Materials: From zero to HERO with Founder and CEO of Swatchbook, Yazan Malkosh

Hello and welcome to another Pi Apparel Podcast. Good day to you all. Today I have the pleasure to be sat down with Yazan Malkosh, CEO of Swatchbook. That's right, Yazan.

Thanks for having me.

Pleasure. Welcome to Pi Apparel Podcast.

Thanks.

I was really interested to know a little bit more about yourself and kind of how Swatchbook got started. So maybe if we can start with when was the first incarnation of Swatchbook?

That's a good question. It's been, it's been a journey. I think Swatchbook has been something that, you know, my, my career probably started about, I want to say, 15 years ago, but now it's become 18 years. Years are slipping by. Yeah, it started off with just doing visualization for design in 3D and moved on to doing consulting for design companies as they transitioned from analog to digital on the visualization side. And we've, you know, in the past we've done work for architectural firms, design for product, and then moved on to a lot more in the apparel and fashion footwear specifically as well. And a part of some of those consultations is seeing, you know, is seeing what in the pipeline of visualization needs to be done. Sometimes there's existing tools in there, sometimes there's a void or a gap, but we— or at least when I was working with these companies, there was a pattern that was emerging, which is basically material library translation.

Okay.

And that was this, I think this, I at least in my mind, the starting point for the need to look at materials. And a lot of it had to do with materials being tied to specific platforms, either the render engines or the simulation engines. And if you were— if you had your material library in one of them, you're kind of stuck in that platform. You can't go out because you have to translate now your content, which is the actual materials. And so I spent a good amount of time helping brands put together the conversion formulas to go from one platform to another. It's painful. I bet. And, uh, and by the time you're done, you're outdated because now updates have come up, the render engines have come— have have new updates that have gone through and everything you've translated is already out of date. And because obviously you're not dealing with 100 materials, you're talking about tens if not hundreds of thousands of materials. And I think that was the starting point of seeing that repetition and that need at multiple different brands that we were helping out. And I then moved over to creating products for a company called Foundry that make modeling product— modeling and rendering and visualization as well. And we helped a good amount of brand companies, either in automotive or footwear or apparel, to improve their— again, they have an existing pipeline, but they need to improve that with focus on visualization. And that became a pattern that I just kept seeing over and over and over. Like, materials became for— because you're making products, obviously, and a lot of them in fashion and footwear, they revolve around the material that you're making from. So it's not just the design itself, but the actual content you're making it out of. And that type of product meant that you had to look at that material all the way from the idea, what we call from the zero to hero.

[Speaker:MAX] Yeah.

[Speaker:JASON] And the focus was materials. And we just fundamentally felt that the tools that are out there, one, they're not created for designers. So they were really created to either, if you're using PLM systems, for example, they're used for sourcing and manufacturing. That type, that part of the process, which is obviously vital and important. But for visualization, there was nothing there talking to the designers. And I think about 2 years, 2 and a half years ago, that seed of, well, what if this wasn't a, you know, custom consultation for every single brand? What if it was actually a product instead that's able to cater to multiple brands? And that's how Swatchworks started.

Okay, so long story short. Well, yeah, we're talking about a 10-year story, but yeah, I mean, so, so 10 years ago this was just a, a little spark of— correct, just a thought, you know, maybe this could happen. But I— even if you had the willingness back then to put it into practice, did you have the technologies available to you to make it happen?

Not that it— I think the technology was there, the accepting of technology was not.

Yeah.

I think cloud, you know, obviously for us Swatchbook is cloud-based. You know, 7 to 10 years ago, no brand in their rightful mind will tell you will be using cloud-based systems. It was just a big no-no, security issues and things of that nature. But man, that transition was quick. It was kind of like a 2-3 year transition where now brands are like, oh yeah, well We'll use it, that's fine. I don't know what— I mean, I know what happened to make that switch. I think people, you know, brands got it, executives got it at the top level that yes, we can try to shield everything away from our user base, but the reality of it is, you know, we used in the consultation side of things, nobody was using Dropbox, for example, at the time, or officially not using Dropbox, but yet every single file that was sent to us for consultation was through Dropbox. The designers were using it, you know, IT was not happy about it, and/or they didn't know about it. But I think that little corner kind of already had made its way, and now it's more accepting. So cloud was there, yeah, it's an old technology, but now it's being accepted as a normal side of things. Yeah. I think the other aspect that wouldn't have been possible is that now render engines are moving towards physically based rendering, which meant that your materials look a lot more real and more consistent across different platforms and different rendering engines. I think without that, you had to rely on a lot of artistic manipulation specific to each different platform to make things look consistent, which is very difficult.

Yeah. I mean, materials may look real, but especially when I have conversations with people, one of the major challenges is drape. —and—

It still is.

—digitalizing Drake. Yeah, I mean— It still is. How— What are your thoughts on that for the future? How are we gonna overcome that barrier? Is there an answer? I mean, can you see technology—

Oh, yeah. —getting there? It'll get there, but it's— We're, in my opinion at least, on that side of things, it's still in its infancy. Yeah. The reason being, one is typically in 3D right now, you're modeling, whether it's simulating or modeling, it doesn't really matter, You're representing the— specifically fabrics, I'm gonna— and textiles. Let's remove things like hard surfaces, like plastics, metals, and things like that, and just concentrate on soft goods. Yeah. But you're fundamentally modeling or simulating a geometric 3D mesh weave of fibers and representing it on a 2D surface, which is not fair because there's a lot of things that need to go in there. So By doing that, you have to rely on a lot of faking in the materials from the rendering side to make it look like textiles. But the reality of it is, if you really wanted velvet, you really need to have— you have to have actually the fibers looking there, standing vertically on the surface, where in our case, we were making it look like velvet. With draping, it's a difficult conversation because I think also the simulation engines out there, one, everybody's different. They've got different algorithms and different formulas, but also their approach. You know, if you wanted to physically and fully simulate the weave itself or a knit, you'd have to actually solve for the geometric— the actual fibers at the yarn level, which nobody's going to do that. There's some white papers out there that do these kinds of things, and your frames will take you anywhere from, you know, 9 to 15 hours per frame just to solve a very simple, you know, square, let alone something like an actual piece of garment.

So it's just not scalable. It's not scalable.

And imagine, imagine telling a designer who's used to iterating at, you know, at the speed of a line sketch— that's how fast they're drawing— to tell them, no, you got to wait 15 hours for that one frame to finish. They're not— they're— but they might hit you with a sketchbook first of all.

But it doesn't exactly help the creative process. No, no, it doesn't.

So I think, but obviously computers are getting better and stronger. You know, the CPUs are getting more powerful. Uh, cloud computing is coming along. Software is getting smarter. Um, so it'll happen, but it's, it's going to take some time. In addition to, you know, the actual complexity of it, uh, different software has, you know, they have different approaches. Some are looking for accuracy, some are looking for speed. And depends on your user base. Is this designer looking for just quick iteration to make it look as good as possible? So looking good doesn't mean it looks real, or looking real doesn't mean it looks good. So some software concentrate on things that look good out of the box. So it's like, I, you know, I crank a few sliders and things look great. Are they real? Maybe not. And other software, you have to crank up a bunch of other sliders much more than the first one, and you get something that looks real, which means that when you develop it as a product you're less likely to fall into issues. Yeah, but then again, it's that give and take. The designer— the designers, I guess it's, it's the pain threshold. How long are they willing to either wait for the simulation or learn the technicalities of the simulation, right? You don't want to have 50 sliders that have, you know, you have to go in there and fine-tune with different values that you have to either remember or something like that. So drape is difficult. It's certainly been easier for the visualization on the rendering side to solve than the drape, but we'll get there. It's just going to take a bit more time. And there's a lot of opportunities in, you know, machine learning and things like that to help, but that'll take time as well.

Yeah, I mean, not just for the technology, but also with the cultural change for designers and within organizations to accept these new technologies. I'm sure that you've experienced when designers moved from analog to digital. And in some companies that's still happening, right?

Oh, for the, for the, for the fashion industry, that's at, again, at the starting point. Look at automotive. They started about 25, 25 years ago. Now they're getting to that point, but there's still a lot of, you know, resistance from the— especially not all, all of design, but automotive specifically, there's still a lot of resistance from the car designers themselves. There's a a sense of, if I'm not sketching, I'm not a designer type of thing. And I think the designers really have to move to, I need to design in 3D. You're solving for a 3-dimensional product. Why are you, you know, you look at the design from a car perspective, it's a side profile, it's typically elongated more than a real car. It's typically got rims that are 40 inches, which is never gonna happen on that car. It's— they're selling you on a concept, they're selling you on the idea, not necessarily the physical reality that you're going to build.

And I think designers have struggled to let go of that. I think naturally though, because I think if you're a creative person and you create in analog, let's say, you're always going to be a purist, let's say. Yeah, I mean, I think some people are like that, you know. Absolutely. And whether that will ever change, I'm not sure. And you can't ask them— you can't—

or not ask, you can ask them as much as you want whether they'll respond, but you can't You can't force them. They either will do it or not. But we've seen, at least from my perspective, we've seen some designers transition really well to a point where, you know, when you're sketching, you're typically looking for something that looks right to you, right? In your head, you've got an idea. You'll sketch. You'll sketch over it, sketch over it. You're looking for the perfect curve, the perfect shape, the silhouette. And when you find it, you know. Right. But for the most part, you're still sketching in that 2D space. It's a silhouette for the most part, and there's nothing wrong with that. But now that you're creating a physical 3D product and with the technology that we have today, you know, with cars, that shape is almost the same thing. You've got the 4 wheels, the side profile, only see 2 of them, and that's about it. But now we're making products that can be 3D printed and they have a different manufacturing that relieves you of a lot of those, um, I guess structural limitations, that now you can do something that looks funky where I want to create a toaster that isn't this the normal shape because now I can 3D print the things instead of have tooling for it. Um, we've seen some, not a lot, but some designers make that jump to 3D successfully and start actually experimenting in 3D. So, your sketches become a 3-dimensional surface versus a 2-dimensional line. And now you're looking at how do these things interact together, and you're using the computer to generate maybe, you know, 10,000 random variations where your design eye looks at them and says, "Oh, I like these 100 and I'll move from there," versus imagine having to draw that 3-dimensional form or shape. 1,000 times. So it definitely isn't for everybody. And I think it also— the biggest struggle is to show the designer the value of it first, because I think that also needs to be a part of the question. Why am I changing? Sometimes it's not for them. I mean, realistically, sometimes you're changing because the brand needs you to change, because it's not just about you, the designer. It's about the entire pipeline, the entire zero to hero. You're making the sketch, but somebody has to go build it. Manufacture it, source it, sell it. It's not just— I know it sounds bad to tell to a designer who'd love to think that that product is theirs, which to some sense that creation probably is, but they're helping other people and helping their own brand and kind of creating that product in a much more efficient, maybe, maybe also environmentally more friendly, less samples.

This is really important because the design process is somewhat artistic sometimes. But the whole product creation process really isn't. The manufacturing process is, as you say, creating less waste is hugely important these days. And certainly with the community that we bring together, a lot of the focus is on being more efficient as organizations and using technology to achieve that. No, absolutely. Yeah. I think I was speaking to another gentleman before with a big focus on sustainability and doing things for the right reasons that might not make commercial sense right now, but actually in the future this is the way that you actually have to do business, right? I mean, do you think the Swatchbook can help companies achieve that side of things as well?

Uh, I mean, absolutely. We, you know, we believe making products, we believe that touch and feel won't go away. You will need a physical sample at the end of the day, whether it's to check out the materials or to approve the final design. But we also are aware that there's a lot of pre-touch and feel and post-touch and feel decisions you can make that can be extremely useful and powerful in the design process. They're faster, they're more iterative, they allow you to bring play back into the designer's world. And a lot of You know, a lot of the physicality of these swatches almost dictates to the user, if I'm making a product, you know, you only have 5 different physical samples we can create. Well, why only 5? Well, that's the time allowed for that product. Well, imagine if now you can make 15 or 20 or 25. That is not the whole idea of design, is you iterate and you find the failures and you remove those and you go to the success. With SwatchBook, we're able to provide the right tools to grab one material and branch off 20 different variations based on what you've created in patterns or Illustrator files or whatever and see them in 3D directly and make decisions on, "Yes, I want to continue and go and talk to a supplier to get this material made," or, "No, that looks like it's a terrible idea. I don't want to make this material for this product specifically." But we do believe in the virtualization world and creating virtual samples as quickly as possible so that somebody, again, down in that design process can make decisions on yes or no. Whether it's a designer, the design team, the customer, the retailer, whatever it is, there's always a chance for that. And samples, you know, physical sample creation is full of opportunities to really make minimize on how much waste we create. Not just— I mean, obviously there's physical waste, but there's also a lot of energy. People are— there's— you're— one, you're spending a ton of money on it, a ton of time on it. There's travel budgets specifically just to have people to go and see what the physical samples are going to look like. And the energy of actually making those physical samples— would imagine just removing that, because it's not— everybody thinks of it's just the actual physical waste that you're touching with your hand, but all that energy of everybody in the team trying to get there, what if they didn't do that and they did something else that's more beneficial to the brand?

Yeah, well, it's really interesting because a lot of the talk is about in the early stages from design to manufacturing. But actually what really becomes interesting is once the consumer gets involved and the rendering and the decisions that they can make for buying products becomes so good that you haven't even made it yet. Absolutely. You know, that's— that they're saying, I want that. And then you're like, great. And that's the minimum wage, surely.

Yeah. I mean, there's— we've already seen some— we've already seen some brands starting to play around with the idea of brand, like virtual brand ambassadors on Instagram. So on Instagram, you can go and follow somebody who's a brand ambassador, wears, you know, Brand A and will showcase a whole outfit of that. And people are starting to look to that as the real marketing and advertising. Because I'm looking at what they think at least is the real person looking, you know, dressing up and showing me what looks like it's in style, what's looking good, what's available this season. That type of approach means I can grab that person, 3D scan their body, pose them, put the garments on there, render it pretty real to the point that most of the population won't know it's not real. And as you said, make decisions on it prior to cutting a single yard of fabric. From that. That's pretty powerful and that's been typically just something that you would do for marketing campaigns or to showcase to the final kind of buyer what things look like. Now you're talking to the consumer. So you can do, you know, focused groups, studies to see what they like, what they don't like, vote for your stuff that you want us to, you know, to release, what's the next Yeezys or whatever they're gonna have in there. But I mean, that's the engagement, right? That's what you want. You don't want to make something and have it just stocked up on shelves where people are not buying it. You want to get something that everybody's just, you know, waiting in lines for. Yeah, absolutely.

Absolutely. Well, look, it's fascinating stuff. And if anyone out there hasn't heard of what Swatchbook are doing, you know, look them up because I think it's a great product and it's been an absolute pleasure to sit down and learn a bit more about it and have a chat about the industry with you. Awesome.

Yeah, it was great having this conversation. Conversation. Thanks for having me. Pleasure. Take care. Everybody get up.