Brick by brick: How LEGO Rewrote the Rules of Innovation and Conquered the Global Toy Industry - BusinessNews Publishing (2014)

Part II. Mastering the Seven Truths of Innovation and Transforming LEGO

Chapter 8. Attempting a Disruptive Innovation

Aiming for the Stars with LEGO Universe

This is the largest and most complex project that LEGO has ever undertaken. It touches every part of the company.

—Mark Hansen, project lead, LEGO Universe

THE EMAIL FROM DENMARK WAS TANTALIZINGLY CONCISE. “There was just one line: ‘Would you be interested in working on an online world project for LEGO?’ ” recalled Scott Brown, the cofounder of NetDevil, a developer of massively multiplayer online (MMO) video games, where thousands of players connect, cooperate, and compete in a virtual world. The NetDevil studio, situated in Louisville, a suburb north of Denver, Colorado, had just begun casting about for a big brand with enough muscle to help them create and market a new MMO game. They never imagined the opportunity would come winging across five thousand miles of ether, from the Danish maker of innocent little plastic blocks.

“When we got the email from LEGO,” said Brown, “we were all like, ‘This is it!’ ”

It was October 2005, when LEGO was back in the black. By year’s end, the company would post sales of DKK 7.0 billion ($1.2 billion) and a pretax return on sales of 6.5 percent; in 2006, sales would rise only slightly, to DKK 7.8 billion ($1.3 billion), but profitability would more than triple, to 20.2 percent of sales. Revitalizing core brands such as LEGO City and DUPLO had helped the LEGO Group boost its earnings; the relaunch of Mindstorms exceeded the company’s most optimistic forecasts.

By refocusing 90 percent of the company’s resources on enhancing its core products, Knudstorp and his senior management team had clearly put LEGO on a path to profitability. But the journey had barely begun. To sustain its momentum, LEGO couldn’t simply refine and adjust its product portfolio. The third and final phase of “Shared Vision,” Knudstorp’s road map for transforming the business, set a goal of markedly increasing profits and market share, all in a bid to make LEGO “the world’s premier toy company.” That meant creating visionary toys that changed the way kids play. To build new growth drivers, LEGO would once again seek out untapped, blue-ocean markets. (We will trace that effort in the next chapter.) And it would attempt to unleash a disruptive innovation—a low-end, low-quality product that improves over time and eventually upends the industry incumbents.

In a New Yorker profile, Clayton Christensen cited the pint-size transistor radio as a prototypical example of a big, disruptive innovation. Launched by Sony in the late 1950s, the tinny, tiny radios, using the then-new technology of the transistor, had nowhere near the same sound quality as the big vacuum-tube RCA and Zenith consoles found in many middle-class homes. But the radios’ easy portability and low price made them a hit with teenagers. By the time the radios’ sound quality improved enough to pull in adults, RCA and Zenith were too far behind to catch up.²¹

In 2005, the LEGO Group’s brain trust believed the digital brick could be another disruptor, just like Sony’s little radio. An online LEGO game, where kids from all across the globe could connect and create their own virtual LEGO worlds, would offer a low-cost alternative to plastic bricks. And as the technology improved, so would the play experience, potentially pulling in the masses of kids who loved gaming but were indifferent to brightly colored plastic blocks. Digital bricks might even become as ubiquitous as plastic bricks. Based on those insights, LEGO decided it would disrupt its own business model before another company did, and thereby create a new platform for growth.

Hence the email to NetDevil, which kicked off one of the company’s most ambitious bids to create a break-the-mold play experience. Developing the game, dubbed LEGO Universe, would consume countless person-hours and $30 million in start-up costs. Along the way, LEGO discovered that of all the seven truths, the toughest challenge was to launch a disruptive innovation. Unfortunately for LEGO, it failed to heed the bright red warnings that it was taking a flawed approach to disrupting its plastic brick business.

First Warning: Don’t Jump into a New Business Before You Understand It

The LEGO Group’s foray into global video gaming, an industry of which it knew little, was fraught both with risks and with potentially outsize rewards. As Universe’s development team imagined it, the concept was decidedly immodest: millions of children from across the globe would enter an online world, interact with other kids through avatarlike minifigs, do battle with a dark force called the Maelstrom and its evil architect Baron Typhonus, and collect bricks with each successful mission. Because it was a virtual world, kids could do almost anything they could imagine: fly their minifigs over mountaintops, leap across vast canyons, or drive a car into a head-on collision with another car. Best of all, they could use their stashes of virtual bricks to build castles, dragons, spaceships, and forts whose scale and complexity far exceeded anything they could assemble with plastic bricks. At least, that was the idea.

To surmount the difficult technical challenges of creating such a vast virtual world, LEGO would have to seek out and successfully collaborate with an unfamiliar partner such as NetDevil, which had deep expertise in developing MMO games but knew nothing about building with the brick. Creating and launching the game would directly or indirectly involve more than 350 stakeholders across LEGO—staffers from the corporate IT department, marketing and sales, LEGO.com, product design groups, operations, finance, and customer service, as well as most of top management and thirty outside suppliers. That was an early sign of trouble, as it’s tough for a bloated project team to create a low-end, disruptive product. At the outset, LEGO estimated it would take three years to develop the game. It ultimately took five years and three missed deadlines.

“Building an MMO is like launching a space shuttle,” said LEGO Universe chief Mark Hansen. “It’s the largest and most complex project that LEGO has ever undertaken. It touches every part of the company.”

At first glance, Hansen seemed an unlikely choice to be the point man for bringing LEGO Universe to life. Whereas many video game developers have the beanpole physique of teenage skateboarders and the clothes to match, Hansen stood well north of six feet tall, was built like an NFL linebacker, and preferred polo shirts to tie-dye. In a company where the majority of the managers are Scandinavian, Hansen was an American, a former Navy SEAL who spent twelve years in the United States military.

Hansen, however, was a bit of an old LEGO hand. He was the original architect of LEGO Factory, which bridged virtual design with real-world construction by allowing anyone to create DIY LEGO models online that could then be ordered as a set for assembling offline. Moreover, his imposing presence frequently inspired others to get things done. Those were qualities that served him well in his role as the force behind LEGO Universe.

Some LEGO executives required a good bit of convincing that investing prodigious amounts of capital, talent, and executive attention would produce a breakthrough business. They took the view that the video game market was a big sinkhole, given that roughly 70 percent of new launches failed to do better than break even and just a minuscule 3 to 4 percent of MMO games managed to soar into the rarefied world of certifiable megahits.

By launching Universe, LEGO would face off against media giants such as Disney—and, later, Nickelodeon and the Cartoon Network—whose marketing platforms let them kick out new MMO game titles every six to twelve months. LEGO, on the other hand, had never produced a single successful video game for solo players, let alone a 3-D game that could simultaneously handle thousands of kids the world over.^* In fact, more than a few LEGO managers were still smarting from the pounding they had endured during the Darwin debacle, their previous big push into the digital world.

Nevertheless, Hansen and a dedicated group of like-minded managers kept pressing the case. Their pitch rested on three reasons for launching LEGO Universe.

First, developing Universe would evolve the LEGO System of Play on a vast scale. In the online world that LEGO envisioned, kids would not only build and create but socialize. Universe was really about bringing legions of kids into the LEGO play experience, where they could engage in construction, competition, and collaboration, all in a safe online environment.

Second, the business logic for gambling on Universe was persuasive. Although most MMO game launches quickly fizzle, there are some that strike it very, very hot. One of the most popular titles among MMO games, Blizzard Entertainment’s World of Warcraft, wowed the video gaming industry by capturing twelve million monthly subscribers and generating more than $2.2 billion in subscriptions between 2005 and 2012. With more than twenty million unique visitors clicking into LEGO.com every month—by far the most in the toy industry—LEGO had the magnetism and the hosting capacity to build a massively successful online world of its own. And once LEGO did the heavy lifting of developing an underlying code base that would allow customers to create whatever they could imagine, LEGO could leverage the Universe platform to regularly launch new MMO titles, just like Disney.

Finally, there were the disruptive possibilities that came with launching a digital building platform. A team of software engineers had already developed the LEGO Digital Designer (LDD), the computer-aided design program that allowed customers to go online and create one-of-a-kind LEGO models using virtual LEGO bricks. LDD, which was at the heart of LEGO Factory (later renamed Design byMe), was a powerful tool, but it was difficult to use. Any four-year-old can snap together two plastic LEGO bricks. But even twelve-year-olds who used the Digital Designer found it difficult to build anything complicated. Like the transistor radios of half a century ago, the LDD was an inferior technology with lots of upside potential. If the first iterations of Universe proved good enough for early adopters, LEGO just might capture greater numbers of kids as the virtual play experience improved.

After nearly forty meetings, the LEGO Group’s brain trust felt confident enough to give Universe the green light. So it was that in the spring of 2005, LEGO began to take its first big stab since Darwin at building a digital future.

Unbeknownst to the NetDevil founders, theirs was not the only company that LEGO had reached out to. LEGO identified fifty-one MMO game developers throughout Europe, North America, and Asia; Hansen and a technologist visited every single one of them. It made for an exhausting—and exhaustive—lap around the world. Hansen met the NetDevil team in their Colorado headquarters on an early November morning; he then flew to a meeting in Texas and finished in California that night, in what was a typical day on the road. But to Hansen’s mind, there was no other option. To find the right partner, he needed to get an up-close look at every potential team and its technology.

“We saw every part of the game industry,” he recalled. “We had one guy who just put his feet up on his desk and said, ‘Give me $10 million and get the f___ out of my office.’ There’s a lot of interesting people in this business, to say the least.”

After spending two months visiting MMO outfits, Hansen slashed the list of potential partners from fifty-one to eight. He then assembled a five-person team of LEGO managers—including executive vice president Lisbeth Valther Pallesen, as well as managers from purchasing and marketing—and paid a return visit to the eight semifinalists. Although much of the visit to NetDevil was dedicated to digging into the company’s culture, financials, and technology, Scott Brown kicked off a discussion of Universe by running a trailer for Gears of War, which was initially released as an exclusive title for Microsoft’s Xbox 360.

Developed by Epic Games, Gears was a “third-person shooter,” a 3-D video game in which a player’s on-screen avatar blasts away at other avatars, with much mayhem and bloodletting. The game focused on a squad of hulking, heavily armed, armored warriors who battle to save the last human holdouts on a fictional planet from a swarming subterranean enemy called the Locust Horde—not an obvious model for a LEGO game that would feature kid-friendly minifigs. But the trailer’s ability to quickly establish characters of consequence gave the LEGO team a feel for how heroic personalities could be a compelling draw. With a running time of barely a minute, the trailer delivered an oblique glimmer of what Universe might become.

After the trip, LEGO narrowed the list of potential partners to NetDevil plus an MMO developer in Seattle and another in Norway. NetDevil emerged as the winner, but one event foreshadowed a little problem that would eventually loom large. LEGO sent out a request for proposal (RFP) to the three finalists. The highly detailed document, which totaled hundreds of pages, focused mostly on policy and procedure. There were very few questions about the actual game, which surprised the NetDevil founders. They could conclude only that LEGO hadn’t developed a deep sense of the kind of experience it wanted to deliver. LEGO, it would seem, was seeking to disrupt an industry that it hadn’t figured out.

Second Warning: Don’t Demand That a Young Technology Deliver a Near-Perfect Experience

Hansen believed that NetDevil couldn’t build a great LEGO MMO without deeply delving into LEGO itself. So in early 2006, he took a dozen NetDevil staffers to Billund for a week. They met with developers and marketers, visited the design studios and the LEGO factories, and immersed themselves in all things LEGO.

“It was the most inspirational week I’ve ever had on a project,” said Ryan Seabury. “People assumed that LEGO was into heavily childish designs with these rainbow-colored bricks. But when you see the level of sophistication that can be achieved with LEGO bricks, it absolutely just sparks your imagination.”

Accurately rendering a virtual version of the LEGO brick, however, proved to be a formidable undertaking. LEGO insisted that Universe deliver the cleanest, most detailed bricks of any video game on the market. It took a team of eight engineers working flat out for six months to produce a brick that met the LEGO design specs. The result was that a single Universe brick, with its lovingly rendered tubes and studs—each imprinted with the LEGO logo—was more technologically complex than a fully decked-out World of Warcraft character.

Although they had created an authentic, virtual version of the brick, NetDevil’s developers discovered that it took just a few bricks to exhaust a computer’s ability to display them. In a typical Universe scenario, such as when a minifig walks toward a car, a player’s computer would drain all of its available horsepower by drawing and redrawing the vehicle as the minifig moved closer and closer to it.

Only after running down many dead ends did the developers find a way to work around the problem. They created a system that made the car look hazy and blocky in the distance; because bricks viewed from afar had less detail, the computer could conserve more of its memory when rendering them. The bricks grew sharper only as more of the car came into view. It was a clever solution, but the demand for greater detailing ultimately reduced the number of illustrations—cars, minifigs, spaceships, and more—that the game could display at any one time. The LEGO Group’s quest to deliver a more beautiful virtual brick had the unintended effect of making the game less visually interesting.

“That was one of those things where we just never agreed,” said Brown. “But what could we do? It was their IP, and LEGO would never give in on the quality of the brick rendering.”

The push for perfection breached the logic of a disruptive innovation. At first it’s fine if the disruptive product is low-end and low-quality, so long as it’s easy to use and works well enough to attract less-demanding customers who don’t care if, say, a LEGO logo is etched into each stud of every brick. It’s better to put a cheaper price tag on the product, launch it even though it’s less than ready for prime time, and then make improvements (and charge more) as it moves upmarket. LEGO, however, spent years working to get Universe just right, which cost it the disruptor’s advantage.

The LEGO Group’s insistence that Universe deliver an ultra-premium play experience even showed up in the RFP we mentioned earlier. A LEGO lawyer inserted a provision that Universe, upon its release, must be entirely free of bugs. It was a preposterous demand, especially for a game that would require somewhere between half a million and a million lines of code. The request signaled that the LEGO Group’s innermost value, “only the best,” would bump up against the disruptor’s “good enough” ethos, perhaps best expressed by a maxim credited to Facebook cofounder Mark Zuckerberg: “Better done than perfect.” In the months and years to come, LEGO and NetDevil would ceaselessly struggle to smack back bugs and achieve a rough equilibrium between less than perfect and better than merely done.

The disconnect between NetDevil and LEGO over Universe’s quality loomed larger as development got under way. At any one time, there were thirty or more programmers working on Universe, all changing code in different parts of the game. As Stephen Calender, a former Universe developer, put it in an interview with the blog MMO Fallout, producing a flawless game with nearly one million lines of code “would be like getting a perfect score on a math test with that many questions.”²²

Five years later, when Universe was finally launched, the development team had slashed the game’s most bothersome bugs to well below the industry standard. Nevertheless, seven months after the game’s release, Ronny Scherer, a LEGO veteran and Universe’s project manager, still wasn’t satisfied. “My house here in Colorado has all these nice features, but its questionable quality means it’s going to need a big overhaul in ten years,” he told us. “NetDevil was making software the same way. And fundamentally, LEGO doesn’t think like that. We want to make sure that when we build something, we build it to last.”

Third Warning: Insulate the Project from the Demands of Other Business Units

Had Universe been a physical LEGO product, the game would have been “locked down” as soon as the development team defined the core concept and play experience. That is, no major design changes would have been permitted, since the team would have begun the process of readying the product for manufacture. But because LEGO had a remarkably underfocused view of what the game should be, Universe remained unlocked for months, which eventually grew into years. As a result, it was subject to the influence of many product lines throughout LEGO, each of which had a stake in the MMO.

As originally envisioned, some of Universe’s “worlds” would be designed around classic LEGO play themes. For example, one world would feature knights and castles, another ninjas, while a third would have traditional LEGO City buildings. Getting those play experiences to work in the MMO meant that the Universe development team had to coordinate with the leaders of those product groups back in Billund and ultimately get their sign-off.

Every Billund product group scrambled to ensure that Universe wouldn’t dilute its sales. The result: managers from other business units pulled the Universe development team in different directions. To reduce that risk, LEGO should have separated Universe from the main business and essentially run the Universe development process in parallel with the LEGO Development Process. Big companies that attempt to launch a disruptive innovation are better off setting up something akin to a spin-off unit far away from headquarters, where associates can do what is best for the product, even if it means breaking some company rules. But that never happened with Universe. Even though Colorado-based NetDevil was an ocean plus half a continent away from Billund, it was never permitted to act independently from the mother ship.

“We had to work with everybody, and everybody wanted a say,” said Brown. “The game kept changing. And because the game was a little bit amorphous as to where it was going, we’d get even more requests.” As the requests and changes accumulated, the Universe team found itself drifting from a course that it had never really set in the first place.

Setbacks almost always invite greater scrutiny; LEGO was no exception to that rule of thumb. Hansen soon found himself making repeated trips to Billund to report in on Universe’s delays and to continue championing the project to an increasingly skeptical management. Over Universe’s lifetime, he estimated he made more than a thousand presentations to senior management and other constituencies throughout LEGO. “This project,” he later said, “just Power-Pointed you to death.”

A big-company mind-set likewise thwarted Hansen’s attempt to open up the Universe innovation process, as LEGO had done with Mindstorms NXT. In 2007, Hansen recruited a small group of dedicated AFOLs to contribute to Universe. He even riffed off the Mindstorms MUPs moniker to conceive an appellation for Universe’s volunteer developers, who came to be called the LEGO Universe Partners, or LUPs. In the months that followed, Hansen concluded that Universe’s scale required far more volunteers, and so the LUP team grew to include nearly one hundred members—a crowd, not a clique, that proved devilishly difficult to manage.

As Universe’s development dragged on for months past its deadlines, many LUPs grew discouraged and nearly half dropped out of the program. When it was finally launched in October 2010, Universe included at least three worlds that were created entirely by LUP teams. Even so, the LUPs’ contribution was considerably less than originally envisioned.

“When a project takes five years to complete, it’s very hard to keep people motivated,” concluded Hansen. “We probably didn’t get as much out of it as we would have liked.”

Fourth Warning: Make the Product for Customers, Not Managers

Despite Hansen’s frequent visits to the LEGO Group’s headquarters, he and the NetDevil team still found it difficult to bridge the five thousand miles and eight time zones separating Colorado from Billund. A problem that normally would require a two-minute telephone conversation to resolve often dragged on for two or three days, simply because, say, a NetDevil developer couldn’t quickly connect with the right IT staffer in Billund. As the number of missed connections multiplied, little setbacks grew into big delays.

Equally problematic, each partner had made assumptions about the other that ultimately failed to hold up. “We thought Mark had this team of [LEGO] experts who knew everything about online gaming,” said Brown. “As it turned out, they didn’t know much at all.” More months were burned through as each side struggled to get a clearer line of sight into the other’s shortcomings and shore them up. Increasingly, more and more deliverables were delivered late or not at all. By the spring of 2008, LEGO had deep doubts that the “build” and “socialize” pillars of the Universe experience were articulated enough for the game to hit its 2009 launch date.

In an attempt to bridge the communications chasm between the NetDevil studio and Billund, Lisbeth Valther Pallesen and a handful of other LEGO managers flew to Colorado every six weeks to review the project’s progress. The check-ins garnered decidedly mixed reviews.

Hansen believed the meetings were invaluable for making timely decisions and ensuring that the project kept its momentum. Others took a different view. NetDevil’s Chris Sherland, who, as lead producer, worked to keep the project focused and moving toward its deadlines, believed there were times when LEGO and NetDevil never achieved sufficient trust and confidence to truly expose the project’s problems and make hard decisions.

“There was a lot of, ‘LEGO’s coming, let’s put lipstick on this pig and gussy it up as best we can,’ ” Sherland recalled. “Part of the problem was born out of the culture of third-party game developers, where we’ve got to impress the publisher so we can keep the contract. It was never that blatant, but there was certainly that undercurrent. The founders of NetDevil had to mortgage their entire lives to make this company work. Those check-ins were very important for them. And so we did that dog-and-pony thing.”

The result was that even though the Universe development team did extensive focus testing with kids, the review sessions with Pallesen and other executives made the play experience less appealing for gamers. During an early stage of Universe’s development, the NetDevil team brought in a group of LEGO managers for a demo, the purpose of which was to show off the game engine and the multiplayer environment. They watched as a handful of kids clicked into a network through separate computers and entered a game where they could hit a LEGO monkey with a hammer. With each whack, the monkey would run off and then return for another blow. The kids loved it. The LEGO execs were horrified. One stood up and essentially asked if the NetDevil team had lost their minds: “Do you really think we’re going to go out publicly and have kids smacking monkeys?”

Billund’s stipulation that Universe must be free of any hint of violence was deflating for NetDevil, which, as the creator of unabashedly brutal titles such as Auto Assault and Warmonger, knew well that violence is a turn-on. But once NetDevil began focus-testing some Universe demos, kids delivered an unequivocal verdict: the game was too babyish. The level of conflict just had to be amped up.

There then ensued months of debate between LEGO and NetDevil over just how scary or dark the game should be. The problem was somewhat resolved when TT Games came out with the LEGO Star Wars II: The Original Trilogy single-player video, which proved that chunky minifig versions of Luke Skywalker and Princess Leia battling Stormtroopers could be humorous in an off-kilter kind of way. More important, parents were okay with the video’s blockheaded form of conflict, so long as there wasn’t any bloodletting.

For all that, Universe’s dark-versus-light problem never completely dissipated. In an interview with PC Gamer, former Universe concept artist Mike Rayhawk recalled that he often had to “tone down” images that were deemed too frightening. “I was always pushing to make the look of the game more epic and scary, because I wanted to give real weight to the struggles of these boxy rainbow-colored plastic figures,” he said. “In general, I was more successful selling [management] on the epic than on the scary.”²³

When NetDevil would demo a new Universe feature for a review session, one of the most persistent criticisms from the Billund side of the table was that the feature “doesn’t feel like LEGO.” Such feedback was less than helpful, largely because no one could vividly define the LEGO play experience in an MMO context. It was agreed that the game should stand on three pillars: build, play, and socialize. But translating those themes into bits and bytes was a trial-and-error undertaking. “We found out pretty quickly that LEGO didn’t understand what LEGO Universe should be,” said Sherland. “They could only tell us whether we were or weren’t hitting the mark.”

Each time the steering committee rejected a feature that wasn’t “LEGO enough,” NetDevil had to dump at least some of the underlying code and start over. The team would then spend months rebuilding entire maps—the game’s locales, stages, and missions—in hopes that this time the new feature or play zone would pass muster. And with every do-over, the project would veer just a little more off course. “We’d go off in a new direction, and then everyone would realize that was a mistake,” said Brown. “Then we’d go backward again.”

It took years rather than months to get the game to a place where LEGO managers deemed Universe to be authentically LEGO. Even more troubling than the delays was the fact that LEGO was pursuing the wrong customers. Transistor radios were disruptive because they appealed to a new customer segment—teenagers—who lacked the money and the desire to buy big, luxe radios. Universe might well have been a truly disruptive innovation—the transistor radio of its time—had Billund allowed the toy to become dark enough and aggressive enough to attract kids who could care less about brightly colored plastic blocks but who just might like the prospect of building and competing in an exciting, somewhat sinister LEGO world. Instead, LEGO targeted kids who loved plastic blocks—and never gave them a compelling enough reason to go digital.

Final Warning: Price the Product to Meet What the Market Will Bear, Not to Recoup Your Investment

LEGO Universe finally launched in October 2010, to decidedly mixed reviews. Although many commentators were thrilled to see the LEGO brick come to life in an MMO environment, they complained that the game had an unfinished feel. Macworld pretty much summed up the consensus view when it opined, “It doesn’t seem that Universe has really arrived yet.”²⁴

A big part of a disruptive product’s appeal is that it’s low-end and therefore cheaper than the industry standard. LEGO, however, never gave Universe a disruptive price. Whereas most MMO publishers attract consumers by letting them play a few levels for free—the idea being that once they get hooked on the game, people will pay for more challenging levels—LEGO Universe had a forbidding entry process. Before children could try the game, their parents first had to order a DVD for $40, install it, and then sign up for a $10 monthly subscription. For most parents, that was simply too high a barrier. It took nearly a year for LEGO management to realize its mistake and restructure the game by giving kids a no-cost entry to a couple of Universe’s worlds.

When we visited the NetDevil studio in Colorado a few months before it closed, the LEGO Universe team tried to summon some cautious optimism, though the underlying mood was as dark as the afternoon thunderheads that swept over the Front Range’s snowcapped peaks. On one hand, Universe had achieved a resubscription rate of 87 percent, which far exceeded the industry standard of below 20 percent. But it had registered just thirty-eight thousand subscribers, not nearly enough to sustain the business.

Unfortunately for its staff of more than a hundred, plus the tens of thousands of kids who loved the game, LEGO Universe never came close to meeting the company’s sales targets. Just fifteen months after its launch, LEGO quietly pulled the plug. On January 30, 2012, the company shut down Universe’s servers. With that, Universe became the one major product failure during Knudstorp’s seven-year effort to turn around and transform LEGO.

LEGO Universe’s collapse shows how difficult it is for any big company to change its product development system so as to deliver a disruptive innovation. As it turned out, the strategies that worked for the LEGO Group’s plastic brick business hindered its efforts to build a disruptive digital brick business, as Christensen himself might have predicted. Rather than let NetDevil act like the start-up that it really was and ignore what counted as prudent in Billund, the Danish company’s senior management pushed the Colorado coders to develop Universe the LEGO way.

LEGO, largely an analog company, failed to adapt to a digital world where change is unrelenting and noncompliant. LEGO jumped into the MMO business before fully understanding it. Because executives took many months to fully define Universe’s play experience, the game’s design specs kept changing, resulting in further delays and cost overruns. Then, instead of tailoring the game to what the underlying technology could actually deliver, LEGO executives demanded that Universe create an “only the best” play experience. When the satellite studio failed to do so, the number of miscommunications and misunderstandings between Billund and Colorado spiked upward. And when LEGO finally launched Universe, it didn’t price the product to meet what the market would bear, but instead sought to quickly recoup its investment. No doubt LEGO will keep chipping away at the big, important challenge of creating riveting digital play experiences. But after that fifteen-month run, the company’s patience for Universe ran out.

A video on the LEGO Universe website announced the closing by proclaiming that the game had “shot for the stars and reached the planets.” Not a bad epitaph, given that Universe extended the fun of the physical brick into the virtual realm but never became a sustainable business.

An Innovation That Might Well Disrupt LEGO

In May 2009, a little more than a year before LEGO released Universe, a Swedish video game programmer and designer named Markus “Notch” Persson posted the first crude version of a game he’d created. Like Universe, the game let players freely roam a virtual, 3-D world and build complex creations such as spaceships and volcanoes. Like LEGO, it featured textured building blocks reminiscent of the brick, though Persson’s were not nearly as complex. It also had a survival challenge that proved to be a magnet for kids and adults. Each game began at dawn, with nightfall coming eighteen minutes later. Players had to quickly build a shelter to protect themselves, because at night monsters came out—zombies, skeletons, and (later) green “creepers.” Persson called his game Minecraft.

Based on feedback from users, Persson steadily expanded Minecraft, adding new building materials, new monsters, a multiplayer capability, and the capacity for users to create “mods,” their own DIY versions of the game. He started charging for the game soon after releasing it, and steadily increased the price as the game’s features expanded and its popularity exploded. As it worked its way upmarket, Minecraft began to take on all the hallmarks of a disruptive innovation.

When it launched, Minecraft was decidedly a low-end product. Its crude, blocky look and feel had none of the LEGO brick’s precision and Universe’s attempt at perfection. Early on, Minecraft’s “building experience,” which involved hitting things with an axe, offered none of the brick’s limitless capacity for creative play. And the characters—zombies and creepers—were barely recognizable. But Minecraft was inexpensive. An unlimited, lifetime license for Minecraft cost just $13, which was far cheaper than a typical themed LEGO set consisting of plastic bricks and just a little more than the price tag for a month’s play on LEGO Universe. And the gameplay, where the primary goal was to escape attacks by monsters, proved addictive. Minecraft spread like World of Warcraft through the global gaming community.

In January 2011, Minecraft passed one million purchases. By April 2011, Persson estimated that Minecraft had made $33 million in revenue. By August 2012, Minecraft claimed more than thirty-six million users, nearly seven million of whom had purchased the game. That translated into a sales rate of about $100 million per year for the game’s PC-based version. At the time, Minecraft was the fastest-selling game on Microsoft Xbox, and it featured an extensive catalog of merchandise, including a partner product with LEGO.

With the explosion of 3-D printing tools in 2012, Minecraft moved further into LEGO territory. The MakerBot Replicator, just one of more than a dozen new 3-D printers introduced in 2012, can make objects out of ABS plastic—the same plastic that LEGO bricks are made of—in any shape, up to about the size of a small loaf of bread. At the time of this writing, 3-D printing is roughly equivalent to the digital photography of the late 1990s. Back then, the price of a digital camera was comparatively high and the picture quality was inferior to film-based photography. But the benefits were clear—you could take as many pictures as you wanted for almost nothing, and print out only the images you liked. Similarly, 3-D printers are expensive (in early 2013, the MakerBot Replicator cost $1,749) and the quality of the product is decidedly on the low end. But the cost to “print” a single 3-D copy of a LEGO model is 5 to 10 percent of what it would cost to buy that model in a store.

With a program called Mineways, developed by a Minecraft fan named Eric Haines, Minecraft users can “print” their creations in plastic, stone, ceramic, silver, or gold-plated steel. The program starts with a Minecraft creation and automatically produces a file that can be sent directly to a 3-D printing company such as Shapeways or to a personal 3-D printer such as the MakerBot Replicator.

To illustrate the Mineways/Minecraft threat to LEGO, Gordon Robertson (the author’s son) built the LEGO Fallingwater kit (part of the Architecture series we described in the previous chapter). He then used it as a model to create a version of the famous Frank Lloyd Wright house in Minecraft, a four-hour project on a rainy weekend afternoon. (Insert photos 17 and 18 show the LEGO Architecture kit and Gordon Robertson’s Minecraft version.) Using the Mineways tool, he then “printed” a physical reproduction of the virtual Minecraft version of Fallingwater, both using Shapeways (see insert photo 19) and the author’s MakerBot Replicator (see insert photo 20). The cost of the MakerBot model is about one-tenth of the LEGO kit, which retails for about $100.^†

So now, using a combination of Minecraft, Mineways, and a MakerBot Replicator, kids can create any model they want using as many blocks as they wish, and print out a half dozen copies for less than the cost of a similar LEGO kit. Just as with digital photography, the price of the machines is dropping fast and the quality is improving. Although the machines are designed for sophisticated hobbyists, it won’t be long before anyone will be able to build, print, and rebuild any block-based plastic model they wish, for a fraction of the cost of a LEGO kit.

To be sure, the LEGO Architecture Fallingwater set offers a beautiful re-creation of Wright’s iconic house. But Adam Reed Tucker’s kit includes just the pieces and colors he believes will create the best reconstruction. If you want to add Wright’s guest house (which sits above the main house) or create your own modifications to Fallingwater, it’s difficult and expensive to find the right LEGO bricks. With Minecraft, Mineways, and the MakerBot Replicator, you can build many different versions of the house and print the best four or five, all for less than the cost of a single LEGO kit. It’s easier and faster to redo a room in the Minecraft version of the house than it is with the LEGO version. With LEGO, you’d have to disassemble the model, find the bricks you need, rebuild the room, and then reassemble the floors above it. With Minecraft, you simply “walk” into the virtual house, make your changes, and print out the new version.

Minecraft is a long way from posing a serious threat to LEGO. It lacks the sweeping variety of LEGO play offerings. It hasn’t come close to matching the LEGO brand’s global appeal. And it can never duplicate the essential thing that’s drawn millions to LEGO—the tactile pleasure that comes from making two bricks click. But LEGO executives have long known they had to disrupt their plastic brick business with a digital brick business, lest a competitor do it for them. Minecraft, however, got to the digital future first. In 2012, the year LEGO shuttered Universe, Mojang—Minecraft’s parent company—earned $90 million on sales of $235 million. As 3-D printing improves, LEGO might one day find that Minecraft (or something like it) has jumped the digital divide to become a more compelling, richer, and easier way to construct plastic buildings, characters, and vehicles. As many disrupted companies have learned, what was low-end yesterday can be mainstream tomorrow.

One of Clayton Christensen’s most counterintuitive arguments is that managers sometimes fail because they are smart, not because they are stupid. They abide by the same management strategies that have worked so successfully in the past. As a result, the disruptive effort goes wrong. Such was the case with the LEGO Group’s handling of Universe.

The company’s management concluded that digital brick-based play experiences might one day overthrow physical brick-based play experiences. So it raced to get to the future first with Universe. In doing so, Lisbeth Valther Pallesen and her managers reverted to the very same practices that had helped revive LEGO itself. They insisted that the Universe team stand in the customer’s shoes, through rounds and rounds of focus testing, to truly understand what core LEGO kids wanted out of the game. They built a full spectrum of LEGO experiences into the game—including a new pricing scheme and physical models of virtual creations—which meant coordinating with many business units in Billund. And they opened up the development process to expert AFOLs, just as LEGO had done with Mindstorms NXT. Taken together, those innovation strategies, which had helped LEGO succeed so well in its traditional markets, burdened the company’s effort to thrive in unexplored territory.

With a novel product such as an MMO game, at first it’s uncertain what the technology can and can’t deliver and what market will ultimately want it. LEGO might well have been wise to let its Universe team act like an independent start-up—in other words, like a Minecraft. Freed from excessive executive attention and the demands of other LEGO product groups, the team might have launched a far more modest, less costly version of Universe. Compared to other LEGO products, Universe would have been low-end, a little shoddy, and in almost every way inferior—all the hallmarks of a truly disruptive product. Some kids would have been drawn to the novelty of a LEGO MMO, which would have allowed the Universe team to see what worked and what didn’t, and thereby refine the play experience even as it “lived” in the market. As Universe improved, it likely would have attracted more customers and moved upmarket, just as Minecraft has done. Universe might even have become the transistor radio of its time, a product that ascended from the bottom of its market and eventually became ubiquitous.

What’s certain is that Universe’s failure and Minecraft’s success demonstrate that to create a low-end disruptor, it’s far more effective to put a tight lid on resources, free the development group from outside distractions, kick the product into the market before it’s completely “finished,” learn from customers’ feedback, and make improvements in real time. What’s surprising is that LEGO, as we’ll see in the next chapter, had already set itself up for exactly that approach to innovation.

* Through a worldwide licensing agreement, the British video game developer TT Games—founded by former LEGO staffers who had been let go in the 2003 downsizing—created popular console games such as LEGO Star Wars and LEGO Batman.

† The cost of the LEGO Fallingwater kit is about 15¢ per gram, which is three times the cost per gram of the plastic used in the MakerBot. But with the MakerBot you can hollow out the inside of a model, reducing the amount of plastic used to make the model.