Jake Nickell stepped to the front of a small classroom on the MIT campus in Cambridge, Massachusetts, and looked around. It was an autumn morning in 2005, and before him sat a dozen executives from some of the country's largest companies -- General Mills, Pitney Bowes, Clorox, and Google -- and a contingent of innovation researchers from MIT's Sloan School of Management and other business schools. The meeting had been organized by Eric von Hippel, an MIT bigwig and the foremost authority on something called user innovation. Von Hippel had heard about Nickell from a graduate student and had invited him to Cambridge to share his story with the group.

Nickell was somewhat befuddled by all the attention. He was not familiar with the term user innovation -- or, for that matter, the term Eric von Hippel. Business at Nickell's company, Threadless, had been growing quickly -- annual sales were on track to hit $5 million, and he had lately started getting curious calls from venture capitalists and large retailers. But Threadless didn't quite seem like MIT material. At 25, Nickell hadn't even graduated from college.

Von Hippel, a Harvard graduate, entrepreneur, and former McKinsey consultant who was 40 years Nickell's senior, called the room to attention and began lavishing praise on Threadless; he called the company a "perfect example" of a new way of thinking about innovation. Von Hippel's theory, which he had introduced in the late 1970s, was that most product innovations do not come out of corporate research and development labs but from the people who use the products. Nickell shot a confused glance at Jeffrey Kalmikoff, Threadless's chief creative officer, and Jacob DeHart, his chief technology officer. The meeting had barely begun, and they had already learned something.

Nickell started talking about his company. Threadless, he explained, ran design competitions on an online social network. Members of the network submitted their ideas for T-shirts -- hundreds each week -- and then voted on which ones they liked best. Hundreds of thousands of people were using the site as a kind of community center, where they blogged, chatted about designs, socialized with their fellow enthusiasts -- and bought a ton of shirts at $15 each. Revenue was growing 500 percent a year, despite the fact that the company had never advertised, employed no professional designers, used no modeling agency or fashion photographers, had no sales force, and enjoyed no retail distribution. As result, costs were low, margins were above 30 percent, and -- because community members told them precisely which shirts to make -- every product eventually sold out. Nickell's company had never produced a flop.

The audience members listened, rapt. For years they had suspected that this kind of business model was possible -- even inevitable. They had seen the beginnings of it in the open-source-software movement, and they had been trying to make it happen in small ways within their own companies. But somehow, this T-shirt guy had gone whole hog. He had built an entire business around the idea that an online community could drive innovation. "We were blown away," says von Hippel.

Nickell does not seem like the kind of guy whose presence would blow anyone away. He's 5 feet 10 inches but seems smaller, with a frame that barely fills out the medium-size T-shirts he wears like some kind of uniform. His fine features, unkempt hair, and scraggly red beard make him look almost elfin. His mannerisms, including a whisper-quiet voice and a tendency to punctuate his sentences with a faint giggle, do little to dispel this impression.

But Nickell is at the vanguard of a new innovation model that is quietly reshaping a host of industries. Whether it's called user innovation, crowdsourcing, or open source, it means drastically rethinking your relationship with your customers. "Threadless completely blurs that line of who is a producer and who is a consumer," says Karim Lakhani, a professor at the Harvard Business School. "The customers end up playing a critical role across all its operations: idea generation, marketing, sales forecasting. All that has been distributed."

This idea goes against a basic principle that has been taught in business schools since the invention of mass production: Employees make stuff, and customers buy it. But this notion seems anachronistic in a marketplace of ever-narrowing niches and nearly unlimited consumer choices. Meanwhile, a generation of so-called Web 2.0 companies has succeeded by encouraging customers to contribute to, and in some cases create, the product being sold. Not only do we have instantaneous access to countless television programs though video websites, but anyone with a YouTube account and a digital camera can create a show of his or her own. Professionally edited, dead-tree newspapers are besieged by digital news sites that are produced and edited by their readers. The 240-year-old Encyclopaedia Britannica finds itself eclipsed -- at least in terms of readership -- by Wikipedia.com, which pays its writers nothing and requires that they possess no expertise at all.

In a superficial way, Threadless resembles these Web 2.0 firms. It is an online business, built around a social network, in which users collaborate with one another. The difference is that Threadless is not a software or media company. It designs, manufactures, and sells actual stuff. "They're the beginning of a new wave," says von Hippel, the author, most recently, of Democratizing Innovation. Von Hippel envisions a future in which most companies essentially abandon market research and product design and instead rely on communities of users to figure out which products to sell. Tim O'Reilly, the founder of O'Reilly Media and the guy who coined the term Web 2.0, goes even further. "As manufacturing technology gets richer, this model will be much more widely applied," he says. Already, the prices of rapid prototyping gadgets like three-dimensional printers and laser cutters are plummeting, a fact that O'Reilly believes will allow for open-source electronics, furniture, and toys. "How far off," he asked in a 2006 essay posted to his blog, "is a future in which the creative economy overflows the thin boundary that separates 'information' from 'stuff'?"

Nickell had no such vision as he put the finishing touches on a T-shirt design in late 2000. It was for the New Media Underground festival, an informal gathering of Web designers in London. He had no intention of attending the event, but he cared about it deeply. At the time, Nickell was 20 years old, living in a tiny Chicago apartment. He spent his days on the sales floor at CompUSA; at night, he was a talented if unenthusiastic part-time student at the Illinois Institute of Art. Though his girlfriend visited him each weekend, he had few close friends.

When he wasn't working or studying, Nickell was tinkering with Web design, a hobby he indulged in on Dreamless.org, an Internet forum for illustrators and programmers. He would spend hours at a time cruising the forum, talking with his online friends and engaging in a pastime called Photoshop tennis. In it, designers pass digital photographs back and forth and challenge one another to manipulate the images in the most outrageous way possible.

Nickell's design for the New Media Underground festival -- three lines of gray text that mimicked the layout of the Dreamless website -- was an entry for a contest that the festival's organizers were holding online. The design was simple and not quite pretty. But it was strikingly clever -- a physical representation of their digital community. The Dreamless members agreed. Nickell won the contest.

In concrete terms, this accomplishment meant exactly nothing: He got no money or even a copy of his winning shirt. But the experience was exhilarating. Dreamless members spent a lot of time batting ideas back and forth, but their creations rarely made it out of the digital realm. Suddenly, Nickell had an idea: What if the best designs were printed on T-shirts and sold in the real world? He suggested as much to Jacob DeHart, one of a handful of friends he had met on Dreamless. DeHart, a student at Purdue University, loved the idea, and each pitched in $500 -- enough to pay a lawyer to set up the business and print the first round of shirts.

Nickell and DeHart held their first contest in November 2000. They asked the designers on Dreamless to submit their best work and to pick their favorites. The grand prize: two free shirts and the promise that any proceeds would be reinvested in future contests. They called the competition Threadless, a play on thread -- either a clothing item or a discussion topic on an online forum. In all, they printed two dozen copies of five shirts out of slightly fewer than 100 submissions with in-joke titles like "Evil Mother F---ing Web Design" and "Dead Sexy Designer." The shirts went on sale in January 2001 for $12 each and sold out quickly. In the months that followed, Nickell and DeHart ran regular competitions using an automated rating system that allowed users to score designs on a scale from 1 to 5, but it never occurred to them that they had a real company. "It was just a hobby, a way for people to get their artwork out," Nickell says. By 2002, the hobby had surpassed $100,000 worth of T-shirts and attracted more than 10,000 community members, mostly artists in their teens and 20s. Even so, Nickell, DeHart, and Kalmikoff -- who joined the company that year -- spent much of their time doing freelance Web design to pay the bills.

Shortly after founding the company, Nickell and DeHart began awarding small cash prizes to the artists whose T-shirts were selected. Initially the prizes were $100 per winning design, but they gradually climbed to $2,500, plus reprint fees. But the appeal of Threadless to artists has never had much to do with getting paid. "It wasn't so much the money," says artist Glenn Jones, who won $150 in a contest in 2004, at age 29. "It was how cool it was to get your shirts printed." Young illustrators had few outlets in which to display their art, and within a few years of the launch, Threadless had acquired a sort of American Idol cachet. It was where unknown designers went to make their names.

In addition to attracting a lot of talent, the contest format encouraged artists to tell their less artistic friends about the site. Designers labor mightily on their submissions; they spend weeks tinkering with their work and soliciting advice from other members. Then they post links to their submissions on their websites, blogs, and MySpace pages, asking their friends to click, vote, and, the artists hope, buy. (Threadless helps with this, sending the artists digital submission kits that include HTML code and graphics to help them create professional-looking advertisements for their designs.) "Threadless was a huge word-of-mouth thing," says Tom Burns, a 30-year-old freelance designer in Murfreesboro, Tennessee, who discovered Threadless through a mention on a design forum in late 2004. When he submitted his first design, he told all his friends and posted links on every design forum he frequented. "I spread the word as much as I could," says Burns, who eventually won for a design called Spaghetti Western.

Threadless users are not required to join the social network or vote in order to buy shirts, but many users have offered their opinions on thousands of designs. There's something enjoyable and empowering about playing critic in a never-ending gallery of pop art. "Participation on Threadless is not just about voting for designs you really want to buy," says Frank Piller, a management professor at Germany's Aachen University and a researcher at MIT. "It's an exploration of new designs, and it's fun." For a 2006 paper he published in the Sloan Management Review, Piller surveyed Threadless customers and found that only 5 percent were buying shirts without first voting on designs. "Almost no one was simply consuming," he says. "They were all participating."

This rabid engagement propelled the company through four years of phenomenal growth, beginning around 2004. The user base grew tenfold, from 70,000 members at the end of 2004 to more than 700,000 today. Sales in 2006 hit $18 million -- with profits of roughly $6 million. In 2007, growth continued at more than 200 percent, with similar margins. Though Nickell refuses to disclose the exact revenue number -- perhaps because he now counts Insight Venture Partners, a New York venture capital firm, as a minority shareholder -- it seems fair to assume that Threadless sold more than $30 million in T-shirts last year.

Ask Nickell what he makes of his company's whirlwind success, and he will respond rather sheepishly. "I think of it as common sense," he says. "Why wouldn't you want to make the products that people want you to make?" Indeed, the idea that the users of products are often best equipped to innovate is something many entrepreneurs know intuitively. And it is supported by a growing body of research. A study published last year in the Strategic Entrepreneurship Journal suggested that the vast majority of companies are founded by "user-entrepreneurs" -- people who went into business to improve a product they used. Meanwhile, studies by von Hippel and others show that in industries as diverse as scientific instruments and snowboard equipment, more than half the innovations generally come from users, not from research labs.

Some major corporations are beginning to experiment with these ideas. Pitney Bowes is building an online social network for direct marketers who use its mail machines, and Ford now allows drivers of its Focus sedan to add third-party hardware and software to their vehicles' navigation and entertainment systems. But most companies still prefer what von Hippel calls the "find-a-need-and-fill-it" paradigm -- which involves market research, focus groups, testing, reworking, and retesting. Not only is this method extremely costly, but it fails to capitalize on a company's most dedicated customers -- who often are already improving existing products to fit their needs. Think of the hacker who tweaks his iPhone to allow it to run Skype, the mountain biker who builds an improved chain guard, the teenage girl who cuts the collar off her Gap T-shirt.

Some companies actually punish these people by cracking down on unauthorized innovations. Apple has famously "bricked" -- that is, electronically disabled -- iPhones that have been enhanced by their owners. Other companies pay lip service to user innovation but have trouble following through on the concept. "Companies are very good at creating platforms for external input, but they're very bad at using this input," says Piller, who has studied BMW's use of an innovation portal, a website that invites consumers to submit ideas. "BMW gets a thousand good ideas each year," he says. "Maybe they use one every two years." In other words, no matter how much technology goes into prettying up the suggestion box, the suggestions tend to get dumped in the trash at the end of the week.

Threadless is an exception to this. "You could say that what Threadless does is trivial, but it's not," says Harvard's Lakhani. In fact, the very triviality of Threadless's product -- something as low tech and as commoditized as a T-shirt -- proves that vibrant online communities can drive all sorts of nontechnical businesses. This should be encouraging news to entrepreneurs. Customer communities have become exceedingly inexpensive to build and manage; blogging software and social network platforms, for example, are now available for free from a handful of start-ups. "We thought that open source could only work in software, and now it's being successfully applied to a product as mundane as a T-shirt," Lakhani says.

Threadless's headquarters is in a former printing plant on Chicago's Ravenswood Avenue. The 25,000-square-foot office is open to customers, a dozen of whom stop by every day to pick up shirts in person. They sometimes stick around for hours and hang out in a space that resembles a college dorm room constructed on an impossibly large scale. There are video game consoles, go-carts, a giant television, beanbag chairs, action figures, a singing-buck trophy, a Ping-Pong table, and a full-size Airstream trailer that the company uses as a studio in which to produce podcasts.

These diversions and many others are captured in a host of videos published at Threadless's website. There's the one of Nickell smashing a television with a homemade potato gun, the Halloween video with employees throwing pumpkins off the roof, and the one that depicts Threadless employees in a fierce pillow fight. Many videos depict employees dancing, often in ridiculous costumes. In short, the impression is of a teenage paradise -- a company that has nothing but fun.

But take a closer look, and the company is suspiciously companylike. The go-carts generally stay parked, the buck stays mute, and the Ping-Pong table serves as a gathering place for impromptu meetings. "When I started, we spent half the day playing," says Lance Curran, a bearded 29-year-old wearing a beanie, jeans, and a flannel shirt. "That doesn't happen anymore." This is not to say Curran doesn't like his job. On the contrary, he nearly glows when he talks about his rise from a temporary warehouse worker in 2005 to the warehouse manager in charge of a staff of 18 today. When Curran arrived, the average time from an order to shipment was a full month during peak season. Today, it's one day, thanks to a system by which the warehouse is regularly reorganized based on what is selling well. Hot products are placed near the packaging area, which means that packers, who will often walk three miles in the course of a shift, don't have to travel as far to get shirts. "Visitors come in here with degrees, and they can't believe how this place has evolved without any outside help," says Curran.

Like Curran, most of Threadless's employees come with no obvious qualifications for their jobs. The oldest staff member is 33, and many are under 25. The employees do, however, arrive with a deep and abiding love of Threadless, having joined the community long before they entered the work force. Joe Van Wetering, a 21-year-old illustrator who works in the production department, was a frequent visitor to Threadless's offices as a teenager before taking a job in the warehouse in 2006. Ross Zietz had won seven competitions while studying art at Louisiana State University before he took a job as the company's janitor in 2004. He has since been promoted to art director, charged with helping the winning designers get their entries ready for printing. In fact, 75 percent of the company's 50 employees were community members before they were hired.

Nickell makes hiring decisions based primarily on one metric: trust. "It's pretty much the only thing we talk about when we interview," he says. The goal is to find people who can work independently. "It takes a while for people to get adjusted to this place," he says, adding that those who do not display an ability to figure things out on their own are quickly dispatched. (The first and only person over the age of 40 to work at Threadless, the CFO, left after only 60 days on the job in early 2007.)

Trustworthiness is especially important at Threadless because the company's most important asset -- its vast online community -- is managed collectively. Threadless employs no moderators, and no single person or group is charged with keeping the community happy. Nor, technologically speaking, is the social network itself especially advanced. It lacks many of the features found on MySpace or Facebook. There are no virtual friends, no messaging features, and no status messages. Users' profiles are made up of their blog postings and their submissions.

But what Threadless lacks in flashy features, it makes up for in steadfast dedication to staying close to its customers. Both Nickell and Kalmikoff spend much of their time cruising Threadless.com -- posting comments on blogs, inspecting designs, and tweaking the website. They publish their instant-message addresses and regularly query the public about changes to design or contest policies. "If someone changes something on Facebook, there's no expectation that some random 14-year-old from the middle of Idaho is going to be able to get in touch with Mark Zuckerberg," Kalmikoff says. "On Threadless, if people see something they don't like and want to talk to Jake, they get Jake."

Employees have long served as the models for the company's shirts; this puts community members on a first-name basis with them. "Ross makes one hot Dylan," reads a recent comment from one of Zietz's admirers in response to a Bob Dylan-inspired pose he struck. (Users are also invited to upload photos of themselves wearing Threadless shirts. For each photo submitted, the company doles out a credit worth $1.50.) Meanwhile, each employee is encouraged to talk regularly with users. For instance, when Curran is planning on cleaning out the warehouse, he alerts the customers on his Threadless blog. The posts typically generate dozens of requests. "Keep an eye out for the XL Corporate Zombie," reads one from a customer eager to score a lost copy of a sold-out design.

In 2005, Nickell got a call from a buyer at the retailer Urban Outfitters (NASDAQ:URBN) about carrying Threadless's shirts in the company's 150 stores. Around the same time, Target (NYSE:TGT) sent him a several-hundred-page contract for a "test" involving tens of thousands of shirts in a handful of stores and on Target.com. Nickell reluctantly declined both offers, fearing a backlash from a community that often uses "Urban Outfitters" as a synonym for uncool. "We would do a deal with Target or Urban Outfitters," Nickell insists. "The only stipulation we need is to have some kind of presence in the store where people are able to easily learn about where the designs come from. You go to Target or Urban, and it's just shirts on a wall. You have no idea where they came from or who designed them," he says. He would like to see a computer kiosk that allows shoppers to score designs and read about artists, but when he pitched the concept to Urban Outfitters -- which approached Threadless again in 2007 -- the clothing giant demurred. "As long as the story isn't lost, we're OK," he says. Urban Outfitters and Target declined to comment, citing company policy.

Still, Nickell is not averse to pushing his model in new directions. In late 2006, he sold a minority stake to Insight Venture Partners for an undisclosed amount. He informed the community with a release titled "Holy Crap, Big News!" With Insight's cash and expertise, Nickell began work on a Threadless retail store and started looking into the possibility of opening a European warehouse to speed international shipments. He also broke with DeHart, who had lost interest in expanding Threadless and wanted to start something new. (DeHart declined to comment for this story. He maintains his ownership stake in Threadless and a board seat but no longer works at the company.)

Last September, Threadless opened a two-story store in Chicago's Lakeview neighborhood. Nickell spared no expense in designing the space, which is appointed with zinc panels, hardwood accents, and 20 flat-screen television monitors. But the most striking thing about the store is how few products are on sale. The upper floor is dedicated exclusively to art, and 20 or so T-shirt designs are sold on the lower level. This layout makes the space feel worlds away from American Apparel (AMEX:APP) stores, which are usually crammed floor to ceiling with all things cotton. Nickell imagined the store as a marketing channel -- a physical embodiment of Threadless.com that would help attract attention to the website, give artists a chance to see their work sold in a real-life setting, and serve as a venue for events such as concerts and art exhibits. He figured it would lose money.

Six months later, the store is profitable, and Nickell is already planning a children's shop in Chicago and a second store in Boulder, Colorado. Eventually, he hopes to open stores in midsize cities such as Austin, Seattle, and Minneapolis. The expansion, which has been greeted positively by much of the Threadless community, is not without its detractors. "Please promise me that this will be the ONLY store you will be opening," wrote a member several days before last year's grand opening party. "Next thing you know, everyone on every other block will be wearing the same shirt." And LovesThreadless.com, one of several independently run fan sites, greeted retail expansion coolly. "Congratulations, guys, but keep it real," wrote the site's creator, Chris Cardinal, a 22-year-old Phoenix-based Web developer.

Such tension, Nickell says, is inevitable in community-based businesses. "Even before now, we've been losing our core people and gaining new core people," he says. "It's kind of like a band in its infancy: As soon as a lot of people start listening to the band, the core fans go away." Nickell points out that the look of Threadless's T-shirts -- what other clothing companies might call the brand -- has changed drastically as his audience has evolved from a small collection of geeky Web designers to include tens of thousands of teenagers from middle America. Webmaster jokes have been replaced by cultural references; wordplay has given way to painterly richness. The Threadless brand is not the shirts but the community experience. As Nickell puts it, "Our brand is a fun boys' and girls' club."

Now, Nickell is set to let his club loose on other businesses. In addition to expanding to children's clothing and retail, Threadless will begin selling prints and posters online. And later this year, the company will add a range of products, including handbags, wallets, and dinnerware, under the brand Naked & Angry. Each item will be adorned with patterns submitted by users, with a new product launched each month. "I think Naked & Angry, if handled properly, has the potential to be way bigger than Threadless, because we have the flexibility to do everything," says Kalmikoff, who envisions moving into high-end clothing as well as housewares. Jeff Lieberman, managing director of Insight Venture Partners and a board member, is even more bullish. "To say it's just a T-shirt company is absurd," he says. "I look at it as a community company that happens to use T-shirts as a canvas."

That a quirky T-shirt company can elicit such glowing statements from a private equity professional who invests mostly in an industry called software-enabled services is something of an accomplishment in itself. Indeed, nearly everyone who touches Threadless seems to come away feeling a little cooler. The executives and academics who met Nickell and Kalmikoff three years ago at MIT still speak glowingly of the kids from Chicago who bailed out on a group dinner at Legal Seafoods in order to attend a customer meet-up in a Cambridge bar. "They're a bunch of guys in T-shirts, but they're incredibly thoughtful," says Jim Euchner, vice president of growth strategy and innovation at Pitney Bowes. Euchner marvels at the way Threadless has built a defensible business based not on proprietary information or technology but on an extremely loyal group of customers. "It's just a different way to think about business," he says.

The way Eric von Hippel sees it, Threadless has tapped into a fundamental economic shift, a movement away from passive consumerism. One day in the not-too-distant future, he says, citizen inventors using computer design programs and three-dimensional printers will exchange physical prototypes in much the same way Nickell and cohorts played Photoshop tennis.

Eventually, Threadless-like communities could form around industries as diverse as semiconductors, auto parts, and toys. "Threadless is one of the first firms to systematically mine a community for designs, but everything is moving in this direction," says von Hippel. He foresees research labs and product-design divisions at manufacturing companies being outstripped by an "innovation commons" made up of tinkerers, hackers, and other devout customers freely sharing their ideas. The companies that win will be the ones that listen.

This may or may not come to pass, but the lesson of Threadless is more basic. Its success demonstrates what happens when you allow your company to become what your customers want it to be, when you make something as basic and quaint as "trust" a core competency. Threadless succeeds by asking more than any modern retail company has ever asked of its customers -- to design the products, to serve as the sales force, to become the employees. Nickell has pioneered a new kind of innovation. It doesn't require huge research budgets or creative brilliance -- just a willingness to keep looking outward.

Max Chafkin is an Inc. staff writer.