Home Featured Conversations: Eaton Donald, ExactFlat

Conversations: Eaton Donald, ExactFlat


Just last month, Eaton Donald, President & CEO of ExactFlat (specializing in 3D to 2D patterning solutions for a host of industry sectors), dedicated some time to talk to our Editor. Eaton speaks passionately and with penetrative insight about the issues we face in the fashion industry, and exactly what this transformation to digital will mean for everyone. Our longest conversation within this series, this piece is a must read for anyone interested in challenging the norm, and moving with the industry.

Like all our ‘conversations’, this piece has been transcribed verbatim, to ensure Eaton’s thoughts were never condensed.

Name: Eaton Donald

Occupation: President & CEO, ExactFlat

Likes: Keeping it real, keeping it simple, full moons on bright nights, tough demanding mentors

Dislikes: The smell of fresh coffee; blech, unresponsive touch screens – that’s it, no more dislikes.

Words to live by: “Simplicity is the ultimate sophistication.” – da Vinci

Lydia Hanson: Eaton, to begin, can you give me a quick overview of ExactFlat?

Eaton Donald: ExactFlat is software for sewn products. We provide our technology as an add-in to leading 3D CAD companies, like SOLIDWORKS, Rhino 3D and Autodesk. In our opinion, manual processes used for the last 200 years, are in their sunset period. The era of digital design, digital patterning, and digital production has already begun. Our goal is to assist in the textiles industry’s overall transformation to digital.

We have made core contributions in industrial sectors in this transformation. Apparel markets are a key priority for us going forward.

Lydia Hanson: You mention other markets before fashion. Is there anything that’s happening in these industries – automotive for example – that can be directly applied to fashion?

Eaton Donald: 3 aspects come to mind: First, pattern makers are disappearing. They are exiting the profession at a rate that’s faster than new patternmakers are entering the profession.

The second aspect is what to do about it: invest in training, outsource, or automate. There has been a decades long trend in industrial markets towards automation. In the case of textiles this implies digital process and new kind of pattern maker digitally proficient in 3D to 2D workflows.

The third aspect is a more fundamental change. It is this phenomenon towards mass customization, and the compression of the time between design and product retail. Mass production has offered lower unit costs. However, that is only true when inventory is fully sold. The problem is it may not, and often does not, get fully sold. That introduces risk in the system. The risk of unsold inventory gives rise to discounting and even dumping product.

The response has been a move away from mass production to shorter production runs. Products that are produced “on-demand” and customized to end markets are now more profitable. The 3D to 2D digital process in industrial markets can support this change well. This is the big lesson for apparel and fashion markets. On-demand production, and zero inventory production is a viable option today. It is no longer necessary, nor is it economic, to invest in over-production, followed by discounting, followed by discarding unsold product.

If we look at this from the consumer or retail perspective, each retailer wants unique product for their channel and companies have been adapting to this reality. Delivering design variation by retail channel is essential for many manufacturers

Lydia Hanson: How do you see this need to deliver different design variations changing the manufacturer’s process?

Eaton Donald: You must design with variation in mind: core design elements, and design elements that are open to variation.   Companies that get this right, will produce compelling products, at a lower cost, with less waste. So, you would offer customers “variations” within a prescribed set of options. In order to reduce the inventory distribution complexity, you would then make the specific product on demand after the customer has bought the product. This is a major shift. It will also be an essential shift.

We used to write a book and print five or ten thousand copies. That is a lot of money up front. Then, 2 or 3 months later, our books are in retail and that’s when we get paid. Now, we have this print on demand process where customers buy the book first, which triggers the manufacturing of the book next, and then it goes out.

For this to be successful, it requires a compression of the time between the customer buying it, and the product being made. That time is now compressed to almost zero in the book industry and the music industry. That’s what’s coming to fashion. What we see coming to fashion from our industrial markets is a shift to letting the customer buy the product first, which triggers a manufacturing order to make it and send it out.

This, of course, is going to split the industry: those that do, and those that don’t. Those that don’t will have to have a good economic reason why they will stay with the mass production model. And they will need to have a way to compete with those that move to the economically advantageous produce on demand model.

We don’t think that fashion is going to be able to exist like it has been: designing products and showing collections 6 months before they’re in retail, risking that a collection might be out of sync with weather patterns if there is a cold spring or a warm winter. This mass production model creates risk, and people have to get paid for risk. So the fashion industry puts up their prices because they know there’s going to be unsold product – which means higher prices for the consumer, risk for the manufacturer, and risk for the retailer. This is just a broken model. And digital transformation is going to re-shape all of that.

Lydia Hanson: Mass customization is obviously just one facet of this move to digitalization, and this move to everything being about the consumer and the consumer experience. In terms of fashion, do you think mass customization is a positive?

Eaton Donald: Well, it’s a paradoxical phenomenon. We all want to belong to a group, but we all want to be unique. So mass customization is going to come, but here’s how we see it playing out…

There are going to be an influx of new competitors. Upstarts poised to create fashion brands. In the book publishing world, print on demand ushered in an explosion of publishers. This will happen in fashion, and existing brands will need to compete with a long “tail” of new producers. You look at, for example, people with a fashion-centric Instagram following of 100,000 or a million followers. They are ripe for creating a brand for themself, but what they lack is the expertise and the tools to do so. The digital transformation will democratize that production process; it will give access to these people. Fashion houses and branded apparel have known that they have a problem for a while now.  Their margins and their way of doing business are all constrained by processes of planning the line, showing the line, making the garments and then managing and selling the inventory.  It’s a tremendously risky guessing game and it’s just not sustainable.  There are threats on the horizon when it comes to technology simplifying the process dramatically. Once that happens, anyone with a big Instagram following, well the next step is for them to “self-publish” and create their own brand and distribution. That means that they can compete directly with the larger more established brands.

Here is another example of a new non-traditional channel. You’ve probably heard of product placement in TV shows and movies. Well, you can imagine a scenario where you have a TV product that is funded in part by the ability to sell the clothes the characters are wearing. The character’s fashion could be a distinct character in that production. Television as a merchandising and retail channel is an incremental add-on to product placement. If you like what you see, you like the fashion sense, why not buy the clothing?

The real change is going to be people who have a good connection with their following. And clothing and fashion is such a central part of our own identities. And those people we idolize or aspire to become, have their own fashion sense and identity. Why not see a TV show, go on the show’s website, pick that piece of clothing, and have it produced and sent out? Now, the TV show is unlikely have inventory; they just won’t because it’s not their business. But, to connect to a rapid on-demand manufacturing system based on a couple of clicks here to get something shipped out the next day, we think that today’s technology offers that as an option.

Now, we’re only at the beginning of imagining these kinds of things. It will take bright minds to bring the full vision into reality. But, in our company we try to think five or ten years in the future: where are people going to be? What will the state of technology be? What will it allow customers to do? There will be a shift from customers orbiting around a brand to the brands orbiting around the customer. They will both co-exist; one won’t replace the other, but one will be a force that the other will have to reckon with.

Digital design, digital patterning, and digital production will change everything: products, customers, competitors, pricing, business models and opportunities.

Lydia Hanson: Fantastic points. I supposed instead of us having customer loyalty, in the future we’ll be seeing brand loyalty to the customer. There are all of these changes happening in our industry, and even in recent years there’s been a push towards 3D that perhaps wasn’t there before. Am I correct in thinking that you guys specialize in 3D to 2D, rather than the other way around?

Eaton Donald: Absolutely. For us the workflow is simple: 3D CAD for digital design, 3D to 2D digital patterning, and finally digital production on a textile printer and CNC cutter. This is the apparel on-demand workflow.

We started our existence in industrial markets where 3D CAD is well established. You have a conceptual design, which can be a whole variety of things – a paper sketch, a digital drawing, a photograph, a laser scan – and from that you’ll create a 3D design. The 3D design is different from the conceptual design in that it will have dimensions and it will be accurate in terms of perspective and dimensions. So, there’ll be a proportioned width, height, etc. From there, we transform it from 3D to 2D. And here’s the magic of what we do well. We translate patterns in 3D to patterns in 2D accurately. We can do the same with graphics and textures applied to those patterns; flatten them accurately with images perfectly applied.

This is a fresh break from the 2D-to-3D-back-to-2D block centric approach that still requires manual patterning.

Lydia Hanson: Simulating what’s really happening to something as unique as fabric sounds like a difficult problem to address. How do you intend to make something like this work?

Eaton Donald: We’ve spent many years developing digital simulations for fabric, and we know how fabric stretches under tension and how it bunches under sag. All of the mathematical algorithms and simulations for how fabric behaves under certain conditions, we’ve embedded in our technology.

We are able to take a 3D form and flatten it so that the flat patterns accurately reflect what the designer had intended. We have already done this in many demanding industrial environments with very exacting tolerances.

However, as you know, some 3D shapes are not manufacturable as they are – you have to put darts in, or relief cuts – and we’ve figured that out. Sometimes you want your piece to be under tension; for example compression clothes or under garments. We have included the right tools and feedback for skilled digital patterns makers to be proficient.

We call what we do ‘digital patterning’ and that’s in contrast to the current 2D-to-3D-back-to-2D way of working. It brings to mind Henry Fords statement If I had asked people what they wanted, they would have said faster horses.” The current state of the industry is “the faster horse.” Our core suite of technologies is going from a 3D design to a 2D flat pattern that you manufacture and it fits perfectly in physical form.

Lydia Hanson: Very interesting. So, you’re idea is essentially the reverse of what is typical today when it comes to 3D working and 3D visualization. It all sounds very complex, as almost all technologies are.

Eaton Donald: I can actually simplify it for you. Suppose you take a piece of fabric and drape it over somebody’s shoulder, you’re naturally going to get some sag and some wrinkles. So what do you do? You would take your hands and smooth those wrinkles out. That’s exactly what our technology does, and it’s called ‘digital smoothing’. And it’s analogous to what a human being would do when you see some sagging or some bunching in a piece of fabric that’s undesirable – you literally put your hands on and spread it out.

Now, suppose you have the opposite – suppose you have a piece of fabric that is under strain. What happens when you pull a piece a fabric? It’s going to get longer; it’s going to stretch. And what also happens is that it gets narrow in the orthogonal direction. You can really see this if you pull a rubber band – it gets longer but it gets skinnier in the middle. So, we have to make sure that the elongation is within the amount that we allow, and the narrowing is also within the amount that we allow. When patterns are made manually, by eye, it may take 5, 10, 20 or more iterations to get it right. We have automated that process.

It is very complex mathematics, but in concept it’s really quite simple. Our latest innovation extends this to graphics and textures. For apparel designed with graphical images or textures it is quite simple. Apply images and textures to 3D designs. Flatten the patterns from 3D to 2D. Our technology creates flat patterns with the images and textures perfectly applied in 2D. Once flattened, these images will flow accurately across seam edges. This gives designers new possibilities for mass customization of apparel on-demand since it was prohibitively expensive and in many cases impossible to do before.

Lydia Hanson: That’s great. There has obviously been a lot of development of your product(s) throughout your journey. When you are developing something new, how difficult is it to develop your product to cater to multiple industries that will have different characteristics to work with?

Eaton Donald: [Laughing] Well, it’s not without challenges, as I’m sure you can imagine.

We have developed considerable expertise over the years. We know how to deal with rigid, semi-rigid, flexible, stretchy, and non-stretchy objects. We can handle all manner of geometric shapes. It does not matter if it is a car seat, a pair of pants, a knee replacement, an aircraft wing, or a satellite – we can handle it all.

So, we started with rigid objects, and we got the 2D to 3D transformation solved. Then we went to flexible objects, and got the 2D to 3D solved there. Next was non-woven, so vinyls and things like that. Then we went to wovens – which have a different stretch characteristic and attributes – and solved that. Then we went to things like composites – which again have different characteristics, because they don’t stretch but they have sheer – and we solved that.

It took many years with a team of scientists in multiple disciplines – mathematics, computer science, fabric science etc.

Lydia Hanson: I’m assuming you have a lot of cooperative partnerships, and connections to vendors of cutters and plotters and things. Do you currently link to digital printing systems?

Eaton Donald: Ah, that’s what we’re getting into now. Now that we have the 3D to 2D graphics, our vision is this: you design a garment, you attach an image or texture to it, you flatten it and that output goes to a digital textile printer and a cutter. 100% digital workflow, no manual intervention, and you’re able to create variations very quickly.  This is our vision: complex cut and sew apparel with beautiful designs ready to print on demand on a digital textile printer. It will reduce waste by nearly 90%, lower cost and give consumers great looking and great fitting clothing.

We are an ecosystem player. Our plan has always been to integrate our technology into points of the workflow best suited for our customers’ needs. This includes both upstream 3D CAD and design, and downstream digital textile printing needs. Currently our output files are compatible with all digital printers and all cutters.

Lydia Hanson: Amazing. You seem to be doing a lot of development around fashion and apparel. Forgive me if you’ve already mentioned this, but how long have you been catering to our industry? And, in terms of clients, have you seen a good reception with apparel?

Eaton Donald: We certainly find that the fashion and apparel industries are aggressively experimenting. They want a solution. For the longest time, our feeling was that the cost of not adopting digital technologies was zero, so they could wait. Now, I think the industry perceives that if they don’t shift into digital they will see bigger and bigger problems.

We have done a lot in related fashion markets: accessories, handbags, and technical apparel. With respect to consumer fashion and everyday clothing, we’re just starting to make a meaningful penetration. We have high fashion brands, which seem to be the most aggressive. These are companies that have both printed items on their fashion accessories, as well as conventional fashion items. They are, right now, trying to get an understanding of what the workflow is so we’re giving them our perspective: concept, digital design, digital patterning, and then digital output.

As this overall transformation to digital unfolds, we have different players in the workflow – the digital printing player, the 3D CAD player, companies like us that we would call digital patterning, there’s conceptual design software, virtual prototyping software and so on. What’s happening right now is that the eco-system is coming together; it’s evolving to the point where you have better integration between the various pieces, and the whole technical eco-system is becoming more complete. As the eco-system comes together, you’re going to get the mainstream coming into it. Some people were left out before because they didn’t have the output, or didn’t like 3D, but now with the whole system coming together we think the industry is going to start to adopt it on an accelerated basis.

It’s an exciting time. If you’re looking to get into the fashion market, this is when the new kings and queens are made. We have a change in the way the work is done, a change in the way you go to market, and a change in the relationship with the consumer. All of that change is open to new companies coming to take advantage of it.

Lydia Hanson: I couldn’t agree more. It’s very rare to speak to any retailer or brand these days that doesn’t have a digital transformation roadmap. Like you said, there are so many different players within this digital landscape coming to market with solutions you never would have thought possible just a few years ago. How do you keep the sheer volume of change from overwhelming the client?

Eaton Donald: We must anchor ourselves in the problem we are solving. Then choose the best solution. Start with the problem. Second, define the optimal workflow. Next, look at the technical enablement of that workflow using various tools. Lastly, put together an implementation road map. Pick the easier and more beneficial aspects to address first. Improve over time. It can be a lot of change, but should not be overwhelming. If it feels that way, put together a small team and experiment.

The worst-case scenario is doing nothing. Change and indecision are not good bedfellows. When you do nothing you perpetuate your problems. In my experience this does not serve the interests of your customers.

Lydia Hanson: One question I have – for which the answer may very well be “nothing” – is whether there are any limitations you’ve come across in your offering?

Eaton Donald: There are limitations. When designing in 3D, designers must define the pattern boundaries. Pattern shapes are a result of designer intent so you must know what you want.

The better quality your 3D model, the better the flattening and downstream processes. Issues can start with the wide spectrum of 3D models. They can be designed natively in the program you are using. They may be imported or the result of laser scans. In these instances, there may be preparatory work to get the 3D model to accurately reflect your intent. Vendors, freelancers, and outsource partners may work in different environments, all of which may impact the speed of your workflow. Products made from layers or very complex designs with many pattern pieces also have an impact on workflow speed. Designers have to manage this complexity. There are certainly design tools and product development strategies to meet these limitations. Another current challenge in fashion and apparel segments is the lack of experience with 3D design.

What we expect as the industry matures, is that there will be libraries that develop just like there are in the solid, mechanical world. Nobody models a screw today – you have a library of 10,000 screws to pick from. If you imagine modeling a collar, a cuff, or a pocket, there will eventually be libraries for these things, and you’ll pick pieces from different libraries to assemble. The big limitation right now is that we just don’t have the fashion and apparel industry trained in 3D. That’s a major barrier.

Lydia Hanson: I think that’s a barrier that fashion is well aware of – it certainly crops up a lot in our conversations with key players.

We’ve spoken a lot about your developments over the years. Are there any other plans underway – that we haven’t touched upon – you’re able to share with us? In the next year, or the next decade.

Eaton Donald: Our plan is to actively partner with other industry participants to make the eco-system of technologies less complex, and more easily adoptable by the fashion and apparel industry. We’ve done this already in industrial markets; we took our whole product and integrated it into mechanical CAD software so the customers didn’t have to go out into other environments. We’re in active discussions with 3D CAD, digital printing, with cutting table companies, with complimentary tools for bringing products to market. And this will make the offerings much more easily adoptable by the industry.

We do have plans for our technology. We are getting a lot of requests for an API, rather than selling our software through our own distribution channels and partner distribution channels. We think the 3D to 2D problem is a difficult one to solve. There are many 3D to 2D technologies out there, but we haven’t seen any that makes production ready flat patterns like ours does. We feel that’s a unique aspect, and we may develop an API to let other people innovate on top of it, rather than restricting it to what our team can get out.

From our perspective, we want to go from the 3D to 2D downstream to the cutter. That’s where our focus is right now. With respect to the upstream aspects… we aren’t prioritizing developing more hooks into the 3D technology. We want to make sure that the 3D designs – from wherever they come – can get to their output destination as quickly and as easily as possible.

Lydia Hanson: It’s great to hear you say you’re considering developing an API. There is an issue inherent with technologies – from various markets – being ‘closed’. If you’re a retailer or a brand and all of your solutions don’t ‘speak’ to each other it obviously makes it very hard.

We’ve covered an awful lot of ground but, before we go, is there anything else you’d like to mention?

Eaton Donald: Our society has allocated a great deal of talent in support of creating beautiful and functional clothing. This has made meaningful contributions to our lives. However, it comes with some problems too. The financial risks of mass scale production have put the industry in an inefficient cycle of overproduce, discount and discard paradigm. This is like lighting your investment dollars on fire to cook your dinner and heat your house.

The digital transformation can change all that and it can do it now. Waiting comes at a cost. Our goal as a company is to be an advocate for the transformation to digital, and we have technology to support the change. Anyone in the digital transformation workspace… we want to work with.

Lydia Hanson: I fully support those points. And there are a couple of things you’ve discussed today that our other recent interviewees have touched upon as well. It’s great to know that businesses from all different aspects of the industry have aligned views on so many things.

You can keep up with ExactFlat here, and find out even more about them here.

This marks our final conversation for this series. Take a look at those you might have missed here.

Lydia Mageean Lydia Mageean has been part of the WhichPLM team for eight years now. She has a creative and media background, and is responsible for maintaining and updating our website content, liaising with advertisers, working on special projects like our PLM Project Pack, or our Annual Publications, and more.Joining mid-2013 as our Online Editor, she has since become WhichPLM’s Editor. In addition to taking on writing and interviewing responsibilities, Lydia has also become the primary point of contact for news, events, features and other aspects of our ever-growing online content library and tools.