What happens when anyone can manufacture anything they can see, whenever they like?
What happens if you can take a few snaps with the camera built into your phone, send it to some clever software that turns it into a 3D model, and produce it on the 3D printer sitting on the counter in your kitchen? There’s a long way to go, but we are heading in that direction. If you make physical things for a living, and you want to avoid going the way of the music industry, now is the time to start thinking about how to deal with a world of digital manufacturing.
We have explored what has happened to the products that have already made the transition from physical atoms to digital bits. It becomes easy to share them. It becomes easier for fans to become pirates. As it becomes cheaper and easier to distribute books and games and music and films, so fans find it easier and more desirable to share them, or to find ways to own them without paying. It becomes possible for competitors to find new ways to give stuff away for free and to find different ways to pay for the production by building close relationships with fans. Businesses and creators who make physical items – widgets, jewellery, kitchen utensils, spare parts – have been spared the impact of digital and casual piracy.
But not for very much longer.
*
Chris Thorpe always wanted to be a boffin. ‘When I was younger, I always really wanted to be a boffin. Maybe that’s why I did a chemistry degree – the lab coat came with the course.’1
A boffin may be a peculiarly British expression. It’s an egghead. A scientist of a particular breed. It’s the phrase used by tabloid newspapers whenever they talk about science, as in the headline the Sun used to report on proof of the existence of the Higgs Boson: ‘Boffins spot “God particle”.’2 Thorpe is hoping to be at the forefront of the new breed of boffin-entrepreneurs.
In November 2012 Thorpe travelled to the Bala Lake Railway in the mountains of North Wales. The railway is a charming narrow-gauge tourist railway laid on the trackbed of a former standard-gauge railway along the side of the lake. Thorpe was not just going to view the railway, though. He was going to see and record for posterity one of her famous engines, Winifred.
Winifred is a narrow-gauge steam engine built for the Penrhyn Slate Quarry in 1885. Her working life lasted eighty years, starting in the reign of Queen Victoria and ending in the reign of Queen Elizabeth, encompassing two world wars and countless technological changes. In 1965, the Penrhyn Quarry decided to sell its steam locomotives and three of them, including Winifred, were bought by Tony Hulman, an American businessman who ran the Indianapolis Motor Speedway. Hulman planned to make them part of a new museum he was building. The museum never happened and Winifred spent most of the next half-century sitting in dry storage in the shadow of the Indy 500 grandstands.
In 2012 she was returned, in almost the same condition she left, to Bala, home to four other similar quarry locomotives. Thorpe made a pilgrimage to see her. Not just to see her, but to record her in glorious 3D. He partnered with laser-scanning experts Digital Surveys who in the course of a single afternoon took fifteen detailed laser scans of Winifred and 1,000 high-resolution photographs. They took the data from the scans and images and, at a cost of just £4,500 ($7,000), converted it into a complex 3D model that could be used to reproduce Winifred at will. Which is just what Thorpe did. He placed an order with Sculpteo, a business that allows you to upload a model and have it printed by a 3D printer at their facilities just outside Paris. A week later he had a 1:25 scale model of Winifred, accurate to the bent brake handle and the dents in her dome.
The experiment with Winifred was part of the launch of Thorpe’s business, the Flexiscale Company. Thorpe uses modern surveying methods on real-life objects, like the laser scans he took of Winifred, to make precise three-dimensional plans. Flexiscale then splits the plans up into kits that are simple but aim to be fun to assemble, detail and paint. The kits are produced on demand using 3D printing. The thing that is special about Flexiscale is that a customer can order the kit at any scale they want. Every kit is printed on demand, individually. The 3D model is accurate at a 1:1 scale, that is to say full size. The only limits to the size at which the items can be printed are the consumer’s desire to pay and the physical capacity of the printer.*
On Flexiscale’s website today, you can buy a Dinorwic Quarry Slate wagon, a simple vehicle once used to carry slate around Penrhyn Quarry. Flexiscale has three scales available: a 16mm model costing £34.50 ($55), a 9mm scale costing £22.50 and a 7mm scale costing £14.50. All printed on demand and all using the exact same underlying geometry stored as a digital file on Flexiscale’s computers.
Thorpe estimates that the model-train market is worth £100 million annually in the UK alone. He hopes to disrupt the two major incumbents, Hornby and Bachmann, by harnessing a community of modellers and train fans to help him create, curate and manage a huge array of three-dimensional computer models that can be used to create replicas at any scale on demand. Thorpe is connecting two technological revolutions – the cheap cost of distributing digital files and the ability to produce cost-effective, physical items one-by-one using 3D printing technology – to change the way people think about making models.
As he does so, he will have to wrestle with the challenges that have faced the music industry for the past decade: if people love what Flexiscale does, they will want to share it. If the thing they want to share is digital, they can make and share a perfect replica, while still keeping the original for themselves. Casual piracy has just come to the manufacturing world.
*
Thingiverse is a website filled with 3D models. It currently boasts over 30,000 digital models that can be downloaded and manufactured using laser cutters, CNC (computer numerical control) machines or 3D printers. It offers a holder to turn your iPhone into an electronic picture frame, a drain trap that can be scaled up and down to fit the size of your plughole and earrings in the shape of Captain America’s shield. Many of the items can be adapted or customized to your exact specification.
Meanwhile, Defense Distributed, a Texas-based company run by twenty-five-year-old law student Cody Wilson, gained notoriety in 2013 when it announced that it had successfully fired the ‘Liberator’, a gun that had been created on a 3D printer. In the two days that the files were available on the web before the State Department asked Defense Distributed to take them down, they were downloaded more than 100,000 times.3 With those files now spread around the web, it is only a matter of time before they resurface. While gun experts remain sceptical about the danger posed by the Liberator and warn of the risk that it might explode in the firer’s hand, the control that governments have on access to firearms has just become a little bit weaker. To understand the importance of the Liberator files, it is helpful to understand how digital manufacturing works.
Digital manufacturing is a three-step process. First, you need a data file that contains the three-dimensional geometry of the object to be printed. Users can make these themselves in computer-aided design (CAD) software packages, they can scan the objects as Chris Thorpe did with Winifred or they can download existing files from the web. The data file is interpreted by software (step two), sent to the hardware (step three) and, hey presto, the object appears. It doesn’t always appear fully formed: the Liberator handgun comes in fifteen pieces that have to be assembled after manufacture. The final piece, the firing pin, needs to be metal and is not currently printed, but a nail that can be purchased from any hardware store will do the job.
A 3D printer is not dissimilar, in principle at least, to the ordinary inkjet printer that sits on the desk in my office. There is a print-head that moves backwards and forwards, left and right, just like an inkjet. To create objects in three dimensions it needs an additional motor that moves up and down as well. On some printers, the bed on which the object being printed rests stays stationary while the print head moves around it. On others, the print head stays still and the object is moved.
The term 3D printer covers many different manufacturing techniques. Some 3D printers such as the Makerbot use a technique called fused deposition modelling (FDM), an additive process where plastic is melted and squeezed through a small heated nozzle. The plastic hardens to form layer on layer to build the model. More expensive machines use lasers either to harden liquid resin in a bath (stereo-lithography or SLA) or to harden layers of powdered metal, plastic or ceramic (selective laser sintering or SLS). Machines based on lasers can use a wider range of materials than the plastic-extruding ones and have a higher resolution, but are much more expensive. The Liberator handgun was printed on a second-hand Stratasys Dimension SST that cost $8,000 while a Makerbot 2 can be bought new for $2,199.
3D printers can work in a variety of plastics, in metals like stainless steel and silver, and in ceramic. The market is changing so rapidly that materials are improving in quality, decreasing in price or both at a fast pace. Anyone starting out in digital manufacturing in 2013 can expect rapid improvements in both quality and price, year by year.
Alice Taylor is chief executive of Makielab, a London-based company that aims to use 3D printing to take on the might of Hasbro and Mattel in the market for dolls.4 Makielab allows users to design their own personal doll on a web page or tablet. They can change the shape of the doll’s face, its nose, eyes and ears. They can choose different skin tones. They can personalize their online doll to make it completely unique in the entire world. Then, for just £70 ($105), they can get the unique, personalized, 10-inch poseable doll custom printed and posted to them.
‘Moore’s Law is at work here’, says Taylor.* ‘A commercial grade printer can cost anywhere between $50,000 and half a million dollars, but the quality is going up, prices are coming down and the results are amazing. Even in the first year of Makielab, we’ve seen the quality of the Makie dolls improve substantially as our printing partners have improved their processes.’
Makielab’s vision is to have a virtual world or game where players can come in, design the virtual Makies, create new items for them, play or just hang out. That will be free, although there may well be charges for virtual items, the model that has worked so well for free-to-play games elsewhere. Some players will choose to make a physical Makie. Some of those will continue to spend money on physical items for their Makie: clothes, hats, accessories and so on. Makielab finds new customers through its website and tablet experiences (as well as through more traditional PR stunts like presenting personalized dolls to Prime Minister David Cameron and Prince Harry). Those customers who choose to spend will start by paying £70 for the basic doll. Superfans can buy multiple personalized dolls, and spend lots more money on accessories and accoutrements.
So Makielab is a Curve business. It grows its community through free digital content, and it allows those who love what it does to spend lots of money on things they value. Taylor and the team at Makielab are not fighting free or just targeting the expensive. They are embracing both ends of the Curve and everything between.
Taylor is under no illusions that this new world comes with challenges for intellectual property. ‘I think that one of the biggest issues is going to be copyright and trademark. Anyone will be able to pick up anything, scan it into a computer and replicate it easily. What does that mean for the person who made the original? I believe the solution is to design for piracy: most often it’s simply a service issue: pricing or access, for instance.’5
Taylor’s comment is one that everyone who manufactures anything should take to heart. When every physical object can be reduced to a digital file, availability is no longer the only consideration for end users. If they want something they can visit Thingiverse, The Pirate Bay or BitTorrent to download a legal or illegal version of it. For many that is a terrifying vision of the future, where manufacturers have no place and every home is a factory. But while we are heading in that direction, this is not going to happen to manufacturing at the same pace that it happened to the record industry – if manufacturers start now, there is time to adapt to be successful in the digital age.
Let’s start by considering what 3D printing and the associated digital manufacturing pipeline excels at, and where it is weak. If you want to produce lots of the same item very cheaply, traditional ways of manufacturing are still likely to be far superior to 3D printing. Injection moulding involves a heavy spend upfront to create the tooling for the mould, say $10,000. If you only produce one widget, you are paying $10,000 plus a few pence for the raw materials. If you making a hundred widgets, it’s $100 each plus the raw materials. By the time you are making a million, the initial set-up costs are a trivial fraction of the per-unit price. Injection moulding means that the underlying cost of a product like a Barbie doll is less than a tenth of the retail price, with most of the value being captured by the retailer and Mattel, the owner of the brand.6
With 3D printing, creating a single widget would be much cheaper than with injection moulding. Let’s say the first widget cost $20, compared with $10,000 plus a few pence in the injection moulding process above. The second widget costs $20. So does the third. So does the millionth. There are no economies of scale in this model. If you think of 3D printing as a direct replacement for traditional manufacturing methods, you would be right in thinking that it sucks.
The benefit of digital manufacturing is the new opportunities that it brings to do things differently. Digital manufacturing subverts the old order. It changes manufacturing in two fundamental ways: it changes the prototyping and creation of new products; and it changes what can be manufactured cost-efficiently.
In the days before digital manufacturing, it was hard to be an inventor. In his book, Makers, Chris Anderson tells the story of his maternal grandfather, Fred Hauser, a Swiss émigré with twenty-seven patents to his name. Only one of them ever saw the light of day as a product, an automatic sprinkler system licensed and manufactured by Moody and called the Rainmaster. Hauser patented his idea in 1943. It took seven years, lots of pitching and substantial legal fees before a product based on his patent reached customers.
Hauser was trapped by the tyranny of the physical. He had to negotiate with many potential manufacturing partners to find one who was prepared to turn his idea into a product. Each one had to do market research and feasibility studies to determine if there was a market for his product (although I’m sure some rejected it based on nothing more than gut instinct). Moody had to decide if the initial investment in the manufacturing capability for the Rainmaster was worth it. The success of Hauser’s product was in the hands of the gatekeepers who controlled the means of production: the factory owners and manufacturers.
That is not true any more. If Hauser wanted to give his invention away for free, he could upload it to Thingiverse and those early adopters with access to a 3D printer could make a sprinkler system for their garden in the course of a weekend. If he wanted to commercialize it, he could print it himself on a 3D printer. He could sell it via his own website and start to build a community of gardeners to help him to adapt and improve his product to meet their needs. He could take the sprinkler system to garden shows and fětes and sell his products one-by-one. He would get feedback from real users which would allow him to improve on his product with every iteration, and in the world of 3D printing, every single copy can be a new iteration. Some customers might want a small system for an urban patio, others an enormous set of sprinklers for a rural estate. They might pay a premium for Hauser to visit them to advise on the system and to install it. Hauser would have harnessed the Curve.
If that all sounds like a lot of work, that’s because it is. Nothing about the Curve says that you can sit back and make money for doing nothing. The ending of the tyranny of the physical and the erosion of the role of the gatekeeper mean that more people will be able to test their ideas in the open market than ever before, although finding customers will still be hard. Many will fail. Others will create viable businesses that would not have been viable at the scale needed under the old manufacturing model. A few will create huge businesses which will compete directly with the industrial titans of the twentieth century. Inventors and creators will spend more time talking to their customers, the people who actually use their products, to iterate on, improve and refine their ideas. Digital manufacturing will enable products to get better faster.
The second change is that personalized manufacturing changes what is cheap and what is expensive. In the traditional manufacturing model, it is the fact that every single product is identical that makes it so cheap. If the objective is to reduce the price as much as possible, it is critical that you make as many of the same thing as possible. If that means that each customer has to have exactly the same product as everyone else, so be it. That’s why Henry Ford is reputed to have said his customers could have their Model T any colour they wanted as long as it was black. It is the sacrifice consumers had to make to get a high-quality product at a very cheap price. These limitations are different for personalized manufacturing.
When you are manufacturing from a digital file to a 3D printer, you can make every single product unique. Gone is the need for expensive set-up fees and the requirement for scale (although the unfortunate corollary is that economies of scale have gone as well). If a computer is driving the manufacture of each single item, it can make each one different. The items can be complex too, because a computer can just as happily perform the calculations to print out a complex model of the Eiffel Tower as those for a simple cube. Inventors and creators will be able to make mistakes because they can fix the mistake before the next product rolls out of the 3D printer.
You get variety, complexity and flexibility at no extra cost. These properties are no longer scarce, and successful businesses will work out how to take advantage of this new abundance. On Thingiverse, there is a whole category of ‘Customizable things’. You can get collar stiffeners that are monogrammed and personalized for you. You can make a pair of dice with customized text on each face. You can manufacture a hundred key fobs and make every one of them different.
This is a very different proposition to traditional manufacturing. If you want to make tens of thousands of the same item, traditional methods are likely to be more cost-efficient. But consumers are going to start seeking out the personalized more. As we move through the journey from commodities (is it available?) to goods (how much does it cost?) to services (is it high quality?) to experiences (how will it make me feel?), manufacturing is only just beginning to offer that personal, social, emotional bond that is where real value will be created in the twenty-first century.
*
If you type ‘overdesigned lemon squeezer’ into Google and hit the image search, it is likely that half of the images that you see will be of the same product. A tall, spindly, bulbous blob on three slender legs, it looks like a three-legged arachnid or the tiny offspring of one of the invading Martians from The War of the Worlds. It is one of the most iconic pieces of kitchen equipment ever designed.
In 1989, Italian manufacturer Alessi asked Philippe Starck, the French designer, to come up with a new lemon squeezer. The result was the Juicy Salif described above, twelve inches of cast aluminium that was almost, but not entirely, unpractical. It’s hard to use on a kitchen surface given its height. It takes up a lot of space. The grooves are hard to clean. It is quite pretty, though.7
The Alessi lemon squeezer costs £48 ($72) from the online website of British department store chain John Lewis. A digital model of this iconic design is already available on Thingiverse and it won’t be long before I can print a bootleg copy, at any size I like, in resin, plastic or stainless steel. People who love Starck’s work will want to tell their friends. To share it. Some of them will send digital files containing everything you need to know to make one of your own in your very own home. In a stroke, manufacturers have a new intellectual property risk to worry about. They don’t have to worry about organized gangs of criminal counterfeiters creating knock-off assembly lines; they will have to worry about the piracy taking place in every house in every nation in the world.
They are not ready for that.
*
A Makerbot Replicator 2 is an amazing piece of kit. In December 2012 I went along to the Wired pop-up store on London’s Regent Street to watch one in action. The head moved across and around, pumping out layer after layer of gloopy plastic, depositing and building the 3D shape of a Christmas angel, perfect for hanging on the Christmas tree. I brought one home with great delight, marvelling at how I had seen this thing manufactured before my eyes. I showed it to my wife.
‘I’m not sure what you’re excited about. It doesn’t look as good as any of our other decorations. I wouldn’t have paid 50p for that in a market.’*
She was right, of course. 3D printing is still in its infancy. The Makerbot Replicator 2 costs $2,199. It is aimed at ‘prosumers’, those early adopters for whom the fact that you can manufacture whatever you like is the heart of the product, not the quality of its finish. Alessi, which can afford the significant upfront cost of creating a production line optimized for churning out thousands of Juicy Salifs, has nothing to worry about. Yet.
Clay Christensen is the author of The Innovator’s Dilemma, the classic work on how disruptive technologies are at first derided by incumbents before overtaking and ultimately destroying them. His book focuses on how many of the businesses that are destroyed fail not because they do anything wrong, but because they do everything right. They focus on quality. They deliver what their customers want and would pay for. They improve. Meanwhile, some scrappy little business has a product that costs a tenth of the price of that offered by the incumbent. It isn’t as good. It lacks features, or quality, or something that the incumbent believes gives it a long-term sustainable advantage. Over time, the upstart competitor gets better. It keeps its prices low but it adds features. It improves that quality. It starts to be a viable alternative to early adopters, and then to the mainstream. It knows how to deliver an experience in a way that the incumbent can’t match. One day, the incumbent wakes up and realizes that its competitor can make an equivalent product for a much lower price, and goes out of business.
That’s where we are with 3D printing at the moment. To the management at Alessi, comparing their Starck-inspired kitchenware to a knock-off 3D-printed lemon squeezer, it might be hard to imagine the Makerbot as a real threat. The whole point of an Alessi lemon squeezer is that it exudes quality. Quality of design. Quality of raw materials. Quality of manufacture. A rip-off 3D print in striated plastic doesn’t fit the bill.
Meanwhile Makerbots and other 3D printers will get better. Someone will take the 3D image of the Starck classic and improve it. Maybe they’ll make its legs a bit shorter so you don’t need to raise your shoulders or stand on a chair to squeeze a lemon. Maybe they’ll experiment with different grooves that clean better in a dishwasher. Maybe someone will adapt one that will fold, or come to pieces for easy storage. I don’t know what they will do.
Which is the heart of the threat to the traditional business. Most businesses that make physical products are used to being one of the few organizations with the time and resources to experiment with a product, to take the risk on a production run and to bring it to market. In the world of 3D printing, experimenting becomes incredibly cheap. Ordinary consumers become designers and manufacturers. All those tinkerers and makers in the general population get to tinker and experiment with new designs. Many, maybe most, of the alterations that are made will make the product worse. Some, a few, will improve it. Those improvements will spread through the web as people search for ‘overdesigned lemon squeezer that works’. Alessi’s lock on what gets made will be over.
In 2000, to celebrate the tenth birthday of the Juicy Salif, Alessi launched a limited edition. Ten thousand pieces were produced. Each was gold-plated. Each was individually numbered. You can currently buy one for $640.8 It has no use as a lemon squeezer, because the citric acid in the lemon juice would discolour the gold. It is an expensive artefact, much like the Ultra-Deluxe Edition of Ghosts I–IV that Trent Reznor released to his biggest fans.
This, in fact, may be Alessi’s salvation. Instead of being a volume player, trying to keep the price fixed while driving up revenues by selling more units, Alessi could go the other way. It could give away the designs for its products, which is not much of a concession, given that anyone with a hand-held scanner and the product in question will be able to produce a 3D geometry file in a matter of moments. It could still make the basic product for those who don’t want the hassle of making it themselves. It could have a version available at $100, $1,000, $10,000, each tapping into some emotion of status, exclusivity, belonging, self-expression or whatever.
In short, Alessi will move away from the business of making things and move towards the business of making people feel. That is how to fight casual piracy in the internet age.
*
Bill Gates, the founder of Microsoft, says that we overestimate the short-term impact of technology but underestimate its impact in the long term. That is where we are with 3D printing. In my lifetime I expect that as many homes will have 3D printers as currently have 2D paper printers. The process will be slow. First, the early adopters will take their expensive playthings and use them as a hobbyist would, simply for the fun of doing. Next, local convenience stores will start having their own 3D printers so every town will have its own. Just as every town once had a shop that processed camera film, it will instead have a 3D printing shop, in many cases occupying the same premises. In the same way that most of us could, if we chose to, print out our holiday snaps on our own home printers yet often conclude it is easier to upload them to a website and collect the finished, high-quality prints from a local store or receive them in the mail, we will first get exposure to 3D printing through third parties.
Perhaps we will need a replacement part for a broken cot or child’s buggy. A drawer handle that is no longer in production has broken, and we either have to replace every handle in the kitchen or find a way to get one matching element. A personalized key ring as a gift. Whatever it is that we want, something will eventually make most of us consider 3D printing as a solution to some need.
The office supplies company Staples has already announced that it will be rolling out 3D printing services through stores in the Netherlands and Belgium in 2013. That is how it will begin. Eventually, 3D printing will be all around us. Exactly how this will pan out, nobody knows.
We do know that we are in the first experimentation phase, as entrepreneurs test the market, searching out what customers want and will pay for through trial and error. There will be failures and bankruptcies, because that is how the capitalist economy sorts and filters the ideas that can work at scale from those that cannot. 3D printing, or some similar form of personalized, individual manufacturing, will be a substantial part of the Western world’s manufacturing capability within the next twenty years.
Hopefully, traditional manufacturers will learn the lessons of the record industry. They won’t seek to sue their customers. They may try to delay the onslaught of digital manufacturing in their traditional businesses by using Digital Rights Management and litigation, but in the end, it will be their competitors figuring out how to use the power of free to reach more customers that will be their biggest threat, not piracy. They will learn to adapt.
The Curve suggests three linked strategies for building a business in the digital age. Use the cheap distribution of the web to find customers; use technology to ascertain who the best customers are; let them spend money on things they really value. Manufacturers have one big advantage here: there are many people who are unlikely to want the hassle of personalized manufacturing. That means that the web is less disruptive for physical industries, in that it doesn’t eliminate the existing business, but also more challenging, since manufacturers will need to adopt two totally different strategies simultaneously.
The first and most urgent order of business is to start building relationships with end users. Manufacturers might use content marketing to draw users to their websites; the ways they build relationships with their audience will become more sophisticated.
A business which makes kitchen utensils will provide all sorts of free advice on kitchen techniques. A tool manufacturer will do the same with woodworking or DIY. Toy companies will have entertaining websites (many, such as Lego, already do). They will help people cook better, or make better, or have more fun. They will use content to become more relevant to the lives of the people who buy the things they make.
Some will give away the designs of their products, knowing that few 3D printers in the world will be able to match the quality and finish of the products that they manufacture themselves. By sharing, they will start the process of connecting with their fans. They will upload digital files to sites like Thingiverse and The Pirate Bay and sell physical products on sites like Etsy to find an audience who will test out their products at their own expense. They will embed elements in their digital files that encourage users seeking a different experience to visit the manufacturer’s website or to buy one of their products. They will encourage the use of 3D printers as a prototyping tool for users, knowing that quality, brand, scarcity, status and all the other tools at a marketer’s disposal to create value still exist in a world of personalized manufacturing. They will use customer relationship management to put connecting with their customers at the heart of their business, not secondary to it.
Manufacturers will make less money at the cheap end of the market and much more at the expensive end. They will have to become experts in creating desire and in using technology to identify those customers who will be freeloaders for life and those who have the potential to become superfans. They will also cease to be business-to-business businesses. Every manufacturer whose products are sold at retail will need to start building one-to-one relationships with their customers. They will stop relying on a long chain of wholesalers and distributors and retailers. They will have to learn new skills.
They will have to deal with the consumerization of business.