4 Fashion Tech Startups You Need to Know | PI Apparel

Kendall Becker, Head of Trend Strategy, Fashion and Beauty Director, Trendalytics. Trendalytics. And Marita Sanchez de la Cerda, CEO and co-founder of size GmbH.

OK, so I'll get up. We become what we behold. We shape our tools, and then our tools shape us. I'm quoting a 20th century philosopher, Marshall McLuhan, who very nicely captured the challenge that when we build tools to support our processes, they often end up shaping them. And so in fashion product development, for the last decades, we've actually struggled with a trade-off: how do we keep structure and oversight without losing the creativity that makes us special? Now, I'm Heidi, I'm co-founder and CEO of Athena Studio, and in the next few minutes, I want to talk about why this struggle has been around for so long, but also why now is a time to solve it. Now creativity is what defines a fashion brand, and at the heart of it are its go-to-market teams: design, product development, merchandising. They should be focusing their time on products, creating the next new collection that wins customers' hearts. But in reality, they often end up spending over half their time on process, updating mirror boards, Excel sheets, going into emails and chats to find the latest status, or copying data into PLM.

Them.

Now, why is this? Go-to-market is inherently chaotic. It's fast-moving, creative, iterative, and spread out across teams, files, and seasons. But so why didn't we solve it? Well, for decades we tried. But for decades, the only way to bring structure into a process was by introducing rigidity. And now these teams are creative. So even more than other teams, they hate rigidity. And they will always find the quickest, easiest way. And so despite big IT investments and big complicated integrations, critical product data has remained in the tools that are easiest to use. Again, the spreadsheets, the mirror boards, the chats. Now this is not just frustrating, it actually costs brands. So we see skyrocketing sampling costs with high drop rates of up to 70%. We see costly manual mistakes with 1 in 4 tech packs being shipped to suppliers with mistakes. And we see missed revenues from time to market overruns and suboptimal decision-making at the early stage. Now, at Athena Studio, we don't believe it should be like this, and we are building the next generation AI-native operating system for go-to-market teams. No, what does that mean? Think of it as software that across files, teams, and tools keeps itself updated and keeps your teams on track. Now, it's not a software or a system. It is Athena. So Athena is the first— yeah, is the AI go-to-market assistant who joins teams, and they basically add her as if she was a human to their shared OneDrive, shared Miro, and chats. On Teams. Now, Atina that way knows what's going on in the collection. Here we see an example where she's being added to a mirror board, and Atina understands like we do that this colorway is dropped. Here, a colorway is being added by someone, and she also knows that she now needs to monitor this change across all the other files that are referencing this style. In another example, she's going into a very limited assortment plan where the user is going to delete an option. A style. And so again, Atina understands that this action is taking a style out, and she notes to herself to track this across the other files. And a final example is a PowerPoint line review where people are adding feedback on stickers, as they often do. And so Atina is also trained to know, okay, what is a relevant change that I need to take into account for my database? Here, an option being added, and she tracks that as well. Now, this is not just over files and boards. She also tracks conversations and emails being shared. Now, users interact with Athena over Microsoft Teams as any assistant or any colleague. So here she keeps us posted of the changes that were made, and she flags inconsistencies that we need to be aware of. Here she alerts the team of upcoming deadlines, which is also a critical element in the go-to-market process. And here, in addition to sending alerts, she also answers questions. So what is the fabric of this dress? It's a real-time question. What is the latest status? We can also ask questions over time. Which options were dropped in the last 2 weeks? And she also— sorry— pulls that up. Now, as a final use case, Athina is also smart enough to understand the full process. And here she sees that there's a style that has all the information to create a bill of material. So she suggests to fix that for you. Now, Athena is not just a chatbot, which I understand is confusing given the demo. The real magic, of course, is in the AI models that we built to turn the unstructured data across different sources into a structured relational database. And so today, we bring one source of truth. So Athena is very much in the background. She's observing, alerting, aligning. We're already working on tomorrow where Atina will be actually autonomously keeping systems and files up to date and will be able to also do more complicated tasks to really save teams time. And over time, our vision is really for Atina to be the only go-to-market interface that your teams need. Think of it as going from the tools that you love to the enterprise systems that you need. So we really want to help brands get to 100% PLM accuracy right from day one. Now in summary, we are an AWS Microsoft supported startup, and we were one of 6 startups selected for New York Fashion Tech Lab this year by retailers including Lululemon, J.Crew, and American Eagle Outfitters. Now we're currently working with global brands that are spread across US, mostly East Coast, so we'd love that to change, and Europe. And I think to end the presentation, we really believe that great products deserve great processes. And so in the words of Marshall McLuhan, Atina shapes the way that you work, not the other way around. So you can find myself and Arnaud, our CTO and co-founder, at the conference for the next 2 days. So we'd love to have a coffee or a drink tonight at the reception to chat a bit more and show you how Atina can help you. Thank you.

Thank you so much. Okay, my mic works. Excellent start. Um, this panel was not meant to be like a female founders panel, but I absolutely love that it's turned out like that, right? So good. So, um, I'm Phoebe. I'm the founder of Fit Collective AI, and I'm sure you have all trialed a bunch of size and fit tools over the years that have been around for 20 years. I'm I'm not naming any names, but they didn't work. And you're like, ugh, size and fit tools, right? OK, well, that's completely different to how we do it. And I would love to talk to you a little bit about how that works. So where's the next button? There we go. So I love this industry as much as you do. But the 3 main stakeholders in it are struggling— businesses, consumers, and the planet. Businesses barely breaking even. We're seeing like a 4% average profit. And if you've ever been in a business where you're looking at the screen that says so much gross revenue and you're like, how are we still barely breaking even? That's what I'm going to be talking about today. So businesses are struggling. And this is an example of, uh, an enterprise customer of ours where you can see this is a swimsuit that they made. It sold 150,000 of this swimsuit, but the return rate is 58%, and that is which is totally normal in Europe. Germany is 70%, and the US is catching up year on year. So the average price was £50. That means they've made £7.5 million of gross revenue from this product. But they've sent back £4.5 million to their customers, and they've spent £10 with returns processing and redelivery, which is almost another million. And they've spent £2.5 million buying customers to— persuading customers to buy this product that they've actually returned. And the manufacturing cost of the product. So what this is actually saying is this isn't a bestseller. This product is actually losing money. Not talking about OPEX, staff costs, stores, salaries. Just trying to get consumers to interact with this product is losing millions. Now, in Europe, that means 1 in 3 products are not profitable. And in the US, we're talking about maybe 1 in 4, 1 in 5, depending on the brand. So businesses are struggling. Consumers are returning a third of what they're buying. And I love this screenshot from BBC News, my home from home. I'm from London, in case anyone didn't get that. Where it's saying that H&M medium can actually be anything from an extra small to a double extra large. I'm sure we've all experienced that. And consumers getting really frustrated that they're being forced to pay for returns when they're like, I'm ordering my normal size and you guys can't make that right. and I don't need to tell you guys about the impact that this is having on the planet. So the industry is in this tricky downward spiral. They're forced to make products as cheaply as possible, spend way more money on advertising, not sure if the products are going to be successful, return like crazy, actually sell, and then next year there's even less money and you have to make them even more cheaply, and consumers being charged the same amount year on year or slightly more. No one's winning. So the only thing winning from this downward spiral is advertising. So hands up if you're working in fashion because you want to help Facebook make more money. Right? Hands up if you're working in fashion because you want to make great products that consumers love. Exactly. And this is why I love P.I. So I always say the fashion industry is making products people want to keep, and advertising the hell out of them. Most technology and innovation and solution is all about the second part. I care only about the first part. So, I'm Phoebe, this is my business, and I have a Master of Science degree in consumer behavior psychology. I started Fit Collective 2 years ago, and since then, I'm very proud to announce that we've just raised the biggest female-founded pre-seed round in Europe of £4 million. Which is super exciting. Thank you. And we currently work with brands all the way from a million of revenue up to a billion of revenue. And our customers say they would cry without us, and without Fit Collective, they just have to guess. And does— has anyone heard of this street in London called Savile Row? So it's a street, for those who don't know it, where for 200 years they've made men's bespoke suits from scratch. And I'm very proud to say that my business before this was the first women's wear tailor's in the 200-year history of this street. Thank you. You guys are very nice. Thanks for all the clapping. Which meant that I spent 10 years listening to actual paying customers complaining about sizing. It did age me, but, you know, I'm in LA, so anything can happen here. And it meant that I became a world-leading expert in fit. It also meant that I developed the only dataset in the world that actually has the consumer's body measurements as well as the garment products. And so that's how we developed part of what we build at Fit Collective. So we've built an upward spiral, we've proven a reduction in returns of 10%, better products, and happier customers. If you care about the industry and want to be part of a sustainable, more profitable, and more innovative solution, please send me a message on the app, I'm only here for today, and I'd love to meet you.

All right.

Do we have the mic on?

Going.

Okay. All right. Hi, everybody. So today I'm going to be talking to you about utilizing data and AI to empower your retail decisions. To find out what consumers actually want with Trendalytics. So my name is Kendall Becker. I'm the Director of Trend and Editorial Strategy at Trendalytics. We are the leading retail analytics platform. We partner with businesses to anticipate trends, make disruptive decisions, and drive profitability. So I want to start with you at the basics on what is a trend. So this is a big thing. Of course, as we know, we are inundated with trend information on TikTok, the runways, you name it. So when I'm seeding through trends, looking at a few different factors that are 3 things that can help define what exactly a trend is. First of all, I want to know about the relatability. This is setting the stage in the world around us. If we're wanting to dress more conservatively and all of a sudden miniskirts are popping up everywhere, It might not match the vibe of the moment, per se. We also want to make sure that this is adaptable. A trend that lasts long is able to fit a vast variety of consumers, show up at a variety of different brands, retail stores, you name it. And it also needs to be actionable. So this is when we're getting our really great data coming into play. Are we seeing strong turns? Is search growth coming in? Are hashtags popping up and click-throughs really resonating? So I do want to lay the groundwork that this does start with you because we're not in the business of knowing every trend, but it's knowing the right ones for you, your customer, and the ones that are resonating in the world at large. So that leads us into Trendalytics in action. So what we do is we look at these whole variety of inputs here that you can see over on the left, looking at Google Search, social media trends, looking at YouTube, Instagram, and TikTok, influencer insights, and e-commerce data. What happens is Trendalytics isn't just aggregating this data, but it's reverse engineering this with AI to give you a trend forecast that is actionable, informative, and meaningful. So you can use all of this information in, in a variety of select outputs, and we're going to get a little bit more concrete in here to walk you through a timeline of how to put Trendalytics into action. So a big macro trend that I've been looking at from last year all the way into next and so on into 2027 is what I like to call the Age of Aquarius. Yes, it does have astrological relations. I won't dive into that. That's my hobby on the free time. But looking at our short lead forecasting, this is what I want to know what is happening right now. So I would categorize this macro trend as Bohemian. Looking within fashion, we saw the revival of Chloé coming on. We also saw other brands, let's say like a YSL, who wouldn't normally be playing in that space, starting to jump in here too. And of course, as we know, a lot of the trends that have been making an impact this summer into early fall, thinking about airy maxi dresses, suede, cowboy boots, you name it, but it feels very light and bright. So we can say that these trends don't necessarily evolve in isolation. So what we're able to do at Trendalytics is roll up all of this data to look at the trend story as a whole. So as you can see, it is up 463%, which is a really great and meaty number here that all of these bohemian trends are on the rise altogether. So now we want to get a little more nitty-gritty here. We can see the trend trajectory is up a growth at 100%. The searches are incredibly high. But I also want to look over— you can see at our most searched trend is a babydoll top. So a great example of this on how we can action with this information right now. So let's say you're updating your copy for SEO. We really want to get that product you spent so much time and dollars into development seen by the right eyes. So let's say you might have called it an empire top, but that babydoll top is the most searched right now. That's a really nice and easy swap that is going to put money back into your pocket. Other ways you can put this information into action is, OK, bohemian, this is doing great right now. So can we update our assortments online to create an edit that's speaking to this? Adding those buzzwords into social media captions or curating some in-the-moment newsletter topics to be able to feed into this trend. So what we do want to know is as we are continuing to move forward and develop, which of these trends are going to stick stay with us and which might be dying out. So what we want to use this data for is understanding the lifecycle of these trends in particular. So for example, I'm calling out here balloon sleeves. The data has been popping up with balloon sleeves, balloon pants, these voluminous silhouettes starting to pop up. And even looking at the runways, of course, we're in the midst of the spring/summer— oh my gosh— 2026 season coming up. We got Paris kicking off today. Exciting things, but we have been seeing balloon silhouettes all over, even starting from Copenhagen. So looking at the data, you can already see that this is up 263% to last year. This is really exciting. So looking at— we're sending our buyers out to market. Can we direct them to make sure to have this into their assortment? Or let's say you're in design, and this— maybe it's too short of a lead time, but do you have a best-selling body with a manufacturer that you can make a quick sleeve update to then be able to play into that trend? And then maybe you're curating different advertising campaigns. Can you update to be sure to include some of these driving trends that you have these medium lead time to action on? And looking at our long lead forecasts, this is where we're looking at social media. These conversations are coming into play. And yes, we're talking about fashion here, but also looking at what are people talking about online? You may have noticed that TikTok, there is a lot of mysticism growing. For example, there's a TikTok tarot card readers. There's also kind of darker outlooks and jokes that happen on TikTok. We also saw very trend-driven events like Coachella, that a lot of the outfits that people were wearing leaned into almost this dystopian trend with distressed knitwear or really harsh and edgy Moto boots. And so we were seeing this start to grow online, and eventually this sparked me into crystal jewelry. So as we know, crystals can be empowered. As we know, rose quartz, for example, maybe that's helping you channel love. Okay, great, awesome information here. How can we utilize this? Well, now that we have so much time, we're starting to see this trend evolve, and in my perspective, it's taking less of this light and bright, airy bohemian, and it's getting more into this mystic tone that has a darker lens. So that's our evolution of that trend story that the data has driven us to be able to understand. So with that, we can design and develop product that leans into this darker aesthetic. Maybe you're bringing back velvets, for example, or trying out this crystal jewelry. This gives us the opportunity to make sure that that trend is really aligned with our brand DNA and the consumer mindset of the moment, and helps you consider different messaging and partnership opportunities. Maybe you're a jewelry brand that doesn't normally play within gemstones, but you have the opportunity now to partner with a sourcer, a manufacturer to source jewelry that does play into this. So thank you so much for listening to our take on trends here. We'd love to learn more about your business. Have a conversation today to see how our AI trend forecasting can help empower your business.

Good morning.

Am I—

can you hear this? Ah, yes, perfect. Good morning, everyone. Last but not least, um, happy to be here today and talk to you about what seems to be 50% of the panel today, about a very important topic as well, which is sizing and fit. And it's wonderful to be back. It's our third time this year at a PI conference. And earlier this year at Lago Maggiore, we had the pleasure of doing a very interesting think tank actually on the topic of sizing and fit then as a return lever. And therefore I thought, let me, you know, kind of bring a little bit of a glimpse into that, into today's morning session. My name is Marita. I'm one of the co-founders of Size, and that's Size with AI. So, um, sizing and fit is really all that we do, and not the AI part in a buzzwordy kind of way, but in a way that's actually solving one of the industry's biggest problems. And I think today's panel kind of goes to show to that a little bit, um, because as a matter of fact, yes, so returns is obviously a huge a huge issue. We've heard the numbers before as well, and 70% of that are actually caused because of sizing and fit. I don't think that's anything new in the room, but what's interesting, it's really the sizing consistency that we take a look at quite importantly. And what's interesting, one of the partners that we work with just brought you this beautiful graph by one of our data engineers, which is very technical, but at the end of the day really shows that this is one and the same brand. The pants and their waist measurements. And within that brand, these waist measurements alone, they vary by 7%— hip measurements, sorry— by 7 centimeters. So that's a lot, right? So I think we can all agree that obviously, if at the end of the day— and everyone knows this, right? We go in, we try 2 different sizes, but if none of them really fit, it's not a consumer selection problem. It's more of a product fit and size consistency problem. And what we see here is really that at the end of the day, there is a very direct correlation. We look into KPIs like fit consistency, fit rate. So this is really where our passion lies. And then if we compare this to the return rate, what's interesting is really that they're directly correlated. So things that don't fit very well, very inconsistently, that do have a huge impact on the on the return rate. And I think in a, in an area or in a year especially, and in the last couple of years where margins is really everything that we as businesses also often look at, knowing that returns and the sizing and fit being a lever that can help us is a big opportunity. And the question then we ask ourselves is, who actually owns the return rate KPI in your organization? And Coming back to the think tank earlier this year, that was very interesting because a lot of the product development colleagues there, we all asked ourselves, and a lot of times it is actually in the wrong place, so to speak. So I invite you to take a little bit on a journey with me today and think of it almost as this product-customer data platform. It's gonna make sense in just a second a little bit more, but to really look at the organization holistically. And what I mentioned, right? A lot of the times, return rates are located with e-commerce, which is interesting because it's actually a product issue, right? So we should look at it a bit differently from— and I think all of here from a product development point of view can share that kind of vision— that that's where we want to kind of look at it as a root cause of sizing and fit. So what we look at a lot of the times is, This is supposed to be checkmarks, but I think, well, presentation style. Is that the different solutions obviously here to stay, and we heard it earlier, and I share this very much that a lot of the time sizing and fit solutions in the past couple of years were only focusing on e-commerce, but it's too superficial. It doesn't solve anything. So what we really say is it doesn't account for any of the other data points in the organization. So yes, it might be superficial on the e-com part, but we really want to look into to data that goes across the organization to the actual product, into the actual manufacturing and how it came out at the end of the day. So, there's all these different solutions that would look at different kind of steps along the way. So, what we then said is, "Hey, we need to look at this very holistically. We need to collect all those data points and connect it in this unified system that would ultimately have, you know, this trajectory of having an impact." visible impact on the P&L. And you can almost think of us in size as disconnecting the dots officer, especially when it comes to sizing and fit-related topics. And what we see here is, and if you remember anything from today, is really what the way we look at a product and the customer and matching that in a one-to-one to really give us this depth of insights in not only the product development side, but at the end of the day also how we want to communicate this to the end user and to the shoppers that at the end of the day are still the ones that are buying our products. So, it's really the SaaS operating system built on which there's different applications that we view, you know, kind of super helpful, both on the product development side but also on the way we can communicate this. I know this was all very, very, you know, technical in a way. So, what does this actually look like? So, we call it our Size Fit IQ, and what we see is really, um, going the, going to step by step across the, across the organization. So looking into the worst performers, who perform the worst from a return rates point of view, who have actually a margin impact, would it be more profitable to discontinue it, what is it that we need to be looking into, does this product have a fit-related problem, and if it does, we can actually see What are the fit rates for that product, as we saw earlier, related to my return rates? Which products do I need to look at in more detail? What is the fit consistency? Can I really see— and here it's very interesting what we see with the brand partners we work with— that the products with a lower fit consistency are the ones driving return rates. So here, really being able to pinpoint exactly where we need to start to understand what does my customer actually look like, because at the end of the day, They really want to know who are we designing for, who are we creating our products for, does that influence our product development? And then taking one step at a time once we know all of that, being able to deliberately know where can we go into product improvement, and ultimately the holy grail at the end of the day also, how can we make products even better to make sure they fit my audience? And at the end of the day, we know all of this about the product, and we want to be sure to communicate this effectively to the end user, whether that's through a conversational agent. And, you know, if we all read the news over the summer and what that would mean for the shopper experience, but also really knowing that actual product which is going to be recommended to the end user, because ultimately at the end of the day, that's how we're going to unify everything together. So it's not just a returns problem, it's actually a margin issue, and what we see is that here at Size, we were able or are able to deliver a 40% impact on gross profit margins with the brand partners that we work with. To put that into numbers, a $3 billion annual revenue company is looking at a fit-fix of roughly $300 million in margin opportunity. Now, we've proven track record with a lot of the leading European brands and are obviously I'm very excited to be here today and happy to go into more detail on what that would look like. We're Berlin-based but have teams across Europe and the US. So yes, those dots should look a little different, but just to get an idea of where we're based. And if any of that today sounded interesting to you, I'd be super happy to chat with you about how we could rethink about sizing fit in an industry. And if anything you want to take with you from, from this presentation today, that fit doesn't have to be a guessing game, it's a data opportunity. Thank you very much.

All right, we've got some questions. Good job, thank you. All right, so Dom Bonifacio. Athena, what does security or access look like across Miro, OneDrive, Teams, etc.? Do you have any pilot examples, successful use cases tested at retailers?

Yes, I can give a short—

you hear me?

Yeah. Now I can give a short answer. Happy to go in more detail after the presentation. The way that we give access, we made it as simple as possible. Think of Athena as a person. So OneDrive, we're integrated with Microsoft. We set that up with IT very quickly as a guest user. So you can use @Atina to a OneDrive folder, an email that you forward to her, a chat group that you just add her to Teams. And that is all like, yeah, that is the security of your own Microsoft setup, but we can go in more detail about that. Miro is idem dito. So you add her to a Miro team or a Miro board, and that way she gets access get access again as a human would get access. So it's very restricted. Pilot cases, also can tell you more. The only thing is because we work with actually enterprise customers and we cover a very— yeah, it's the IP of a brand, right? It's their new collection and all their data. We are not allowed to say publicly who we work for. We can give one-on-one a bit more information about the type of brands that we do pilots with at the moment. I hope that answered the question.

Yes, you have the second question too. So what does training or feedback for Athena look like? Curious if it's a closed ecosystem specific to each company or if it's a logic shared across many.

Yes, very good question. So we actually set up our brands again linked to the fact that it's enterprise and very, very confidential data. Every brand has their own database. It's separate, physically and logically separated from other brands. And so the ecosystem is indeed very much— yeah, integrations are the same. Every Microsoft setup is similar. Every Miro, I mean, it's the same software. But then the way that brands are using it, it is slightly different. So the core of our AI models that turn data into a structured database, that is the same for everyone. We have, for instance, now a collaboration with brands that are focused on circular design. There, that's very specific. Not every brand is using us for doing early-stage circular design measurements. We have another luxury brand that is much more focused on specific use cases where we needed to go deeper. So it is indeed custom leveraging what we already have. And training, very quickly, also we do that initially, and we are very close to the brands that we work with to check. We follow engagement to see, are people continuously talking to her? Do we see a drop? Do we see inaccurate responses? So all these things we track. Training is very simple because it's simply chatting with a colleague over Teams. So there it's quite limited in terms of training.

Okay. They keep changing. What sources is SaaS using to understand customer fit data?

Ah, yes.

Hello.

Does it work? Yeah.

Okay.

So there's two different ways we can do and go about this. Typically. So one is the most, I would say, low-hanging fruit where we do facilitate an e-com integration. And by that e-com integration, we are gathering data and our human AI, as you saw earlier, is calculating the actual customer and their fit preferences and the way they look like. So really their circumferences and understanding their, the actual customer avatar, so to speak, from the e-com sphere. But we can also have opportunity to either do that through customer success programs or in-store and offline programs. So there's multiple ways and touchpoints we can facilitate with the actual customer because, yeah, as you have probably gathered, I think that, or we think that's a huge lever and opportunity for us to really understand what they look like. Great.

I'm gonna jump to a question for Kendall. Kendall, I I'm curious about Trendalytics and how you're defining when a trend starts and that lifecycle of how that trend flows for how long it flows. Because I think that's very interesting. Looking at something now and kind of looking at where it's at in its lifecycle and making decisions about product 18 months in advance.

Yeah, absolutely. And that's where I think the nuance comes in. Because it's slightly different for every brand and every retailer, right? Exactly where you're falling in that line. So I think how you analyze it, and that's where your creative comes in and you're adding in your own expertise to analyze that data because there's a plethora of options to look at, including TikTok data, which can be a really early indicator of a sort of movement that's starting to happen. Then as time is going on, we can start to dive in more where I love looking— we have a market adoption number and that's a great stat that I love to use to help confirm, okay, great, this is starting to be adopted. And you can see maybe it's by luxury retailers. So if I'm sitting in the mass market section, I can know, okay, great, these numbers are looking really strong. So let me action on it when I know that this is coming up and will work really well for spring, let's say. And then you have those Google search numbers that really can resonate whether it's in that medium to, to short lead time. So it really depends on your use case and where you fall in that cycle. But I think that's what's so great about it, to be able to analyze it and put it into different use cases.

Great, thank you. Okay, this one is for Phoebe. For FIT Collective, is the analytical data you can incorporate for brands their specific customer data or regional data for— of means outside of countries?

So we, when we're in Setting up with a brand, we take their transactions, returns, and reviews. So if you're a small business on Shopify, you just install our app and you give us the API key to your returns and reviews. So a small business can be set up in like an hour. If you're a big business, we also take your PLM data. So those are the four like main sources of truth that we'll use. And if you don't have reviews, then we can also ingest customer service feedback emails when people are sending an email saying, I washed these jeans and all the colors run out, we make sure that that's all incorporated as well. And one of the cool things that we do is we're able to show you how you're performing relative to industry benchmark, geographical benchmarks, and price point benchmarks. So we know that a £600 dress is always going to return at 40%, 50%, and a $10 pair of socks should return at 2%, right? So what we can do is not only tell you how your individual products are performing, performing within your own business, but how that whole category of your business is performing against its peers.

Great.

Thank you. That's the last question for now. There are quite a few more questions. So excellent panel. Thank you so much. And if you have questions, please find them and spend a little time getting your questions answered. Thank you.

Thanks, everyone.

Thank you.