Good morning, everybody. It's Phil Clark here, I look after investor relations here at ASOS. Just very quickly, housekeeping. We're going to start with the presentation, so I'll hand over to José in a minute. Him and Aaron will run through the presentation. We've got some great material in there. I'll then come back and I'll host the Q&A from here. We'll do questions in the room first, and then we'll do questions from the web platform. José?
Well, I don't know why I do that because I have the other one. Morning, everyone. As always, very happy to see you here in our presentation of the first half of this financial year 2026. Let me first tell you a little bit what is going to happen this morning, even though probably you already know. I will start giving you a little bit of the strategic highlights. Aaron will cover the really important thing, that is the numbers. I will go back to give you a little bit of the strategic outlook for the rest of the year, and then we will move to Q&A that I know that this is what everybody wants.
You know I tend to talk a lot, so please don't let me talk too much so that we can really have enough time for Q&A, not like in other occasions. I'll really do my best, but please keep me honest, I think you say in English. With this spirit, let me try to start by giving you the most important messages in one page. That with that, you can probably forget about the rest of the presentation. If I want to summarize everything into one liner, is like we are delivering to our promises. That would be the one liner. Why do I say that? Well, we started three years ago, a big transformation, and what we see is that this is happening, and now we're building on solid pillars. Obviously, I'll give you more about that.
We told you at the beginning of this financial year that we were going to focus on growth, and what we see is that the early signs are positive, and we feel confident about the rest of the year because we have a clear and a detailed plan that is going to take us to where we want. If you want, in a nutshell, we are happy that we see we are delivering. Let me get there step by step. To go there, I'm going to do something that I normally like to do, and probably you guys don't like that I do, but I'm going to do it anyway. That is, I want to go back to which one is our ambition. What is it that we're trying to get here in ASOS?
Three years ago, we really, not decided, but we really stated that what we wanted to be was the best global destination for young fashion lovers. What does it mean? You have heard that 100 times, but I like to repeat it because it helps. It means we bring to consumers the most relevant products. The most relevant product means fashion first, discovery of brands, and this is based on structurally faster way of delivering products, putting together an assortment that is more relevant of our own brands and the best brands on the planet, perfectly blended, which is quite unique. The second thing that makes ASOS different is we want to deliver to our customers the most inspirational shopping experience. That is a shopping experience that is immersive, outfit first, is personalized, and is social.
All this is underpinned by the most efficient operational model, a model that is flexible, that adapts. I'm sure you're going to be worried, and you're going to ask about the Middle East crisis and how this is impacting the businesses. Well, the fact that we have been working on having a more flexible and more dynamic model is helping us to adapt to, for instance, these type of circumstances. What makes ASOS different, or our ambition was to make ASOS different by putting together these three big pillars and delivering on the three big pillars. This is quite an ambition because it's not that easy to do that, as you will appreciate. We were aware three years ago that we were getting on a journey from where we were to become this preferred global fashion destination for young fashion lovers.
That journey had certain steps. We could not jump from where we were to that in one go. First, we really needed to address our legacy issues that you know were our debt and our stock levels, and I feel comfortable we've done it. Second, we needed to transform our commercial model. We needed to go from a commercial model that was not generating. It was not giving us sustainable growth, profitable growth. We wanted to move to a commercial model that would give us sustainable growth, and that was a big transformation from a model that was based on a lot of stock, a lot of promotion, and a lot of performance marketing into a model based on newness, excitement, full price sales, and we feel we are in the right place, and I'll show you why.
Then the third step of that journey is re-engaging with our consumers, focusing towards growth. We have started this year, and as I said before, the early signs are very positive. Let me elaborate a little bit on the first two steps. Three years ago, we told you we wanted to gain much more flexibility with our balance sheet. We want to radically improve our gross margin, and please pay attention to the word radically. We came and said we wanted to take, in the midterm, our gross margin to 50%, and some people thought this guy is crazy, and maybe they were right. We wanted to implement rigor in our cost management and discipline in our stock management. What we have seen over the course of the first half of this year is that our gross margin has improved 330 basis points.
Our profit per order, 30%, our adjusted EBITDA, 51%, and our stock has been reduced by 10%. We are delivering on those targets. Let me take a little bit of a historical perspective here. Our profit per order in the last three years has improved 560 basis points. Our gross margin has improved 560 basis points. Our profit per order has improved 70%. Our EBITDA has improved 14 times, which is quite a number. Our stock has been reduced more than 60%. When I say we are building now on solid pillars, this is what I mean. This is a radical transformation of the underlying business of ASOS, and we feel we have delivered. That's why we feel these first two steps are in the right place, because now the business of ASOS is solid. Now we can build on it.
This is why we said at the beginning of this fiscal year, it's time to focus on growth. The way we are focusing on growth is by focusing on specific areas of growth. We are focused on growth, and growth comes through focus. It's a little bit of playing words. We have focused on two things. Number one is we have focused on the core parts of our business that are womenswear and the U.K. The second thing we have done is that we have focused on rebuilding our customer base, starting from new consumers. Let me give you a little bit more color on these two things. What is it that we have done to focus on the U.K.? Well, a lot of the initiatives we have done have had a U.K. focus. Sorry to repeat focus.
I cannot find a better word in English. Test & React goes women's first. We focus more on Topshop than on Topman. The new partner brands that we have brought are more on the womenswear side. When we expand AFS, we have focused more on stronger womenswear brands. The exclusive collabs we have done, Adidas is only womenswear so far. We have done changes in marketing. We'll talk about that later, is mainly focused on womenswear. We have put a lot of energy in womenswear, and rightly so, because it's the core of our business. The outcome is, in the first half of the year, womenswear performed with a -2% year-on-year, improving 10 points versus the end of last fiscal year. In the month of March or in the beginning of Q3, womenswear is on growth. It's clearly having an impact on the performance on womenswear.
If we go to the U.K., well, we have focused our marketing efforts in the U.K. We have increased the budget more in the U.K. Our out-of-home campaigns have taken place in the U.K. We have launched our loyalty program first in the U.K. We have launched our exchanges on our returns policy first in the U.K. The U.K. has received more focus than other geographies. The impact, H1 U.K. performed 4 points better than the rest of the company during the beginning of Q3. The U.K. is still ahead of the company and is better than what it was at the end of fiscal. It is working. This focus is working, is delivering results. Obviously, our opportunity is to take all these learnings into the rest of the business, into other geographies, into other segments, and obviously, we are going to do it.
The second big focus of these six months have been rebuilding our customer base. This is the evolution of our total customer base during that journey that I took you before briefly. Obviously, during the first two parts of the journey, the customer base was being damaged. We were reducing stock, we were reducing promotion, we were reducing performance marketing. We were doing a lot of things that were not necessarily pushing the customer base, and we knew it. We were expecting it. It's only now that we have gone into phase three, when we see this customer base stabilizing, because this is now when we feel we have the solid pillars to build on and really talk to these consumers. This stabilization is coming from two fronts.
On one hand, our churn is improving, so it's green across, or our retention is improving, whatever you want to call it, because churn retention is pretty much the same thing one way or another. It's been improving significantly during the last more than six months, but certainly the last six months. Second, we are capturing more new consumers. New consumers was in growth over H1 our top four markets. New consumers is in more growth over the beginning of Q3 in all of the group. It continues accelerating. Remember, for us, it's very important to attract these new consumers because we have implemented a big change in our commercial model. We were expecting that we were going to change a certain part of our consumer base. The fact that new consumers are coming means they are connecting with our value proposition.
They are happy with the product. They are happy with the shopping experience. They want to buy with us. What is really behind all this? A lot of actions. Now get ready because I'm going to go one by one. No, don't you worry. I promised I was going to try to be brief, so I'm going to concentrate on a few of them. There are a lot of things happening in ASOS, as always, and rightly so, by the way, because this is, as you know, retail is detail, so it requires a lot of changes and small changes at the same time. Let me focus on some of the big changes. When we talk about the product, so you remember, we always go product, shopping experience, and operational model.
When we're talking about the product, we have brought a lot of things that we have already started in the past, and you will have heard me talking about it for many times. When we talk about our own brands, Test & React continues growing. It has grown double-digit over H1. It's 6 points more participation in the sales of own brands. It's going in the right direction. Super happy with that. We have accelerated lead times by 30 days. We have relaunched our 4505. That is our own brand for sports, 20% growth year-on-year. Overall, the autumn-winter collection has improved the sell-through by 5 points, reaching the highest sell-through ever in ASOS. It is working. It is working. When we go to our partner brands, we have onboarded 60 new brands during H1.
That takes us to a position where during the course of the last 18 months, we have renewed 20% of the brands, which is very relevant. With 50 of these brands, we have exclusive collabs or exclusive products, and I will talk a little bit about that now. We are pushing our Flexible Fulfilment. Both together are more than 20% of the sales of our third-party brands. More than 150 brands with partner fulfils. We have launched ASOS Fulfilment Services. It's working very well with Inditex, is growing, and on full price, so it is really working. I just wanted to give you a little bit of color on one initiative that is one of our collabs, and probably you're aware that last week we launched drop three of our collab with Adidas.
That was a special moment because there were quite a few first moments during last week. It was the very first time that this collab was available for our American consumers. It was the very first time that this collab was going in real life. We took over the Adidas store in Oxford Street. Actually, it was quite interesting because you could see ASOS on both ends of Oxford Street. You could walk by John Lewis and see Topshop in the windows, and then you could get to the store of Adidas and see all the windows taken over by the images of the collab, which was very exciting. The result was really amazing. Sold out in three days. It's 80% sell-through, which is pretty much leftovers. The best sellers is north of 90% sell-through in three days, even though we have doubled the purchase.
It's not only that we're selling the collab, the sales of Adidas overall, the halo effect, pushed them 43% up year-on-year. Not because of the collab, it's like the rest of the stuff, we sold 43% more than last year. If we're talking about the U.S., that was the first time, as I said before, the U.S. almost doubled its participation in the sales of ASOS, and 50% of the customers that bought the collab were new customers. These kind of initiatives have an impact on our sales, have an impact on our partners, and attract new consumers. That is exactly what we're looking for. I think it's a very good and a very powerful tool that we're using. Second, our shopping experience. We have been investing a lot in our shopping experience and just wanted to highlight three things.
We have been investing more in marketing. We have increased our investment in marketing by 50 basis points. At the same time, we have reduced the cost of the visits acquired, which means we're doing more with more efficiency. We have launched The Heart, which is on a specific initiative that I will explain to you briefly right now. We have expanded loyalty. It was in the U.K., now it's already 60% of our customers in the U.K., which is more than 3.5 million people. We launched Germany and the U.S., where already 20% of our customers have joined the initiative. Remember, we don't give any economic incentive to join, so it's quite a number. What we're seeing is that the customers that join loyalty grow with loyalty. Our top tiers grow 9% and 20% respectively, versus the customers that don't join loyalty.
It is having an impact and we will obviously push to roll it out even more and to grow it even more. We have invested a lot in our shopping experience, and I will also give you a little bit of detail about that right now. Let me tell you about The Heart, not the best name, you can blame that on me. The Heart is an initiative that we do every month where we pick the best of our brands, the best of our partner brands, and we put an emphasis on how we're going to communicate that to consumers and how we're buying it. That generates a very specific and structured plan that comes to market every month. It's systematic. It's not improvisation, it's systematic.
The outcome, the sales of these items, we call the rate of sale, which is the number of units we sell on average every week, has doubled versus the rest of the items. Obviously it's working, and we will continue doing it and roll it out to not only womenswear but menswear. When we're talking about the shopping experience, what we have done first is that we have focused on our app. In this case, on our iOS app. Why on our app? Because first we know that our app consumers have a value that is twice as big as the non-app consumers. Obviously the app was important. We have put a lot of emphasis in growing the downloads of the app.
What we have seen is that over the course of the last months, we have gone from a negative trend versus last year to growing more than 30% on app downloads year-over-year. It's clearly working, and it's not only that they are downloading more, they are using it more, but I'll get to that now. The second thing that is important to say in this step change in our shopping experience is that all the changes we have done have followed a systematic approach. It is not the outcome of a hunch of our intuition, it is the outcome of a system. This system is based on a drastic acceleration on the velocity of our tech developments. During the course of H1, we have more than doubled the amount of tests we have put in the market.
At the same time, we have increased the hit ratio by 16 points. This is supported by, as I said, an acceleration of the velocity, and is supported by a radical increase on the quality of the measurement and the speed of the measurement of these experiments. This, obviously. The outcome of that is we have implemented, not tested, 50 new features in our app. I will not run you through the 50. I promise I will give you time for Q&A, but just let me highlight a few of them. Virtual try-on. Consumers can, in our iOS app now, take a picture of themselves or choose a model, choose an item they want, and see how it looks on them. That you have seen. Some other places have also launched it. It is available in our app and is working very well. Ways to style.
When a consumer wants to buy a specific item, we are proposing this consumer different ways to style this item that increases the value of this item for this consumer because it has more usability. You want, last but not least, out of the three that I am selecting here, is this trending section. This is a more immersive way to navigate our assortment and to feel it that is much more engaging. At the end, remember, what we're trying to do is to go from a more transactional into a much more experiential way of buying. The outcome is a completely reinvented shopping experience. We have gone from, if you want, a more transactional app into a much more experiential app. We've seen that consumers like it.
The ratio of daily users over monthly users, which pretty much means how many times the consumers are coming every month, has increased 60 basis points. Even more important, the consumers that come are buying 9% more than before. It is having the right impact. This is what we wanted. Last but not least, always important to talk about our efficiency and our operational model. Here we have also implemented a lot of improvements. Again, rigorous in the costs. Our friends of supply chain have gone with another 150 basis points of improvement. We have reduced the impact of returns by 160 basis points, increasing our profit per order overall by 30%. Quite remarkable. We have improved our customer proposition. Cut-off next-day delivery in the U.K. have gone from 10 to 11. We have introduced changes like exchanges.
Last but not least, we have started to really use AI extensively in a lot of the things we do. Let me give you a little bit of color. During the last six months, we have really gone deep in AI on productivity. AI can be applied on many more other areas, and we will talk about that later, but during the last six months, we have gone deep on productivity. Let me illustrate how. November last year, we signed, or we extended, if you want, our strategic alliance with Microsoft. Obviously, if you want to go into AI, it's important to have a strong partner, and in our case, it's Microsoft. As a result of that, we have rolled out the use of Copilot from 10% of the organization to 90%. That's pretty much everyone. People have started to use it, have started to build agents.
We have saved 35,000 hours. It's only the beginning, but it's quite a number. We have implemented a customer care agent, AI powered. The outcome of that is increase in customer satisfaction, increase in, let me call it, containment rate, reduce of cost per contact by 90%. Today, 15% of our software, 20% of our buying decisions are already executed or delivered or deployed, whatever, by AI agents. That gives you an idea with these three numbers of how deep we're going on AI on these topics. If you want a little bit of a summary of that is, we promised we were going to transform the business. I feel the business is now building on solid pillars. We promised we were going to focus on growth on H1. I see that the early signs are positive. Let me save the last point for later.
Now I'm going to hand over to Aaron, who will give you the really good stuff.
Thank you, José. It's become even more apparent over the last 20 minutes how many of my great stats José has already shared. Sorry if I can't give you any new ones, but I'll try. I think there's a few nice nuggets in there. Welcome, everyone. I'm going to talk through our financial performance for H1, which represents a really strong progression and evidence of three clear things. Dramatically improved business financially is what we have now at ASOS, a consistent track record of delivering on our promises, and a structurally improved economic model. When we return to growth, the business can meaningfully turn that into cash and profitability. You'll have seen, I'm sure you've all read the RNS, you'd have seen many of these, but I want to call out a few important parts.
GMV revenue remained negative, as José has talked about, but broadly in line with expectations. The 5 percentage point gap between those two metrics is based on the success of our Flexible Fulfilment models. Behind this, though, is a number of positives, and that's a really important point. As we continue to say, this is about sustainable, profitable growth. Crucially, what we've done through the course of H1 and into the early part of H2 is continue the sequential improvement in our GMV trajectory. Profitability improved by 51%, underpinned by gross margin performance, another 330 basis points up, adding to the progression that we've seen over the last few years. Adjusted cost to serve has been excellent.
Our revenue deleverage masks the 200 basis points of underlying efficiency gains, which I'll talk about a little more in a future slide, but also includes reinvestment back through marketing of 50 basis points. From a cash perspective, we're in line with our expectations at a GBP 93 million outflow. This reflects our normal seasonal working capital cycle, and within that, inventory remains tightly controlled. We continue to improve the health of our stock, supported by the ongoing benefits of the commercial model and the rollout of those Flexible Fulfilment models that we talked about. Geo performance reflects where we focused our energy, again, as José has pointed out.
The U.K., our core market, outperformed the group, and this follows an enhanced set of marketing campaigns and exclusive collaborations like the Adidas collaboration that José showed some photos of, and advancements in our loyalty program. Also important to note, the trajectory in our other markets has improved. Overall, as mentioned, we've improved sequentially through H1 and into Q3. Another important point to note is the customer database. We said at year-end, the importance of rebuilding our customer database through new customer acquisition, and you'll note that only small marginal declines in the customer database since year-end, particularly in our top four markets. AOV continues to improve, principally through reduced return rates, and this is all supporting our PPO increase of 30% year-over-year. With regards to gross margin, the rebuild continues, and we're really pleased with our performance here.
Since we've begun the transformation, 600 basis points improvement, and this is structural and sustainable. There's no shortcuts here. This is for the long term and for us to take forward into H2 and beyond. There's two core drivers of that improvement, not just in H1, but across the last few years. The reset of our operating model, which includes a reset of our buying processes and reduced discounting, that improves our buying margin and also our exit margin through increased full price mix. It also includes the structural impact of the increase in our Flexible Fulfilment mix, where we have no inventory risk and improved proposition for consumers. We're fast approaching our midterm targets on gross margin, which will offer a really solid level of profitability and give us optionality to reinvest back into the bottom line or into growth.
Our third strategic pillar, the efficient operating model, is fully embedded in our ways of working and processes. We've seen another material improvement continuing into H1. I want to be clear here, though, this is not us saving our way to success. It's quite the opposite. This is about creating fuel, generating efficiency to create investment fuel for growth. While the overall cost to serve has increased by 90 basis points, half of this, as talked about, was a deliberate reinvestment back into marketing for growth, and half the structural increase from the improvement, or sorry, implementation, you might say, of our AFS model. Underlying efficiencies within that totaled 200 basis points, which includes the benefits from our carrier renegotiations in distribution and the continued benefits in our warehouse operation, including it within the mothballing of our Atlanta site.
We'll continue to drive these efficiency gains into H2 and beyond, which will include the benefits from a reduced headcount of 16% as we exit H1 versus the same period last year. These will further improve our options to selectively reinvest back into growth, again, in H2 and beyond. Pulling these dynamics together, the impact's been material. Delivered a 51% increase in our EBITDA to GBP 64 million. The volume declines, as we talked about, more than offset by the efficiencies, both through gross margin and cost to serve, and also facilitated reinvestment back into marketing and growth-driving opportunities. Turning to free cash flow, a GBP 93 million outflow, which reflects the normal seasonal nature of our working capital. The outflow broadly mirrors last year, with increased levels of EBITDA being offset by the interest on the 2028 convertible bond, which wasn't paid in H1 last year.
CapEx remains tightly controlled and other impacts within our free cash flow bridge in line with expectation. Continued discipline on CapEx and working capital as we move into H2 will deliver the inflow in H2 that will enable us to meet our full year broadly neutral guidance. One of our financial priorities under the transformation was to fix the balance sheet. We've taken two strong steps forward in the first half of FY 2026. First, we completed the refinancing in November, which reduced our interest costs and extended the maturity out to 2030. Crucially, this also provides an additional GBP 87.5 million of liquidity through the delayed draw term loan, which remains undrawn at the end of the half. After period end, we repaid GBP 74 million of the remaining 2026 convertible bond from our cash reserves.
The net debt bridge shows a GBP 90 million increase in net debt, which primarily relates to the non-cash interest accretion in relation to the 2028 bond. I expect net debt to improve, the absolute levels of net debt to improve in H2 in line with our free cash flow expectations in H2. I want to talk a little bit more about current trading before I hand it back to José. The momentum in H1 has continued into H2, and the list of green shoots is ever expanding. Our women's wear business has returned to growth. New customers across the group are up high single digits year-on-year for the first time in years, and cost and gross margin remains well disciplined. Combined, these improving trends underpin our FY 2026 guidance, which I'll talk to on the next slide. Our guidance for FY 2026 and the midterm remains unchanged.
H1's been really strong, reflective of our locked-in efficiencies that we've delivered over the last 12 months, and we expect GMV to continue to improve throughout the course of H2. In the midterm, our focus remains on growth as we approach our midterm targets for gross margin and EBITDA. To close, we feel really confident about our progress towards sustainable, profitable growth. We've built a significantly more profitable model, and that will enable us to capitalize on free cash flow and profit from that growth. Crucially, we've got a fixed cost infrastructure and a warehouse capacity that can facilitate growth without further reinvestment. With that, I'll hand back to José to talk about H2 and beyond.
Thank you.
Yeah, we're almost by the schedule. I'm going to try to get it there by the schedule. Let me talk about what is coming for the next six months a little bit. As I have said before, there has been a lot going on in ASOS, and I didn't torture you going through one by one. I will not do it now. Our plan for the second half, first of all, is obviously to continue building and to continue getting results from what we have built and achieved. A lot of the things you will see here, and you can probably read later, have only come to life towards the end of the half, so we still have to see some good results coming from that.
I just wanted to highlight four areas here where we are going to double down on H2, which are on the product side, on the marketing side, on the shopping experience, and I will want to make a little bit of a bigger stop, not maybe on AI, and how AI is becoming an enabler and an accelerator for our business. Let me go fast on the first three points. On product. What is it that we're doing on product? Nothing new. We continue the journey we started a few years ago, faster, better, stronger. We want to take Test & React closer to the 30% target. We will push Flexible Fulfilment beyond the 20% that they already are. We will keep on bringing new brands, both Partner Fulfils and ASOS Fulfilment Services. We continue pushing our exclusive collabs.
Adidas Drop 4 and Drop 5, I think, will take place during H2. We have signed a new collab. I'm sorry, I'm sure I'm going to mispronounce that, with Barbour. Barbour. Sorry. Yeah. It's a difficult one. We have not talked a lot about quality, but quality is a big part of our plan. We are doing a lot of initiatives because we want to bring the right product with the right quality, with the right price, at the right time for our consumers. There is a lot happening in terms of quality from making assurances to consumers about the quality of their purchases, to standardizing how they can see sizing across our own brands and third-party brands, and so on and so forth. There is a lot going on quality to improving the quality, obviously, the intrinsic quality of our products.
On our marketing model, we will continue investing in marketing to the level we have seen. We have increased 50 basis points over last year. We will continue investing because we see that our investment is working, is bringing us good results, not only on the performance marketing side, that, if you want, is a little bit of shelf space type of approach, but also on the brand marketing side. We have had some incipient campaigns that have worked well. We're going to double down on sharpening our brand, on improving our channel mix, and on improving the quality of the people we cooperate with. On the shopping experience, you have seen the good results of our loyalty program. We will continue rolling it out in the geographies where it is on to more geographies.
We are going to roll out the improvements in our iOS app to Android and to web, and we are going to continue improving our iOS app. If we have landed 50 features in H1, we want to land at least 40 in H2, some of them like virtual try-on that is available on a subset of the products. We're going to roll it out to many more products. We are working on our own personalization. This has only started. This idea of this has only started is probably the perfect segue into the last point I wanted to make with you, that is how AI is becoming an enabler and an accelerator. As I told you before, during H1, we have worked a lot on the productivity side of AI.
We see AI doing much more for ASOS, and we have started to work pretty much on how this underpins our whole model. These are some illustrations from productivity to how do we manage complex tasks? How do we forecast future sales? We are now using AI tools to forecast future sales. How do we rethink our organization and move into a frontier organization, a hybrid organization? We are doing that. There is one area where I wanted to share with you a little bit of some of the things we are doing that I personally find especially interesting. That is on the customer proposition. I think AI is by far the biggest opportunity in the market for ASOS and everybody else to completely bring a revolution to customer experience, to make customer experience online much more engaging.
Let me tell you how we are approaching one of these areas, that is what we call ultra-personalization. What do I mean or what do we mean with ultra-personalization? Let me try to explain it in simple words. Ultra-personalization means that every time a consumer walks into an ASOS store, let me put it in these physical terms, she's going to walk into a different store every time she comes to ASOS. Every different consumer will be exposed to a different store every time they come to ASOS. This is a quite ambitious type of vision. Obviously, in order to deliver that, we need two things. We need a very strong personalization engine that understands not only the history of the consumer, the preferences of the consumer, the historical preference of the consumer.
It also needs to understand the in-session preferences, because we know that our consumers are not behaving always the same. They are very, I don't know, the word is not volatile, but they are very flexible, probably that's the best word. It also understands what is coming, where the trends are coming. It's not such an easy way of personalization. We're working on that, but I'm not going to talk about that today. It also requires an unlimited capacity to create content. If we want to expose every consumer to a different store, we have to be able to produce an almost unlimited amount of content that adapt to these consumers. This is what we call AI Studios. What I want to share with you now is how we have started to work on this idea.
This is now more than an idea. In ASOS, we have tested that in The Heart of the last month, where more than 90% of the products of The Heart, the assets were created with this logic. Our ambition is to roll it out to everything we do to completely bring a revolution to the way we create content. What is AI Studios and what is the logic? This is a very complicated chart, so probably it's better you don't care. You don't care about the chart. In today's world, to make a picture, we need to put the model, the clothing, and the photographer, all of them in the same place at the same time, which takes time, cost, effort, and brings lack of flexibility. What we are doing now is that we're taking a picture of the product. We're taking a picture of the model.
We have the photographer somewhere else, and this is coming together into a model, which obviously brings a lot of flexibility and accelerates our time to market. Once we have done that, we're using AI to structure the flat pictures, as you can see here, this type of flat pictures were structured by AI so that the consumer can see clearly which are the different elements of an outfit. We're going to give the consumers the possibility or the personalization engine to mix and match these items in a different way. We will generate the picture on a model looking absolutely perfect, with the possibility to select the model, the background, the hair, the makeup, you name it. That means that we have, one, absolute flexibility, and we can generate unlimited content. As I told you before, we have already tested that.
The Heart of April, we are in April, the heart of this month, more than 90% of the products follow that logic. Our ambition is to roll it out to everything we do. That gives you an idea. It's just an example, but that gives you an idea how AI enables this absolutely superior customer experience that we are convinced is going to bring great growth to online and great growth to ASOS. With that note and a bit at the limit on the timing that I had, you guys didn't stop me. Sorry. Anyway, that's okay. I will not take it personal. Just let me finish by reminding what I want you to take away with you. We are delivering. We have delivered a major transformation of this business model and now we're building on solid pillars.
We are focused on growth and we see early signs of positive growth, and we are confident we have a solid and concrete plan for H2 that will help us deliver our promises for this financial year. With that, I'm going to hand over to Phil so that we get into questions. Thank you.
Great. Thanks, José. Thanks, Aaron. We'll start with questions in the room. We'll do them one at a time. I know what you guys are like, so we are going to be very firm on three questions, and that's including the part A, part B. Anne put her hand up first. Anne.
Thanks. It's Anne Critchlow from Berenberg, and I will ask three, please. First of all, what percentage of third-party product is exclusive to ASOS now? The second question is about Test & React. Just wondering how relevant you feel that will be to menswear and to what extent you plan to implement it in menswear. The third question on marketing. Just thinking about return on advertising spend from whatever you're doing compared to, say, return on investment on promotions and what your thinking is given a very deal-hungry consumer out there. Thanks.
Okay. Let me take it.
You take two.
I take the first two. You take the third? Okay. Well, I have this one. Sorry, I completely forgot I have this one. Let me take the first one. What percentage of our assortment is exclusive from third-party brands? Sorry, Anne, I don't know. What I can tell you is that 50 of these brands, more or less 600 brands, we have exclusive products. If you want at least almost 10% of our brands, we have exclusive products with them. It's not 10% of our assortment, at least 10%. I think that the most relevant thing here or a very relevant thing here is the halo effect these items can generate. I think what I share with Adidas is a good example.
The best, being honest, but we see that when we bring relevant exclusive collabs, that generates a halo effect on the whole of the brand and on the whole of ASOS. Not only on that. On Test & React and menswear, we do have Test & React in menswear today. It's not the same level of focus that we have on womenswear. To what extent is this relevant? It is relevant. To be honest, I don't think it's going to be as relevant as it is for womenswear. We see that our menswear consumer is having different motivations than our womenswear consumer. Our womenswear consumer is very motivated by trend. Obviously, Test & React is an amazing tool to react to trend. Our menswear consumer is less motivated by trend.
It's not that he's not, but less motivated by trend, but it's also useful to react to color, to react to fabrics, to react to shapes. It is important. It will not be as relevant as it is for womenswear, almost certainly. You want to take the marketing one?
Sure. Yeah. Thank you, Anne.
I suppose the first thing to call out when you step back from it is we do still promote. We do still have offers for consumers, particularly a small subset of consumers who are still profitable, who are searching for discounts. We've changed our approach, where this isn't our main hook for consumers. We're focused on making sure we can give customers a great experience, better quality product. Our investment strategy doesn't just focus on promotions, and it doesn't just focus on marketing. We also think about how we can invest in quality, how can we continue to invest in the experience? All of these factors together, we're always reviewing the return on investment from all these factors, and we'll find the right balance between them.
What we won't do is go back to a heavy promotional model that is creating this transactional relationship with our consumers. You've seen from all the stuff that José shared, we're moving away from that. We have a multitude of different opportunities to invest, reinvest back into the customer, which still includes some promotions, but it's much more focused on experience.
Great. We'll do John next, and then could we get a second mic, and we'll do Yash afterwards? Thanks.
Thank you very much. John Stevenson at Peel Hunt. I'll go through as well, please. First question on the current cohort. Can we talk about how the cohort, I guess, is behaving versus previous cohorts and your expectations? Are we seeing more green shoots, if you like, within the new cohort? Second question, just in terms of the U.K. versus the overseas market. Is that gap driven by, is it marketing? Is it localization of the offer? What's the main difference in terms of performance? Is it just focus? I guess, and when do you start to apply that focus to close the gap a little bit? Finally, just on cost efficiency. Huge progression. To what extent is it just an annualization from here, or actually, is there still significant ongoing efficiency to come out?
Same again?
Same again.
Two in one, yeah.
Yeah. Looks like.
Sorry John, I'm assuming when you say cohort, you refer to customers cohort, not stock cohort, right?
Yes.
Okay. Just to clarify. Very early to give you a very deep analysis of the behavior of the new customer cohort. What we are seeing is that, especially over the course of the last months, their, how can I call that? Their repeat rate is improving. It's not only that we're getting more new customers, it's that these new customers are better. It's very early still to tell you if they're going to increase their frequency, because it takes a little bit longer to understand that. Their frequency, their value as consumers, and so on and so forth. It is still early. Age-wise, they are a little bit younger, but not significantly younger. We've been pretty much on this average age of 29 for a while. It's not really changing that.
Obviously, we're getting more from the top four, but especially, obviously, the U.K., but this is where we have put our focus. I don't really think it qualifies that our current offer, it resonates better with certain geographies. It resonates better with the geographies where we have put more focus, if that makes sense. On U.K. versus other markets. Well, I think you have almost answered yourself, John. It is what we have done. We have focused on the U.K. first and womenswear. That's, as I shared before, we expect that as we start rolling out this focus to other markets, it will have an impact. It's important to say that there is a difference between the role we play in the U.K. and the role we play in other markets.
In the U.K., we are, with all the being as humble as possible, but we are one of the leaders in our segment, and we play that role. We are more of a wider type of option for our consumers. A lot of people buy ASOS. As we get out of the U.K., we become more of a specialist in a segment. We talk more to the fashion people. We see that in how our consumers buy, and that's why, for instance, our American consumers. It's not that our American consumers are more fashion forward than our British consumers. It's that in the U.S., we have more fashion forward in our mix than in the U.K., because in the U.K., a lot of people buy from ASOS. We take that into account.
When rolling out, for instance, the strategy we have in the U.S. is not exactly the same as strategy we have in the U.K. We are also seeing good results of that strategy in the U.S., and probably a lot of these learnings can be more applicable to other markets where we might play a role more similar to the one we play in the U.S. than to the one we play in the U.K. That is probably quite different in that sense. The possibility to roll out what we have learned in womenswear or what we are learning in the U.K. and the U.S. is absolutely there, and we are confident that is going to give us future growth.
Thank you, John. Yes. You're right, some of the benefits that we've talked about will annualize. We talked a lot at year-end about the distribution carrier renegotiations, which will annualize towards the end of this year, and obviously the Atlanta closure. They've already been replaced to some degree, replaced is the right term, with new efficiencies. As I mentioned, we exited half one with 16% less central headcount. We've also improved our efficiencies within the warehouse operations, particularly in Berlin. There's a number of these efficiencies that will continue to roll going forward. Look, we're always searching for this. Our focus is on growth, but if there are opportunities there, and we believe there are still more, we'll continue to focus on them. What we'd say, it's not just in cost as well. There's further opportunities in gross margin.
We're continually refining our operating model, and across both of these, we expect to be able to create more fuel going forward to be able to reinvest back into consumer growth.
Okay. Yash. You got one, sorry. Hold on to it, John, for a sec.
Hi. Yashraj Rajani, UBS. Thank you for taking my questions. I've got three as well, please. The first one is on the medium-term CapEx. Can you shed some light on what gives you comfort in that 3%-4% of sales number? Where's your capacity utilization? How are you thinking about spend on tech versus fulfillment, and are you comfortable with your balance sheet structure for that? The second one is just a follow-up on U.K. versus Europe, right? Basis the comments you made, can you give us an idea in terms of what your average cost of customer acquisition is in the U.K. versus Europe, and how do you expect that to change with the advent of AI? The third one, just a technical one on the Middle East, please.
Can you give us an idea till what point are you hedged on the likes of freight, fabric prices, and other external factors? Thank you.
How do you want to do that? You want to take the first?
Yes. Sure. Yes, thank you for the question. We're really comfortable with mid-term guidance on CapEx. I guess, if I go back historically, we spent, historically, a lot of CapEx on growing our infrastructure. As I mentioned in the presentation, we have an infrastructure now that does have capacity for us to grow into. I don't expect, in the mid-term, for us to require investment in our infrastructure to grow. The vast majority of our CapEx within that 3%-4% guidance is on tech development spend. You've seen, as José presented, the velocity improvements that we're making already, and they'll continue. The investment into that capitalized development spend will continue to drive benefits and will continue to improve the effectiveness of that spend.
On U.K. versus EU, to be honest, I don't know what is the average cost of customer acquisition. Being very honest, I cannot tell you. Let me comment a little bit on that. What we have done this year is we have increased our marketing spend and reduced the cost, which obviously is very easy to increase your marketing spend, but to reduce the cost means we have found better ways to do it. There is a lot of that is coming from the performance marketing side of things, how we're doing things. We're doing better. As I briefly mentioned before, the way we're thinking about that is almost like shelf space. That is a way to be present in front of the consumers that were already looking to buy and probably buying on ASOS or something similar to ASOS.
There is another side of it that is the brand side of it, where we still have a lot to learn. We have done campaigns, we have improved significantly, but connecting with these consumers that are not so close to ASOS is more difficult. We have done a lot of things. I didn't get into details because, obviously, I really wanted to give you a possibility to ask, but let me give you a little bit of color here. For instance, one of the things we have done, we have done a lot of in real life events, pop-ups, both in the U.S. and in the U.K. To give you an example, the last pop-up we did in New York increased the sales of the whole New York area by 4%. Obviously, just by that, more than pays for the pop-up.
We are seeing a lot of opportunities to improve in that sense. We see great opportunities there. We need to put ASOS more in front of the eyes of our consumers. How AI is going to have an impact on that? I think the impact is massive. From the quality and the quantity of our assets, to the quality and the quantity of the briefing of our campaigns, to the capacity to attract of our own. It's a total revolution, as I said before. I think it's not going to be similar to anything of what we're doing right now. I think AI is a great enabler, and we absolutely already embrace the idea that this is an opportunity for us, and we are 300% on it.
On the Middle East, I knew you were going to ask that, or someone was going to ask that. This is the talk of the day. Clearly, obviously, it's a terrible situation that we all are suffering, and especially the people over there. What we have seen so far is certain inflation in freight. That's what we have seen. That is the fact. We have not seen an inflation in costs of the product, apart from the freight, and we have not seen a change in the pricing level in the market. That's what we have seen. We can talk about expectations, forecast, but these are expectations, forecast, we don't know. How we are answering to that, how we are hedging, to your word, to that, is through the flexibility of our model.
In the case of freight costs, we have reacted to that by changing or adjusting the country of origin of our products and by adjusting the shipment method. With this flexibility, we are managing the situation. so far, we are in a good place. Again, we have not seen product inflation. It might happen tomorrow. It might happen tomorrow. In the merchandise that we have bought for now and the coming months, we have not seen it yet. I don't know if that kind of answers. If that comes, obviously, we will again use all the flexibility we have incorporated into the model to tackle it.
We'll come this side, so I don't just bias the right-hand side. Matthew, do you want to go first, then Katie second?
Great. Thanks. Matthew McEachran from Singer Capital Markets. Thank you very much. Not much was said about Topshop during the presentation. My understanding is that the program is going pretty well. Could you give us a flavor of, within your EBITDA in the first half, what losses, if at all?
Topshop contributed, and how should we think about that for the full year? Second one, just on marketing. We've seen the charts. Should we expect your marketing investment and cost to serve to increase compared to the first half in the second half, possibly or not? I've got a question about the outfit builder, but I'll come onto that after those two, if that's okay.
Okay. Do you want me to take the first and take the second?
Yep.
Okay. Topshop. It's interesting that you were referring to the losses on the EBITDA. I'm not sure we're having losses on the EBITDA. I'm not seeing that, but anyway. Let me answer Topshop from a different perspective. We are happy with the development of Topshop. We have seen Topshop progressing very well, not only within ASOS, but also outside of ASOS. I have mentioned the partnership with John Lewis is working very well and is selling above expectations, so we're very happy. We have reached agreements with the best retailers all over Europe. I think it's more than 40 partners that we have reached agreements with all over Europe, and the reaction so far is very positive.
Same in our operations. We are happy with the development, and we are sure that Topshop is going to become a vector of growth for ASOS over the course of a short period of time. You want to take?
Yes. Look, as we said, our focus is on growth. That is where our energy, focus, and resource is going. Will that be through marketing in H2? We continually reassess our investment opportunities. Marketing is one. As I mentioned before, we have a lot of different avenues to reinvest back into the consumer, through product quality, through continued investment in the experience. Marketing is one of those options. We have loads of exciting marketing campaigns coming up. José already showed, and we talked about the Adidas drop. There's plenty more to come in H2, and we'll continue with our more of a top-of-funnel approach than what you might have seen previously. All of those investment opportunities will be weighed up as we move through H2. Our focus, just to be clear, isn't on continuing to save. It's creating investment fuel to reinvest back into growth.
Sorry, your third question, Matthew.
Yeah, thanks very much. Just in terms of the outfit builder, can you talk a little bit about where an AI outfit has been rolled out to, and what degree of average order value that's driving?
You are going to get me here, Matthew. What degree of average order value is driving? I can tell you that overall, everything we have put together is increasing 9% the sales of the consumers that interact with the new app, but I don't know one by one. We measure one by one. We make sure that everything we do has a positive impact and is statistically significant. If it's 1%, 2%, 3%, I will not be able to tell you, to be honest, Matthew. It is one of the items that is resonating very well with consumers. It is true it takes some time to learn how to do it, and we're only scratching the surface. We're only applying that to some items, not 100% of the items, and the tests happen with some customers, not 100% of the customers.
There is still going to take some time to be able to answer that properly. Sorry, Matthew.
No, that's okay. Thank you.
Yeah, Katie.
Thanks. Katie Cousins from Shore Capital. Just wondering about the traffic of your new customers and how much of that is coming from the likes of ChatGPT. Also just on AI Studios, are the models remain inhuman during this process, or will they also be AI? Yeah, my others have been taken.
Okay. Yeah. Traffic from ChatGPT as of today is very small. It will grow for sure. We are invested into, how do you call it? GO, so whatever that stands for, sorry, is generative.
Generative AI.
Yeah, thanks. We're invested on that, and hopefully soon we will be able to share more with you guys. As of today, it's very small. That doesn't really mean it's not going to grow. It's going to grow, and we're on it, and we believe that agent ecommerce is going to be a big thing, even though we feel we are well prepared, and we have a valuable position that resonates with consumer and is going to resonate with consumers, whether they come directly or through agents or whatever. With the AI Studios, we are using human models, and we are not against human models. It's the fact that we can decouple the moment we take the picture that is giving us infinite flexibility. I don't know if the industry is going to evolve to the use of pure data models.
To be honest, this is if you want my personal opinion, I think that is not the core debate. Some people try to take the core debate of, oh, you can save money because you don't have to pay the model. In my opinion, that is not the relevant thing. The relevant thing here is you can generate infinite amount of content. We will pay the models, of course. We're not trying to shortcut or cut that part. Generating more and better content is what is going to engage better with consumers, and it's what is going to generate better sales. That is the real value we are seeing. We are not after, okay, let's save GBP X million on models. That is not our approach.
Thank you.
Okay, to George.
Right.
Thanks. Hi, it's Georgina Johanan from JPMorgan. Just a couple from me, please. Just the first one. On the marketing spend, are you able to share how that's tracked as a proportion of either revenue or GMV in the U.K., please? Just so we can get an idea as to, I guess, where it might land on a multi-year view as you roll out those initiatives into other markets. Second one. Just remind us where you are in terms of natural language search functionality on your own website and app, and whether you would see that being sort of a big driver and differentiator, please. Thank you.
Hi, Georgina. Look, it's not something we disclose, our marketing spend in the U.K. What I can say is that, again, as José has alluded to, within the U.K. is where we are investing more at this stage in marketing in some of our more top-of-funnel activities so that we can take the learnings into other areas. We are sort of continuing to redress the split between bottom of the funnel and top of the funnel, and the U.K. is where we're sort of implementing some of those changes first. The evolution of that will continue through the U.K. and into other markets. As we've said, sorry if I'm a broken record, but it will be a case of continuing to reassess those investment opportunities based on the fuel that we create.
Just to build on what Aaron said on marketing. One thing is the brand side, the other thing is the performance marketing side. On performance marketing, we have a very rigorous approach in the U.K. and everywhere. We don't invest if it's not generating value. If we're investing more in the U.K., it's because it's generating value. The moment it is not generating value, we stop investing in the U.K., and the same applies everywhere. We are very rigorous with that. We are not over-investing in a country because we want to make it grow. We are extremely rigorous with that. On natural language search, we have implemented big changes. Next time, I'm going to ask Anthony to join us here, because clearly it's the topic of everyone. We have implemented big changes on our search.
We have implemented a new search engine that is called Algolia, if I pronounce it properly. This is only the beginning of the improvement. It's getting better search, it's understanding better natural language. It's understanding better consumers, but it's also connecting now with our recommendations engine. Now when a consumer is searching in ASOS, the outcome of that search is improved through the filter of our recommendations engine, and we are seeing very positive results to that as well. As I said on the app, I only highlighted three. There are 50. We have really put the app upside down and completely changed it, and that is only the beginning. The amount of innovation that has come and is going to come in the app, it's really massive.
Can I check if there's any more questions in the room? Because we've got a couple on the web platform. No? Okay. David, sorry. No. Anybody else? Right. I've got a few questions. I'll take them in order. Some of them have already been answered. You took a GBP 67 million non-cash technology asset impairment in the first half. Do you need to re-look at your intangible asset capitalization policies, given how much pace of change there is in technology, and take more through OpEx than CapEx? I think that's probably one for you, Aaron.
I think that is for me, yes.
Yes.
Question two is a slightly more philosophical question about the use of AI. Anubhav Malhotra here at Panmure says that he's noticed that there's increasing negative customer reviews on places like Tripadvisor for the industry, not for ASOS, relating to AI having a negative impact on customer experience. What are your views on that?
Okay. The first question with regards to our impairment, no, we don't need to review our capitalization policy. This was a very specific program. Before I come back to that, let me just talk you through what it is that we've done here. As part of the strategic alliance with Microsoft, that's given us an opportunity to migrate our ERP onto Dynamics. This gives us a huge amount of benefits. It integrates our financial ledger with our ERP, which will create huge process efficiencies, a much clearer visibility and real-time visibility of a lot more of our data. This, we're going to be able to implement at relatively low, close to zero additional cost, and it will generate huge amounts of benefits, efficiencies, and process benefits for us going forward. This opportunity was the driver of the impairment.
The impairment is an outcome of this huge opportunity from the strategic alliance to create process efficiencies and greatly improve our data going forward.
Great. José?
It's good to know that people are not leaving bad reviews on ASOS on Tripadvisor.
That was the question resolved.
I was like, "Why would they go to Tripadvisor?" Maybe we have opened a hotels line and I was not aware of it. Anyway. I like the way you phrase it, as more philosophical. We are not using AI for the sake of using AI. We are using AI with a very clear focus on increasing value for consumers. Let me illustrate that with a couple of examples. I think that being able to talk to customer service 24/7 in your own language adds value for consumers. That's why we do it. I think that having the possibility to show any item that I want in an outfit that is closer to my taste and my choice of color, and you guys know that I'm very special with that and that adds value for consumers. That's where we're focusing.
This is about value for consumers, not about using AI for the sake of using AI, or, as we were discussing before, using AI to reduce costs. It's nice to reduce costs, and we have used that to a certain extent. That is the tip of the iceberg. The real opportunity here is the transformational power of using AI in bringing more value to consumers. I don't know why they are leaving bad reviews in Tripadvisor. For instance, another thing, using AI to give consumers a summary of the opinion of other consumers is adding value for consumers.
Great. Perfect. Okay. The next question is just, we seem to be moving back into a higher inflationary environment. As you said in your presentation, José, very strong pillars of strength for the business, given all the changes that you've made. Do you feel that ASOS is better placed today to weather the storm of higher inflation?
Shall I touch on that?
Please.
Yes is the simple answer. I'll give a little bit more. That would be a probably slightly too shorter answer. It comes back to all the things we've talked about around continual review of our efficiency opportunities through our cost base and through gross margin. Also, it's the flexibility that we've built into our model. A lot of these inflationary impacts have come recently from political or geopolitical events, you might say. The flexibility that we've built into our sourcing model and into our supply chain model enables us to react really, really quickly and mitigate or at least minimize the impact of this on ASOS. Both of those two things combined, I think we're in a really strong position to mitigate. The focus for us, of course, will continue to monitor the external environment.
As it stands today, we're in a really strong position to be able to mitigate the current headwinds.
Perfect.
Let me give you a short reflection on that. In the last three years, we have gone through Ukraine war, we have gone through tariffs, now through Middle East war, all absolute, I mean, especially the wars obviously are things that we don't want to happen. In this very same period, we have been able to significantly increase our gross margin and keep our competitiveness. I think this idea of this is about gaining flexibility and reacting is very important. We will see what happens when things come, but it's not that we're talking about an abstract concept here. I think we have proved it over the course of the last three years.
Great. I've got two final questions. Should be quick. Again, I think this is for you, Aaron. To reach your guidance of broadly neutral free cash flow for the full year, what needs to happen in the second half?
Well, the free cash outflow in H1 is as a result of a seasonal working capital outflow that we always see. To reach that, really, I guess getting back to the sort of fundamentals and math behind it, is to achieve our EBITDA guidance and see the inflection of the seasonal working capital cycle that we see. I guess stepping back even further, what is it that we need to do to achieve our EBITDA guidance? As stated at the turn of the financial year, we don't need to grow to deliver that across the year. The continued efficiencies that we've proven in H1 that we'll deliver, and we'll continue to deliver in H2, alongside the continued improvement in our GMV trajectory, is what we need to do to deliver it. We feel really confident in that guidance that we've reaffirmed.
Great. Last question here is, you've got two warehouses that are non-core that have been mothballed. Do you have any update on what's happening with the two warehouses?
Yes. You may have seen, the eagle-eyed among you, in the disclosure in the back half, we have entered into negotiations with interested parties in our Lichfield warehouse. There's nothing to update on at this stage. If and when there is news to share, we will share with you.
Great. That's the end of the Q&A. José, any closing comments?
Well, thank you so much everyone for being here. I am not going to repeat the three things because I said it three times, so probably say it a fourth time would be a little bit too much even for me. We are super excited for this H2. We feel that we're gaining momentum and we have all hands on deck to deliver, and I am looking forward to seeing you again in six months to share again where we are. Thank you so much and have a nice day and a nice weekend.