there's somewhere in the region of kind of 600 basis points at least year-on-year growth in EBITDA margins in that business. I'm just wondering, kind of firstly, has supplier income continued to be a drag as we saw in the first half? Secondly, when you think about the step up in profitability, how much of that is kind of operational gearing and underlying efficiencies versus some of the natural benefits from kind of larger basket size, etc.?
Niall, I'll let you take this one.
Yeah. I think in terms of the margin for this year, obviously, the effect of the pandemic period is that you're seeing bigger baskets, which drives routing efficiency, delivery efficiency, and also throughput efficiencies in the warehouse. You're gaining all of that fixed leverage as the order progresses through the system. I would say that looking forward, it's hard to say that that will continue in terms of when things normalize. We'll obviously have smaller baskets, and that sort of boost benefits will disappear. In terms of the supplier income piece, Simon, I think it hasn't been as much of a drag in the second half, but we'll provide a bit more clarity on that when we get to the full year.
Okay. Great. Thank you.
I think I'll just add as well that the shape of week has meant that Niall and the team have utilized their buildings, have utilized the infrastructure they're paying for, and got much more sales out of it. That is another advantage in terms of not just the impact of a singly larger basket size, but the flat week, meaning that more volume's gone through the same fixed asset, has generated fantastic operational leverage. Whilst there will be more of a normalization, we see a very strong trend of persistent demand to work from home. The people who have started working from home who don't want to go back into their offices, it's a big debate within our own business. I think that even some remaining work from home should help our clients, including Ocado Retail, to retain some form of the flattening that we've seen.
We just don't know how much.
We now go to the line of Simon Owen at Credit Suisse. Please go ahead, Simon. Your line is now open.
Good morning. You mentioned that there's been some kind of big shifts in the early days of the M&S switchover. Can you just give us a bit more flavor as to what products and what types of products have sold well and whether there are other products that you've replaced which haven't sold well, where you kind of feel you still need to make some changes to the offer?
Niall, do you want to pick that one up?
Sure. I think we moved 4,000 products into the sites in a single day. Not quite in a single day, but aiming at a single day. It was a huge operation for us to get through. Obviously, when you do that, you create an operational churn, which is what happened, which is why we really focused on that period. I think in terms of the products that people are buying, I think we're definitely seeing very strong pickup in the iconic line. The chicken Kievs or the Plant Kitchen and the Percy Pigs are all selling well. I think what's more important for us and what's more interesting for us is that actually the top M&S sellers are core grocery lines. It's milk and it's tomatoes and it's oranges and it's just that down-the-middle grocery product, which we're delighted with.
It means that customers have bought into M&S for the range, and they're not just buying the sort of the treats at the end of the day. What the customers are telling us is that they love the freshness of the range. The value perception is good. The trend there for the M&S product is very, very strong.
What are the plans in terms of changing or ramping up the range over the next year or so?
We do not have a fixed number for the number of M&S products that we will add. We are working with them very actively on new product development. We absolutely expect to be adding more to the range, but we already carry lots of the range, of the M&S range. There is an additional opportunity on the home and lifestyle front. We have 800-ish lines of that today, and there is probably a bit more we can do there as well.
Great. Thank you very much.
We now go to the line of Tom Davies at Berenberg. Please go ahead.
Morning, all. Just three questions from me. Firstly, in terms of Brexit, what kind of preparations are you guys putting in place? In relation to that, if there was an adverse outcome, would that disrupt some of the manufacturing side of things at FASTES and the capacity installations? Second of all, in terms of the additional capacity, what do you expect the margin drag to be from the great fixed costs in the new facilities versus the offset of increased capacity installation at Erith? Thirdly, can you provide a bit more of an update in terms of Zoom? Should we see any additional capacity rollouts there announced more quickly? How is that business performing?
Let me just answer the Brexit part on what manufacturer and capacity in terms of its impact on capacity for ORL. We would not expect to see any material issue from Brexit in terms of delivering the capacity for ORL in the U.K. here. In terms of planning, and I'll let Niall add a bit, but I mean, as I said from the beginning, you can't store fresh chilled food because you might have some supply chain disruptions. It's not possible for two reasons for the grocery sector to take enormous precautions because, one, the food's fresh and goes off, so you can't store it anyway.
Even on the longer life stuff, if you think about the amount of space that the grocers have while running a very, very lean supply chain, you can't suddenly say, "Let's stick a month's worth of stock in a spare building," because a month's worth of stock for us would be another four times the amount of warehouse space we already have. I'll let Niall if there's anything else he wants to add to that.
No, I think that's exactly right. I mean, ultimately, if there's a disruption, it's going to be a disruption for the whole market. It won't be a cattle-specific issue. We have done as much planning as we can do, and the way we look at it, we've just got to focus on doing our job to feed the nation. Just in terms of I'll pick up maybe on the other couple of points. Margin drag. I think the way that we're looking, obviously, what you've got is an underlying increase in productivity at all the sites. At the mature sites and at Erith, we're looking at increasing productivity, and you can see that coming through in the U.K. numbers. That's one trend. The other trend is when you bring these sites on, you bring them on in a modular fashion.
It is not like you are bringing them all on day one. We use those two effects, I guess, to mitigate against some of the, as you say, bringing on fixed costs at the new sites. We will probably dig into that in a bit more detail when we get to the full year. Zoom, we are very, very pleased with Zoom. It has performed very, very, it is basically performing at capacity and acting all the way through Q4, Q3. We are actively looking for more sites, and we are planning to roll that out as soon as we can bring on the capacity.
Just to follow, come back to Zoom.
Sorry, let me just add something about the question you asked about bringing on the new sites. Obviously, when we brought on Andover and Erith, we brought on a large site and a giant site, which we intended to ramp both of them slowly because they were both new technology that was being rolled out for the first time. Relative to the amount of existing capacity, they represented a very, very significant increase. The increase over the course of this year is much smaller in percentage terms and absolute terms compared to the theoretical amount that we brought on stream when we brought Andover and Erith on stream. We will ramp significantly faster as we discussed earlier. Therefore, it will not have the same impact that adding the fixed cost of Andover and Erith did to the operating margin to the business.
Brilliant. Simon, I just want to go back to your question in terms of delivering the bot capacity for ORL. Does that have any impact on some of your international solutions partners, or have you been able to?
Okay. I'll make the point again. An ORL call, but no, we don't expect Brexit to have an impact on our ability to deliver robots to ORL or any of our other clients.
Brilliant. Thanks a lot.
Thank you.
We now go to the line of James Lockyer at Jefferies. James, please go ahead. Your line is now open.
Thank you. Morni`ng, team. Just a very, very quick one, really. Can you just clarify where Toronto and Paris have ramped up to now, please?
James, just need to remind you, this is a call about Ocado Retail's trading. No, I can't go into our other clients' details. Also, I'm not sure that that will be asked to release anyway unless we ask them in advance to do so.
Okay. Thank you.
We now go to the line of Rob Joyce at Goldman Sachs. Please go ahead. Your line is now open.
Hi, good morning, guys. A couple from me. Just on the smooth demand, quite a lot of importance on that one. Is there any evidence that that smooth demand is changing from here? Linked to that, I presume the excess demand may give some cushion against that changing. Is there any way of quantifying how much excess demand we're seeing for the platform at the moment for ORL, sorry, in the U.K.? Secondly, just on the margins, again, thinking about the impacts of things that may change next year, I think guidance, as Simon said, implying that 600 basis points or so improvement in the second half of the year. Is there any way of thinking about the relative importance of the increased basket size versus the smooth demand and how that may impact things next year? Thank you.
Niall, do you want to give this one a go?
In terms of the smooth demand, I would say, taking Q4 specifically, there were probably three distinct periods. There was a period when we switched over to M&S in the middle, and then there was the second lockdown. Those two, sort of the first and last, obviously, we saw for different reasons, very, very strong demand come through. People buying into M&S for the first time. Obviously, people sort of, I would not call it stockpiling, but stocking up maybe a little bit into the second lockdown. In the middle period, I would say we saw the basket size come down. As we talked about, that normalization of the basket size, which I think is probably a reflection of some of that normalization coming through. In terms of the demand side, we were always able to add limited numbers, but customers where we could.
In terms of sort of leading into your second piece of, are there new customers registering? Yes, absolutely. We could start to work quite quickly with M&S on marketing to the M&S customer base if we thought that they have more capacity or similarly work through some of that backlog that we have as well. I think we absolutely believe there's a lot of demand out there. I would sort of struggle to quantify it, but it's high demand. In terms of the margins, I think that's probably not one for this call because we need to work through the effect of that as well. It's very dependent on the timing of when we bring on capacity versus what happens in the sort of wider environment through FY2021. It's hard to pick apart how it will play through FY2021.
Tim, I don't know if you want to add to that.
Yeah, I was going to say that, no, as you know, they're both positive impacts, the larger basket and the flattening. The larger basket in operational, because of the delivery cost, it's just easier to get we get more pounds through a delivery route. The flatter week, though, gives us the leverage to get a lot more sales through the same fixed infrastructure. They both have a they both contribute quite materially to the improvement in margins as well as the overall growth in the facilities. Because as you recall, the last few years, we've been operating facilities that are built to significantly more volume than we've been doing in them as we've been ramping them up. In terms of the normalization, there is a lot of demand. We do think that the pattern has changed.
With the continued work from home, even if that's only an average of, say, two days a week rather than the five days a week it is for a number of people today, we still think that will give significant opportunity to continue with this flatter pattern that is very advantageous.
Thanks very much.
This is David Shriver. Unfortunately, we're running out of time. I think we have the opportunity for one very quick call.
Right. In that case, final question is over to Victoria Petrova of Credit Suisse. Please go ahead, Victoria. Your line is open.
Thank you very much. Looking at your fourth quarter results, I see that you increased your weekly capacity by 15,000 orders, which is sort of half of Bristol. At the same time, your revenue was down due to significantly lower average order value. Could you please provide some color on month-on-month performance? I'm assuming November should have been significantly stronger given the lockdown, but any comments would be extremely helpful. Is it fair to assume that if you had average basket of the second quarter, which was slightly above GBP 160 per order, your sales in the fourth quarter would have been 20% higher than you are reporting? Or should we look at it differently when we talk about capacity in pound terms? Thank you very much.
Victoria, I'll just take this and start with and see if Niall wants to add anything. The capacity constraint is not in orders. It's in our pick capacity in our warehouses, and it's largely in items. If the basket goes down, we can do more orders. If the basket goes up, we do less orders at any given time. I'm just going to say that the fact that the orders have been half of a Bristol doesn't mean that something's happened to half of Bristol. It's just that the basket size moves means the number of orders either go up or down. Look at the underlying level of items that we're shipping, which is roughly translatable into looking at the cash that we're taking in sales.
As I said before, the quarter started with a slightly reduced volume because of the very large switchover that we did in range that caused a dramatic and some significant swings in what we were selling, which we could not predict exactly beforehand. Therefore, we had to kind of spend a few weeks, particularly in the new warehouses, they react incredibly quickly to it. That is only the Erith volume for Ocado Retail. In Hatfield and Dordon, that are built with older technologies in them, one has to rebalance those warehouses to get the 100% throughput out of them. That is why this quarter did not have higher absolute sales than last quarter because it had some lower volume for those first few weeks. I do not know, Niall, do you want to add anything, or is that answered enough, do you think?
No, perfectly encapsulated.
Thank you for that. Sorry to interrupt you, but unfortunately, we need to jump on another call. If you'll permit me, maybe I can just conclude this call. Thank you, everybody. We report next on the 9th of February with Ocado Group's FY2020 Results. I'm sure I'll be speaking to many of you before then, but for those I don't, I now take this opportunity to wish you and your loved ones a happy and restorative festive season. Thanks very much, and bye for now.
This now concludes today's call. Thank you all very much for attending, and you can now disconnect your lines.