Flag sculpture by Kyle Bean

Custom-made for a supplier of GM auto parts, this basket holds pump housings firmly in place as they are washed; the mesh top aids water flow.

Marlin Steel used to make bagel baskets. Now it's a supplier to high-tech factories.

Manufacturers use these baskets--among Marlin's few stock products--to hold industrial parts during washing, degreasing, and ultrasonic cleaning.

Drew Greenblatt, owner of Marlin Steel Wire, tours and talks with employees on the factory floor of the company's manufacturing facility in Baltimore, MD.

The Road To Resilience: How Unscientific Innovation Saved Marlin Steel

A little maker of metal baskets shows how U.S. manufacturers can thrive against all comers.

The American economy has some really quirky corners--places so esoteric or tucked away we hardly notice them. In 1998, Drew Greenblatt bought one of those corners--a company called Marlin Steel that specialized in a single product: wire bagel baskets, which bagel stores use to display their wares. Marlin had the market to itself. "You had the guy who made baskets for the doughnut stores, Dunkin' Donuts and the like," Greenblatt says. "You had the guy who made the metal chafer stands that buffet serving dishes sit on, with the cans of Sterno. And you had us, doing the bagel baskets." Marlin's customers were the big chains: Einstein Bros., Bruegger's. "Once in a while, we'd edge into each other's business," he says. "The doughnut guy would try to get some bagel stores. But mostly we did our own thing."

Marlin was based in Brooklyn then. Greenblatt, who was 31 when he bought the company, moved it down to Baltimore, near his hometown, along with nine families who wanted to stay with Marlin. It was really just a metalworking shop: 18 employees, most at minimum wage, using hand tools to bend and weld metal, with $800,000 a year in sales. Marlin didn't own a fax machine, and most of the equipment was from the 1950s. Purchase orders arrived by mail. The pace was methodical and unhurried--each employee made 15 or 20 baskets a day. Still, says Greenblatt, make no mistake: "We were the King of the Bagel Baskets."

But not for long.

Within five years of buying Marlin, Greenblatt was getting killed. Chinese factories suddenly started making bagel baskets. Marlin sold its baskets for $12 apiece and with 36 baskets to equip a typical bagel shop made $450 when a company added a location. Chinese factories were selling baskets for $6 each. Marlin's customers were switching to save $200 a store. And Marlin would never be able to match its Chinese competitors on price. "My steel was costing me $7 a basket," says Greenblatt. "We were going to go extinct." It would have been smarter for him to buy bagel baskets from China for $6 each and sell them for $6.50.

Marlin's crisis came at an interesting moment in the history of making stuff. The bad news about U.S. manufacturing has been stark and recited relentlessly. The United States has 6 million fewer factory jobs today than in 1998, a decline of one-third in 15 years to just 12 million, the fewest since May 1941. Yet U.S. manufacturing is not dead or even moribund. Indeed, part of the job loss springs from growth in productivity. U.S. factories are relentless efficiency machines. The typical American factory worker today produces one-third more than in 2000, which is why manufacturers have been able to increase output while cutting employees. The United States still makes more stuff, by dollar value, than any nation in the world except China, which moved into the top spot only in 2010.

And the United States still has hundreds of thousands of factories. The ones we notice are big--GE, Toyota, Whirlpool--but most are small, like Marlin Steel. The average U.S. factory has just 40 employees. Many such factories get trampled on price alone and disappear without notice, taking a steady trickle of jobs with them.

Marlin saved itself by facing a truth that few threatened manufacturers can stomach: It was failing because it had gotten everything wrong. It had the wrong customers; it had the wrong products; it had the wrong prices. Greenblatt realized--just in time--that even wire baskets could be innovative. The simplicity of Marlin's technology is not what we typically associate with innovation--there's no algorithm, no microchip, no touch screen. Instead, Marlin learned how its products could help its customers, providing the quiet innovation that can give a fellow U.S. factory a critical edge and help keep jobs in the United States.

Today, a decade later, Marlin Steel is outcompeting not just Chinese factories but German ones as well. Its sales are six times the 2003 level, and it has almost double the number of employees. The staffers have health insurance and 401(k) accounts (five employees are on pace to be 401(k) millionaires) and an average wage four times what it was a decade ago. The little Baltimore factory runs double shifts. Most remarkably, Marlin is still making wire baskets--just not bagel baskets. Or at least not very many.


The job that rescued Marlin Steel was small--20 baskets, a $500 order. Greenblatt was handling sales in 2003, so he took the call himself. "It was an engineer from Boeing," he says. "He didn't think I was in the bagel-basket business. He just needed custom wire baskets." The Boeing engineer, who had seen a Marlin ad in the Thomas Register, a pre-Internet manufacturing directory, wanted baskets to hold airplane parts and move them around the factory. He wanted them fast. And he wanted them made in a way Marlin wasn't used to--with astonishing precision. For bagel stores, says Greenblatt, "if the bagel didn't fall out between the wires, the quality was perfect." The Boeing engineer needed the basket's size to be within a sixty-fourth of an inch of his specifications. "I told him, 'I'll have to charge you $24 a basket,'" says Greenblatt. "He said, 'Yeah, yeah, whatever. No problem. When are you going to ship them?'"

Custom-made for a supplier of GM auto parts, this basket holds pump housings firmly in place as they are washed; the mesh top aids water flow.

What got Greenblatt's attention in that phone call wasn't the need for speed or even the quality standards. It was that Boeing was completely unconcerned about price. "I'm trying to sell a basket for $12, the bagel shops are saying, 'I'm not paying more than $6.' I'm ready to jump off a bridge, and here's a guy who just shrugs at the outrageous sum of $24. I was like, Wow. He's price insensitive."

That epiphany marked Marlin's rebirth. The company would keep bending heavy-gauge wire to make baskets, but instead of going to Bruegger's to hold bagels, the baskets would go to the factories of Toyota and Caterpillar, Merck and GE to hold everything from microchips to turbine blades.

Greenblatt almost didn't make the leap. "Actually," says Andy Ratner, one of Greenblatt's office staff, "Drew's first reaction was: This is totally not what we do." Greenblatt adds: "The Boeing guy needed a plus-or-minus tolerance on the wires. We weren't used to that at all. We just used a tape measure."

But Greenblatt had to try something new. On top of the competition from China, the bagel business was collapsing after a period of wild overexpansion in the 1990s. Einstein Bros. and Bruegger's went from 303 stores in 1995 to 1,044 stores in 1997--together opening 30 new bagel outlets a month for two years. Just three years later, Einstein Bros. filed for bankruptcy protection and closed 200 stores. Bruegger's scaled back fast enough to avoid bankruptcy, closing 225 stores in the process. Bagel chains closing stores don't need new baskets.

So when the call came from Boeing, says Greenblatt, "we were in a horrific position." Filling the Boeing order involved a clumsy, painful process of trial and error, and Marlin had to throw away a lot of poorly made baskets. "It was not an elegant initiation to the world of precision," says Greenblatt. Boeing was prickly, demanding, unforgiving. An aerospace assembly line was waiting for the baskets. That's why they had to be perfect and why Boeing needed them fast--and why the price didn't matter. The cost of the baskets was trivial compared to the cost of not making airplanes.

The lesson for Greenblatt was that the wire basket was only part of his product. To Boeing, he was selling engineering, precision, and speed, too.


In a small glass display case at Marlin headquarters is a pile of recipes for the baskets the company made in the 1990s and early 2000s, showing the type of wire and measurements in inches, handwritten in black ink on the kind of memo pad that used to be common before email and smartphones. The pages bear the logo of Colonial Steel Corp., Marlin's old Bronx steel supplier. Marlin didn't even have its own stationery. "This was the intellectual property of the firm," says Greenblatt. "No blueprints, just slips of paper."

Owner Drew Greenblatt thought Marlin would give him an easy annuity. Then he found himself in a showdown with Chinese competitors.

After the Boeing job, Greenblatt did a little poking around, and he made two discoveries. Factories have a huge appetite for wire baskets, which they use to stage parts for assembly; a single big factory has thousands of them. And in the United States today, factories outnumber bagel stores by a hundredfold--333,000 to 3,100. "The bagel-basket business is a teardrop," says Greenblatt. "Factories are an ocean."
Getting 20 baskets to Boeing was one thing. Remaking Marlin so that companies like Boeing became Marlin's core customer was another. "It was a wrenching process," says Greenblatt. "We needed better equipment. We had to retrain our people. We had to improve our engineering. We really had to up our game."

Marlin also had to learn a lot about its customers' businesses, as it did three years ago with Delta Air Lines. Delta has a jet maintenance facility in Atlanta, where it refurbishes the turbine blades from its own jet engines and those of other airlines. "They use heat on the blades," says Kash Alur, a design engineer at Marlin. "They use chemicals. They can't touch the blades with their hands. And they need a basket that holds the parts in a specific orientation."

Alur flew to Atlanta to see what the baskets had to do every day. "Delta did have baskets, but they were falling apart," he says. "I took measurements. I took notes. I took pictures." Turbine blades are heavy but delicate. The baskets had to be durable. They had to be stiff--Delta didn't want any flex at all. And they had to be interchangeable, able to hold all kinds of turbine blades. "They didn't want nine different kinds of baskets for nine different kinds of blades," says Alur. In essence, Delta wanted a high-tech dishwasher rack.

A revamped design process helped give Delta what it wanted. Alur's designs are in Marlin's computer system. He can email them to Delta, share them with the four other design engineers at Marlin, or tweak a design and instantly see how the changes work. "They've reordered a couple of times," he says.

Even in the much more unforgiving world of high-tech factories, Marlin is earning a reputation for quality that is unusual for manufactured products today. Tom Salvador is a senior manager with Power Systems Manufacturing, in Jupiter, Florida, a small division of the $20-billion-a-year global energy conglomerate Alstom. PSM reconditions gas turbines for large commercial power plants, and it needed wire baskets to hold finished parts during a demanding inspection process. The baskets had to be made of stainless steel and able to withstand high temperatures and hot water. "It's a manufacturing environment," says Salvador. "We don't baby the baskets."

Marlin's goods held up where the competition's didn't. "There are a lot of welds on these baskets," he says. "We haven't had a single weld on a single basket cut loose. Not one. I'm actually shocked." Salvador has placed two orders with Marlin in five years but doesn't need any more now because, he says, "I can't seem to break them."

Manufacturers use these baskets--among Marlin's few stock products--to hold industrial parts during washing, degreasing, and ultrasonic cleaning.

Early in Marlin's reincarnation, Greenblatt learned the benefit of saying yes to every customer. That sort of flexibility helped Marlin win a job this year from a longtime German supplier to a U.S. auto-parts factory. The order was for baskets that are more like trays, about the size of the ones in a cafeteria, but weighing more than 10 pounds and made of stainless steel, with thick wire props in the middle. The baskets hold heavy auto parts so they can be washed, treated, and presented to robots for assembly. "Because you're staging parts for the robots, the parts have to be in exactly the right place," says Greenblatt--within three-thousandths of an inch in all three dimensions. The baskets, which will hold parts that go into Chrysler cars, cost $200 apiece. And the supplier needed 1,000 of them--a $200,000 order. "The German vendor had made this style of product for them for over 20 years," says Greenblatt, "and quoted them four months to make the new version." Marlin said it could do the job in four weeks. And it delivered. "If a car company doing a model-year changeover can get the assembly line going faster, the value of that extra three months of production is enormous," says Greenblatt. "The baskets are paid for in a couple hours."

Although it was a big order for Marlin, the auto-parts company considers it a sample--it's testing Marlin. "It's a big deal for a company like that to jump vendors," says Greenblatt. "But once you do that quality of work that quickly, they come to expect it."


Drew Greenblatt is both a success story and an eager storyteller, and as a result, he's become a celebrity talking head--testifying before Congress on trade policy and small business, and attending manufacturing events at the White House. He sits on the board of directors of the National Association of Manufacturers alongside the CEO of Caterpillar, and he has appeared, with his chubby-cheeked smile, on CNN, CNBC, and PBS NewsHour. "I'm talking about motherhood and apple pie," he says. "I'm a little guy. I'm crisp. And I call it like it is. I'm not a policy wonk. I have a payroll to make every two weeks."

In person, Greenblatt is a little frumpy and utterly unprepossessing. He often wears a pullover sweater with his sport coat, his hair is unruly, his shoes have holes in the soles. But people listen. This April, Greenblatt spoke at manufacturing conferences in Denver and San Antonio. The airplane maker Cessna and the chemical giant DuPont--both Marlin customers--have flown Greenblatt to events to talk to their staffs. Timothy Geithner, the former Treasury secretary, made a pilgrimage to Marlin's factory during a Baltimore visit in May 2012. (Greenblatt deputized his father to host Geithner; he happened to visit on the day Greenblatt was off at Cessna.)

Marlin Steel is not gigantic. The manufacturing floor would fit on half a football field, and the office space could not be more austere: whitewashed wallboard, zero-tuft gray carpet, few adornments, except for some wire baskets on display in a few places. When the stamping machines are running full tilt, the banging penetrates every corner of the facility.

Marlin's revenue is modest, too, even though it makes a lot more baskets than it did in the old days--250,000 last year versus 67,000 in 2003. The company had sales exceeding $5 million in 2012, and Greenblatt hopes to reach $7 million in 2013--which would be the eighth consecutive year of sales increases. (For the record, the big U.S. steel company Nucor does $7 million worth of business every three hours.)

But in some ways, Marlin's smallness makes its success more surprising. Greenblatt had no manufacturing experience before buying the company. His chief of manufacturing design, Tony Witt, is 25 years old, and the job is his first after earning a mechanical engineering degree from the University of Maryland. Andy Croniser, 33, Marlin's production manager, started in 2007 as an entry-level production employee.

Greenblatt has slowly accumulated smart machines to go with his smart staff, starting in 1999 with a $200,000 piece of gear that does wire bending and butt welding. (Incredulous Marlin workers likened the purchase to using a howitzer for killing mosquitoes.) Marlin now has $3.5 million worth of computerized industrial robots--a couple that can pull and bend hundreds of feet of wire a minute; an automated router; an automated welder; a steel-punch press; a cutting laser. These aren't the kinds of robots you see on newscasts about assembly-line work; they don't have jointed, anthropomorphic arms wheeling in every direction. They're really just highly automated, highly computerized, and very fast manufacturing machines. Greenblatt nods at one of them. "I could have had a yacht instead of that," he says, though Greenblatt isn't really a yacht guy.

The automated machines have helped make Marlin Steel possible, letting it produce quickly, in small batches, with great precision--and economically enough to compete, even against China. Marlin recently won a job from a Chicago company for 160,000 metal brackets a year, each one smaller than the palm of your hand. The manufacturer had been buying the parts from a Chinese company. Even though Marlin's salaries are 10 times higher than in China, the labor costs in Baltimore are actually lower.

One Marlin staffer earning $25 an hour can tend a massive punch press that turns out 2,000 perfect parts an hour. In China, one worker earning $2.50 an hour uses basic tools to make 75 brackets an hour. Even though you can hire 10 Chinese workers for the cost of one Marlin worker, together those 10 workers are making only 750 brackets an hour. So the labor cost of producing 1,000 brackets in Baltimore is $12.50, versus $33.40 in China--60% less.

The danger now is that China could once again undercut Marlin. The little factory in the scruffy industrial park doesn't look like it would be that hard to replicate. And that vulnerability explains why Greenblatt is always thinking about how to build "the moat"--a metaphor he borrowed from Warren Buffett to describe his competitive edge.

Speed is part of the moat, the willingness to do a job as quickly as a customer needs it. Quality is part of the moat too. Scattered around the factory floor are "check fixtures"--simple wooden boards with the outline of a wire part etched into the surface. To see if the part is made correctly, a worker slots it into the outline on the board. If it's made right, it fits.

Often three nearly identical images of a part are etched into the same check fixture. The one on the right is the customer's requirement--say, a tolerance of 0.12 inches. The one in the center is Marlin's "okay to ship" standard, always better than the customer request--say, a tolerance of 0.06 inches. The one on the left is the "Drew" standard--perhaps a tolerance of 0.03 inches--four times the precision of the customer's spec. "If we send them that left one, they will never leave," Greenblatt says. Toyota--legendary for its obsession with quality--was so impressed with Marlin's exactitude that in addition to baskets, it also buys check fixtures from Marlin.

But for Greenblatt, the most important element of the moat is what Marlin didn't have before Boeing called: engineering and design. No one at Marlin designs baskets on slips of notepaper today. Five of 28 employees are degreed mechanical engineers. "We give people slick, elegant designs that make it worthwhile to use us rather than a commodity-part supplier from China," says Greenblatt. More to the point, says designer Alur, "people come to us with a problem and we try to solve it." Marlin has taken something utterly pedestrian and turned it into a tool of innovation--for its customers.


The final line of defense for Marlin lies in the customers it targets: Greenblatt wants the toughest ones, the kind who make his competitors roll their eyes. "What I realized is that the customers who are a pain in the neck are really the great customers," he says. "Some people might say, 'Those guys [at Marlin] are crazy expensive.' But we find the people who appreciate that. I've never sold anything to Walmart."

What those customers realize is that baskets aren't interchangeable commodities. Zoetis Pharmaceuticals, a recent spin-off from Pfizer, is the world's largest maker of animal medicines and vaccines (including the popular canine arthritis drug Rimadyl). The company was facing a nagging, expensive problem. It has pumps that feed medicine into a machine that fills ampules, and the pumps kept locking up. The pumps are small--10 inches long--but expensive. When one breaks, repairs cost $4,000, and the only place to get the work done is from a company in Germany. The lead time is two months.

Zoetis figured out that the pumps kept breaking because some of the parts were delicate and prone to damage when the pumps were moved around the factory to be cleaned. The company didn't have a container for transporting them. Workers would pull the pumps and dump them in a bucket or a bin. Sometimes they would move them around in plastic tackle boxes.

The problem had some urgency. Zoetis needs to have four pumps up and running at all times, and they were breaking so often the company tried to keep four spares on hand, figuring that another four would be out for repair; often that number was higher. The solution was clear: The pumps needed a container that would cradle them safely and that could withstand the cleaning solution and the temperature and pressure of a sterilizing autoclave.

Lewis Ferguson, at the time a Zoetis reliability engineer, looked at medical equipment makers but struck out. "I thought about calling my pals from the space program at Boeing and Lockheed"--his previous job was at NASA--"but I didn't think we could afford that," he says with a laugh. Then he stumbled upon Marlin Steel's website, where he saw baskets that looked close to what he needed. He talked to someone at the company and agreed on a basic design. Then he asked Marlin to add handles. "They said, 'Okay, handles are no problem,'" says Ferguson. Then he decided the baskets needed a lid. "The folks at Marlin said, 'Okay, a lid, we can do that.'" Then he asked for higher-quality steel. "They said, 'Okay, that's a stretch, but we'll see what we can do.'"

Zoetis ordered 12 of the baskets, each about the size of a shoebox; they arrived in four weeks. "It took them half the time to make the baskets and deliver them as it takes to get one pump refurbed," says Ferguson. The baskets cost $4,000, or $333 apiece, almost 30 times what a bagel basket used to cost. In fact, the baskets are cheap. They'll pay for themselves if Zoetis sends just one less pump to Germany. The company has another five or six pieces of gear with similar problems. It will look for similar fixes.

So Marlin's story is a little more potent than it first appears. Greenblatt reinvented his company--both the products it sells and how it creates them. Now Marlin's baskets are doing the same for its customers: driving innovation in how they make their own products. "We have this myth that innovation is something that happens with CAD machines and scientists in a lab, drawing algorithms on whiteboards," says John Shook, a manufacturing expert and CEO of the Lean Institute, in Cambridge, Massachusetts, which helps companies use "lean" principles in their factories. "The lightning bolt strikes, and everything else is easy. That's not the way it works. Product and process innovation go hand in hand."

The relentless improvements in U.S. factory productivity are a rough measure of that kind of innovation, with whole industries slowly transforming the way they make things and the quality of what they produce. If a little company like Marlin can turn a simple wire basket into a competitive advantage, facing down both low-cost competitors from China and high-quality producers from Europe, then there's hope for whole swaths of American manufacturers.

Drew Greenblatt isn't doing anything close to what he imagined when he bought Marlin Steel in 1998. He thought it would be an easy annuity; all he would have to do was modernize things a bit, beef up the marketing and sales, and, as he put it, "be a gentleman to the customers."

Instead, after almost losing his investment, he's running a high-performance manufacturing facility, thinking about markets and moats, management strategies and employee incentives. (Marlin awards bonuses in each pay period, based on performance--and bonuses were paid in 14 of 26 pay periods in 2012.) Greenblatt's ambitions are expansive. He thinks Marlin's annual sales could reach $50 million in five years--more than seven times this year's projected revenue. "Would I have figured it out without the call from Boeing?" Greenblatt asks. "No. I was clueless."

Marlin's raw material--spools of wire, each as heavy as a car--hasn't changed. But everything else is different.

Or almost everything. A few loyal bagel stores still buy their metal baskets from Marlin.

"It's about one-half of 1% of the business," says Greenblatt.

IBM

When Lou Gerstner became CEO of IBM in 1993, most people expected that the best he could do was break the company into pieces, some of which might survive the tumultuous tech world that IBM had helped invent. IBM was devoted to selling mainframes and couldn't see beyond big hardware.

Gerstner laid off 100,000 employees and forced IBM's disparate divisions to work together both to woo customers and to service them. And he refocused IBM on delivering services and software instead of hardware. IBM figured out it was in the computing business instead of the computer business.

When Gerstner left in 2002, revenue was $81 billion and earnings $3.6 billion (the company lost $5 billion the year before he arrived). Today, two CEOs later, revenue is up 25% and profit up 360%, and IBM looks nimble no matter how fast the tech landscape changes.

Kodak

In 1998, Kodak had 74% of the U.S. film market, and it was a pioneer in digital cameras and home printing. But the film business disappeared more quickly than anyone anticipated, and Kodak never became a leading force in digital cameras, which themselves were swept aside by smartphones with decent lenses.

Kodak filed for bankruptcy in 2012. The legendary company that invented photography as we know it no longer exists: It makes neither cameras nor digital picture frames nor consumer film, and it no longer even has a website for uploading and sharing photos.

The company is expected to emerge from bankruptcy sometime this year as a provider of commercial services. As recently as 2012, the Kodak name still adorned the Los Angeles theater where the Academy Awards are held, but host Billy Crystal welcomed guests to "the beautiful Chapter 11 Theater."

Corning

The glassmaker is renowned for the technologies it has invented over its 160-year history. But one of its greatest talents has been reinventing itself. Corning made the first lightbulbs for Thomas Edison; it hasn't made lightbulbs for decades. It created the technology to make color TV picture tubes affordable; the business no longer exists. It invented the optical fiber that connected the world, sending revenue from $4.7 billion in 1999 to $7.1 billion a year later; that business crashed too, and by 2003, revenue was down to $3 billion.

Corning isn't exactly happy about that sort of roller-coaster performance, but it has plenty of practice surviving. Today, display glass is the biggest part of Corning's business--from LCDs for flat-screen TVs to the Gorilla Glass screens in most iPhones. And it has at least one safe harbor: Corning makes the guts of catalytic converters and diesel engine filters for cleaner vehicle emissions.

OmniGuide Surgical

Before OmniGuide even got going, it faced the Marlin bagel-basket problem: an overpriced product in a shriveling market. The company's founders, scientists at MIT, had invented a new kind of optical fiber. It was hollow, so light traveled through air, not glass, boosting efficiency by a factor of 10. But making the stuff in miles-long reels proved too hard, and no one wanted pricier fiber anyway, no matter how efficient. So the scientists sought a market that might need superfast, but short, fiber.

They found medicine. Surgeons had developed the CO2 laser scalpel, which uses light from CO2 molecules to cut without damaging nearby tissue. But it lacked a flexible medium; standard optical fiber absorbs CO2 laser light. The OmniGuide fiber transmits the light exactly where the surgeon wants it. OmniGuide now has more than $25 million in revenue and 150 employees--and manufactures its products in Massachusetts.

(Ed. note: A previous version of this story stated that Kodak no longer produced any type of film. In fact, Kodak does continue to produce professional, motion picture film.)

[Illustrations by Mike Pfaltzgraff]

Charles Fishman (cnfish@mindspring.com), a longtime contributor to Fast Company, is the author of The Big Thirst: The Secret Life and Turbulent Future of Water.

Flag sculpture by Kyle Bea; photos by Mitch Payne; Melissa Golden (Greenblatt)

Add New Comment

3 Comments

  • Wicked Wolf Gang

    All your bagel basket are belong to us! Great article & keep on trucking, Marlin Steel!

  • Leighton44

    Nice to see Fast Company do an article on "hardware" manufacturing instead of the plethora of articles on short-lived, over-hyped software companies. Please do more. 

  • Mark Chalfant

    Great article - really interesting look at a different kind of innovation.