Gamify the Dressing Room With Stefan Hauswiesner

Hi everyone, welcome to the latest episode of Fashion Made. On this week's show, I speak with Stefan Hausweiser. He's the CEO of Reactive Reality, a company that makes a virtual dressing room solution for e-commerce. Very, very interesting stuff. We try and cut through some of the noise. I know it's a very busy marketplace for a lot of fit, style, filling, virtual dressing room solutions. But we kind of cut through some of the noise, as I say, to get through to the reasons why it's such a busy marketplace right now, why both the consumers and the brands, retailers using them, get so much value from them, and the numbers, you know, the cold hard facts behind that. So should make for a really interesting conversation. I hope you enjoy. Here it is. Stefan, hey, how you doing?

Hi, I'm fine. How are you?

Yeah, I'm well, thank you. I'm very well. Um, after our really good event a few weeks ago, I really— I had some more questions that I wanted to ask you and we didn't have enough time, so let's do that now. For anyone that didn't hear that event, I mean, you can go on to the membership site and you can go and rewatch it if you really want, but let's go and recap a couple of things because obviously Reactive Reality, really exciting company doing some great work, and it's all about virtual fit, correct?

Correct. So Virtual Fitting solves one of the biggest problems in e-commerce. It enables online shoppers to try on outfits before they buy them, just like in a store, but much more convenient.

So into more depth, I mean, why is this needed right now beyond the pandemic? Of course, um, let's say we had retail like it used to be a couple of years ago. Why is Virtual Fit needed in that scenario?

Certainly the pandemic has accelerated a development that was going on for the past years. Online shoppers expect a frictionless experience, but these days also expect an engaging experience that helps them make good purchasing decisions and are fun at the same time.

Is it fun?

We, we put a lot of effort into making it fun, and we believe it is. So our customers have actually measured the user engagement of our technology And they found out that people, users of our technology, spend over 3 hours every month mixing and matching outfits. And this is an insanely, an insane amount of user engagement. This is at the level of, you know, podcasts and gaming apps. So this is really remarkable.

Yeah, it's amazing. 3 hours just sort of swapping out garments. Amazing.

Yes, that's mixing and matching outfits, seeing how these outfits look on your own personalized avatar, taking your avatar into virtual background scenes, right? You can, you can place your avatar on a beach when you're trying on a summer dress. And this is just, you know, this is an experience that people are loyal to and get back to and open up the app of our customers very frequently every month and spend a long time. And as a consequence, they buy more and they make better purchasing decisions. And this is something that we deeply care about as a company, because any such technology, you know, that is good for the environment, that reduces product returns, any such technology needs to be fun and engaging, because otherwise people will not use it, and then it has no impact.

It's like the gamification of the changing room.

It's a sort of gamification, but useful.

Yeah, I mean, games can be useful as well, I guess, sort of function and form as well. How does it work? Let's get into that a little bit. You know, what do we need from the consumer? What do we need from the brand and the manufacturer? Let's start with the consumer. Obviously, we need their measurements, right?

Correct. So our product and platform is called Pictofit. It enables online shoppers to create their personalized avatars on their mobile phones. This only takes a couple of minutes. Users are expected to input a couple of measurements, very easy-to-take measurements, and then our artificial intelligence engine transforms this into a full 3D avatar of the shopper. To personalize avatars. People can capture their, their face and head, and the algorithms actually determine their skin tone, their hair, and customize the entire avatar for it to be photorealistic in the end. And this is something that is super easy to do, only takes a couple of minutes. And on the retailer side, we— our technology transforms product photos into what we call smart augmented reality assets. And this means that the retailers can take their existing product photos and convert them into a format that enables Pick2Fit to try on these outfits on their, on the online shoppers' avatars.

You know, people can be pretty particular about the skin tone. I guess my question would be, how accurate is that?

It is extremely accurate. So the technology actually scans the user photos and determines the skin tone from the photos. And creates a whole 3D geometry of the user's head, including the hairstyle. And these aspects are extremely important to have a personalized fitting experience. And we have, we had so many users who sent us emails who said for the first time they could see whether an outfit looks good really on them, you know, really on their own body with their own skin, on their own hair, their own unique body shape. And this is so important. And we believe that this is the future of e-commerce.

100%. 100%. Is there an option though for anonymity in this?

Yes, privacy is incredibly important to us. And the avatars that people create of themselves, they can live on their phones and not leave their phones after they've been created. So all the fitting, all the size recommendation, all of these nice features would run on the user's phone such that the avatars don't need to be sent around, you know, to all kinds of retailers and other parties.

So the brand, the manufacturer, the retailer, for the people that are creating the digital assets, the clothes, the shoes, what software do they need to use, specific software for this, or can they create these in any software and then translate that into the Pictofit platform?

That's actually the beauty of our solution. We only require conventional product photos, and the retailers and the brands, of course, they have photo studios, so the only thing we need to read in are product photos that get uploaded to our web servers, and there the algorithms transform those photos into smart augmented reality assets. This means that these assets are equipped with the information of how the garment would fit on a user's body.

Hmm, interesting, because there's a lot of people designing digitally these days, so why wouldn't you just take the digital designs in the first place?

So actually we do have an import option for digital assets, and this is something we see our customers have a lot of interest in. But in the practical realities that most retailers do not have meaningful coverage of 3D assets of their product range, you have to see that many of the retailers, they have hundreds of thousands to millions of SKUs in their store, and at best they have you know, a few hundred 3D models that they get from select providers. And we always see our technology as something that is useful to everybody every day and covers all the products, not just a few select ones.

What product categories can it be used for? Is it everything?

It can be used for all products. So on your own, as an online shopper on your own avatar, you can try on all kinds of garments, but also accessories, shoes and bags and hats and sunglasses. And that is extremely important because people tend to buy outfits. They want to see how things go together. And this is something that our technology uniquely is able to offer.

And I mean, you kind of mentioned it before, but I guess this is pretty scalable then for large numbers of SKUs because they would have already taken the photos.

Yes, Pictofit is built to work with millions of products and users. And one of our customers, for example, is adding 25,000 new products every month. And this creates them hundreds of millions of outfit combinations that their customers, the online shoppers, can interactively put together and buy.

We kind of mentioned it at the beginning, but I guess there's been a bit more demand for virtual fit in the last couple of years.

Definitely. We see that all the retailers are rushing online. They have to in the past year, of course, due to the pandemic. But now many of the retailers realize that they have to offer more because in the online business, the winner takes it all, right? For an online shopper, it makes no difference where they buy from. You know, they type in a different URL, but the rest is pretty much the same. And you have to see that all the webshops look essentially the same. You have a product listing page, you have some filters, you have a product detail page, and you see models wearing the garments, right? So it has almost zero personalization. So many of the retailers realize they are not really equipped moving forward to compete in this what has become a really competitive space, right? And we offer them a superior user experience that they can roll out to their customers. And we offer the online shopper a solution that really helps them make good purchasing decisions. And we saw that this has— that this creates a lot of user engagement. So the metrics of our customers show that users who interact with Pick2Fit spend 5 times more time in our customers' app, and as a consequence, they buy 6 times more.

Wow. For anyone that's listening, my eyebrows just jumped up because that's a huge number, really.

Indeed, it is a huge number. And yeah, we were surprised in the beginning as well, but it's consistent. You know, we've observed this for the better part of a year now with our customers, and people are super engaged because it's just such a different experience. And the largest leverage the technology has is it creates a user behavior that is loyal and people come back. And that is usually not the case in a shopping app. Usually online shoppers open a shopping app only if they're looking specifically for something like, you know, a new pair of jeans. But with this solution, people open up the day— open up the app almost every day because they simply like it. They like to explore and they like to engage and interact with it just like with a game. And this creates this huge leverage effect.

Interesting. You know, a lot of people don't even shop using apps as well. They just use, well, Safari or whatever internet browsing app they have. Does it have to be within the app specifically or can— is this just hosted on normal website web pages as well?

Uh, Pictofit is actually compatible with websites too. So iOS, Android, and web. And we offer actually a range of different options to integrate the technology. This can range from personalized avatars to 3D product viewing to virtual background scenes and really branded experiences, all the way to virtual dressing rooms and virtual fitting rooms where online shoppers get the experience that I just explained. So there's, there's a wide range of options for our customers. And it covers definitely all platforms, including web.

Do you see any countries spiking above others in terms of their use of this? I mean, you mentioned some, you know, 5 times engagement, 5 times hang time on their websites, and 6 times purchasing, which still baffles me. It's really, really large numbers there. Are there any countries that use this more than others?

We haven't seen large differences in that. We have seen that the pandemic hit the entire world and all of the continents seem to be similar in that respect. We have offices in the US, in Europe, and in Japan, and all of them report the same pattern. Retailers are rushing online, they're recognizing that they need to provide superior user experiences and are getting into this types of developments.

It seems strange that, um, you know, there wasn't much sort of mix and matching, um, of outfits to my knowledge beforehand, even without a personalized avatar. You know, we didn't often see that on, um, in e-commerce solutions, you know, on the big guys, H&M, ASOS. You, you didn't really see that before. Do you feel like that's going to become more of a thing, whether it's on a personalized avatar or not? Or why don't we see that more often?

I think the reason why this hasn't been established in the market before is the technical complexity. So even fitting out outfits and garments to pre-existing models is technically challenging, which is something that we solved for. And the reason why this hasn't been possible was that without virtual fitting technology, you would have to shoot the product images exactly in a, in exactly the same way how they would fit on the model. And, you know, if you just move the camera a little bit or the mannequin a little bit, the whole thing is broken. So this was in the past the reason why this hasn't been possible and could only be overcome with actual virtual fitting technologies such as ours, which then in turn also enables online shoppers to create their own models. So these two things kind of came in tandem because they are based on the same technical developments that are needed for such a solution.

Yeah, yeah, they're kind of synonymous with one another. Um, but that's fantastic. I mean, it's kind of like bringing two solutions to the market in one.

Correct. Um, so one is the fashion image generation, and there is a lot of appeal in that because you have to see that you can actually provide online shoppers with a huge selection of of models to see the products on. So you could, you know, as an online shopper, you could pick a body shape, a body pose, a skin tone, a background, all of that dynamically, and you get the images. And if you take that, and if you take that experience and basically take one more step, then you end up at personalized avatars, which is something of course that what we offer and that enables users to be the model really.

So you reckon there's a big correlation between the pandemic and the demand in virtual fit solutions?

There was a solid demand before, but it has certainly accelerated.

And you reckon, do you reckon there'll be just more and more online retailers offering virtual fit solutions over the next year or two?

Definitely, because the user behavior has changed and will not easily change back. So there's a large proportion of shoppers who used to buy fashion only in brick-and-mortar stores, and they were forced to come online and they saw the value, they saw the convenience, and they will stay. So this is a development that is here to stay and will not go away with Corona.

Agreed. There's obviously a number of solutions, you know, Reactive Reality aren't the only player in this market, but how can we make this as easy as possible for both the consumer and the retailer, you know, just in case not everyone uses Pictofit?

So our technology can be integrated into our customers' apps and websites. So this makes it really, really comfortable for the existing customers of our customer. Another very interesting integration option is to build this into the iPhone, for example, or into the Android software, such that online shoppers have their avatar independent of any particular Online store.

I personally see that as the most sensible long-term solution because you can't be continuing to add in new measurements. Everyone has a slightly different way. Some people might take a survey, some people might take tape measurements, some people might do a 3D body scan. Um, there are lots of ways to get these avatars built, so we need to land on one that works for everyone. And once you've got one, you don't really want to be doing that again and again and again, right?

That's right. Though the way we see it, you create an avatar, it only takes a few minutes, and then you use it for months. So, you know, the ratio between the effort you put in and the outcome is very good, even if you create it for each provider individually. That being said, of course, the most elegant solution is that your avatar stays with you on your phone and can be used across all the brands and retailers. And you get— you always get personalized size recommendations, right? This is really the beauty of it, because different brands and retailers interpret size charts differently for sure. And with a personalized avatar, you always get the right size because it's grounded in really in metric measurements and in a highly accurate representation of your body.

Yeah, absolutely. You know, then style comes into play, which often can muddy the waters a little bit in terms of what fits, you know, what fits some person might not fit another person depending on their choice of style.

Of course, we have a really good solution for that too. So when giving a size recommendation, we also provide what we call a fit visualization. So you would see in a color-coded way where the garment is going to fit tightly and where it's going to fit loosely. And then it's up to the discretion of the shopper to really decide this is— and say, maybe, maybe this is going to fit tightly, but I'm looking for a tight-fitting pair of jeans, right? So I'll buy the size.

So you use color combinations So it's, you know, it's, yeah, a certain color when you say it fits very tightly. That's really interesting. I haven't seen that before.

Correct. So usually we use red for tight fit and blue for loose fit, and people intuitively get it. And this is also really, really nice, you know, in an international setting because you don't need to explain much. You, you see yourself with a pair of jeans on, you see some red bands, for example, around the thighs, and you will know Okay, it's tight there. So this is pretty universal and gives more color to the size recommendation. It's more than just a number, right? It tells you where exactly it's going to fit in a certain way.

That's interesting. You know, you make a good point. I guess this is available in all languages as well. So it's a real global solution.

Correct.

Good stuff. I mean, it seems like you thought of everything, Stefan, but what are the opportunities for development in the near future? I mean, there's always things that you can be improving on, potentially clothing customization and how that might fit into the app? What are the thoughts for the near future?

Being a technology company, we are constantly developing all kinds of cool things, and I can only give you a sneak peek into, into our lab. And for example, we are working on advanced 3D rendering technologies that resolve even the finest details. So you would see on your avatar individual strands of hair you get the complex reflections of materials like satin, which are notoriously hard for any sort of computer graphics. So this is something that's coming this year. And then long-term, we are making our avatars even more lively and animated such that they become a digital twin of yourself and you can interact with them with voice and they talk back to you. They help you make your daily fashion choices. And really become part of your life.

Wow, so you can be your own personal assistant.

Yes, correct. This indeed, the early stages of this we actually explored together with Microsoft and the University of Arts London, which was a very nice project that concluded this fall, or actually last fall. And we created these so-called digital human stylists, which were really digital twins of users. That you could voice control and really have a conversation with about your wardrobe and your daily fashion choices. So it would be connected to your calendar and to the weather forecast. So it can even tell you what to wear in a given day, knowing where you want to go and how the weather is going to be.

Sounds amazing. It made me think of one other thing that's been plaguing the industry in terms of body scanning for a while, which is soft body tissue. And whether you guys have applied your skills to, to that area because it's notoriously difficult.

It is, and we have been working and are still working with customers from the, from the lingerie business, and we're offering them with body measurement solutions which are unique in a way that they give you something that a tape measure cannot give you. And that is volume measurements. So with a 3D avatar of yourself, you can actually take volume measurements, which is highly relevant to give cup size recommendations for bras, for example, because this is actually a volume measurement, not a circumference one.

Very interesting. So more to come there then.

Definitely. This is an active field of development and we are seeing already very promising early results.

Fantastic. Stefan, look, it's been a pleasure to fill in some of the blanks from our last conversation. It's always a pleasure to talk to you. So thanks for your time.

Thanks, and thanks for having me.

Yeah, thanks to Stefan again. Really, really interesting conversation. Found out a lot more about this world than I knew before. And thanks to you guys for listening. As always, I really, really appreciate it. I hope you can join us again soon on Fashion Made, but until then, take care of yourselves. All the best. Bye-bye.