(FORTUNE Magazine) – THERE's a musical management message in Frank Sinatra's bouncy 1959 1959 hit "High Hopes," wherein an industrious ant confounds the skeptics by moving an entire rubber tree plant. That could serve as the new work song for industrial America. To a degree not seen since the Fifties, factories and labs across the U.S. are junking business-as-usual incremental objectives -- moving a few more grains of sand -- and striving instead to hit gigantic, seemingly unreachable milestones called stretch targets. From Boeing to 3M to papermaker Mead, managers and workers are aiming higher, working harder, and achieving more than they'd believed possible. Of course, there's a catch Sinatra never mentions: Employees are having to endure mounds, not anthills, of toil and trouble along the way. Stretch targets reflect a major shift in the thinking of top management. Executives are recognizing that incremental goals, however worthy, invite managers and workers to perform the same comfortable processes a little better every year. The all-too-frequent result: mediocrity. In the words of Charles Jones of EDS's management consulting unit: "If you don't demand something out of the ordinary, you won't get anything but ordinary results." That's the rationale for stretch targets, which require big, athletic leaps of progress on measures like inventory turns, product development time, and manufacturing cycles. Imposing such imperatives can force companies to reinvent the way they conceive, make, and distribute products. Why have stretch targets suddenly become the rage? The answer is that, despite the upswing in the business cycle, many companies now perceive that they must perform far better to prosper -- or even, in the long term, to survive. Industries from paper to appliances face relatively bleak markets in which prices are falling inexorably, sales volume is growing slowly at best, and low-cost foreign competitors are crowding the market. Dwindling margins make it increasingly difficult for companies to earn their cost of capital. Eventually, they'll lack the money to buy new railcars or build the next generation of billion-dollar paper mills. Companies conclude that traditional ways of doing business are no longer good enough. That's when they reach for stretch targets. CEOs embrace stretch targets with fervor. It's often the one subject that can turn a harried, taciturn boss into an impassioned chatterbox. "We're doing things we didn't think were possible," rhapsodizes Boeing CEO Frank Shrontz, a man not known for overstatement. Another evangelist is Jack Welch, CEO of General Electric: "We used to nudge the peanut along, moving from, say, 4.73 inventory turns to 4.91. Now we want big, stretch results like ten turns or 15 turns." It's a good thing the bosses are so excited. Getting an organization to embrace wrenchingly difficult new goals -- particularly in the absence of a crisis -- can traumatize employees. Managers who can't stand the relentless new pace quit or get fired. Says Steven Mason, CEO of Mead: "A little over half the managers can adjust to stretch targets. For the program to work, the rest have to go." Mason adds that companies don't go for stretch targets unless they must: "It's unsubtle, painful, tough, and no fun." Four masters of what may be called the art of stretch management -- Boeing, Mead, 3M, and CSX -- rely to varying degrees on a set of nuts-and-bolts techniques. --Honesty is the best management practice. To instill the sense of urgency necessary for radical change, a CEO must level with employees, explaining in clear, convincing terms why the company must either change or fall on hard times. The goals must ring true: Imposing arbitrary objectives is the quickest way to turn employees off. A stretch target, such as halving the time needed to produce a mainstay product, must derive unambiguously from the corporate goals.((Classic example: Jack Welch's dictum that in unless each GE unit was No. 1 or No. 2 in its worldwide market, the unit must be fixed, sold, or closed.)) -- Employees must be convinced that they're not being asked to do the impossible. Benchmarking is a powerful persuader: showing that factories or labs at other companies -- often in other industries -- perform at levels that can never be achieved by incremental improvement. Seeing outsiders excel doesn't just teach managers how to, say, cut inventories; it is a potent psychological tool to enlist them in the crusade. If others can do it, they reckon, so can we. So powerful are the forces of pride and peer pressure that companies using stretch targets rarely view bonuses or other compensation incentives as key to the programs. --Finally -- and here's where stretch targets differ from old-fashioned top- down management by fiat that U.S. companies have spent years unlearning -- the CEO has to get out of the way. The job belongs to managers in the field, workers on the plant floor, and engineers in the labs. For CEO John Snow of CSX, the $9.5 billion-a-year railroad and shipping company, stretch targets were a natural extension of his business approach. The lanky Snow, 53, is a champion of management by economic value added (EVA), a form of analysis that calculates a combined charge to the business for debt and equity and then defines success as delivering returns that consistently exceed that hurdle. In 1991, CSX's return on capital hovered well below its capital charge in the range of 10%. Snow's bold goal: to make sure CSX would earn the full cost of capital by 1993 and exceed it thereafter. As Snow foresaw, the goal forced managers to stare smack at the railroad's core problem -- the fact that its multibillion-dollar fleet of locomotives and railcars sat loafing much of the time at loading docks and seaport terminals. Raising the company's return on capital would mean working the massive fleet far harder than had ever been attempted. Subordinates fretted that the task seemed impossible, but Snow refused to back off. He quips: "My son doesn't get better grades than I expect." Having set the target, Snow got out of the way: "It's people in the field who find the right path." The strategy proved a winner. Since 1991, while handling a surge in business, CSX has eliminated from its rolling stock 20,000 of its 125,000 cars -- the equivalent of a train stretching from Chicago to Detroit. That caused capital expenditures for supporting the fleet to shrink from $825 million a year to $625 million. CSX is now earning its full cost of capital. Stretching infused CSX's $1.4-billion-a-year coal division with new life. The division now hauls slightly lower volumes with 31,300 hoppers than it did with 41,500 in 1991. That represents a more than $150 million reduction in capital. Now the division is aiming for 28,800 cars on higher volumes in 1995, another stretch goal. Pete Carpenter, head of CSX's railroad business, decided that hitting such targets will require radical changes in the division, which up to now has run scheduling, sales, maintenance, and other functions spanning two-thirds of the country from its towering riverfront headquarters in Jacksonville, Florida. Result: a fascinating experiment in decentralization that is already yielding spectacular results. Last year Carpenter dispatched five volunteers from Jacksonville to tiny Cumberland in the Appalachians, where they were to set up as an independent profit center in the business of hauling coal for customers in western Maryland. Ray Sharp, the new unit's president, says he gave up his comfortable post as sales manager serving the coal division's biggest customer, utility companies, without a second thought. "I was thrilled to be an entrepreneur," he says. "I knew we could run the business a lot better from on the ground than from Jacksonville." Conditions in Cumberland were primitive. Sandwiched in a ramshackle building next to a freight yard, Sharp and his team labored without secretaries for months. In the summer, weak air conditioning forced them to keep the windows open, allowing clouds of acrid diesel smoke from the locomotives to blow in. The pioneers padded around in golf shirts and Docksiders, and over pool and poker in the evenings, bonded into a feisty fraternity, exhilarated by their newfound independence from headquarters. Carpenter wasn't coy about wanting quick results. He demanded that the region go from breakeven to a substantial profit within a year, and eliminate 800 of its 5,000 hoppers. "If you don't do it, we'll no longer need you up here," he told Sharp and his men. Carpenter didn't say if they could expect jobs back in Jacksonville. The stretch target provided a spur. "I thought the goals were impossible," says Sharp. "But without them, we would have gotten comfortable and kept using railcars like they were free." The team quickly decided that hoppers and locomotives still spent too much time idle. "We'd look out our office windows at the tracks and wonder, 'Why aren't the cars moving?'" says finance director Peter Mills. The problem, the team finally realized, was the railroad's practice of running the longest trains possible -- typically 160 cars. At the port terminal in Baltimore, hoppers often sat for two or three days until enough of them accumulated to form a train. At rail yards in the coal country, assembling a train of 160 hoppers could take the better part of a week. The cause of the problem was a logjam at headquarters. Executives responsible for equipment utilization wanted to keep the hoppers busy by running shorter, more frequent trains. But they clashed with executives in charge of manning, who resisted shorter trains to hold down labor costs. As head of a profit center, Sharp told the bureaucrats he'd gladly pay for extra crews in order to reap far bigger savings from cutting the fleet. Now his unit runs trains with as few as 78 cars, and hoppers never sit in Baltimore for more than a day. "I raise hell with headquarters to keep the cars moving," he says. In just over a year, Sharp's group surpassed its stretch targets by cutting 1,000 of 5,000 railcars and 25 of 100 locomotives. It also mined more business, lifting volumes by 6%. The fleet reduction lowered depreciation and operating costs, causing profits to bubble. No wonder CSX has begun cloning the Cumberland unit in other regions.

THE STRETCH TARGET at 3M has been to vastly improve the creation of new products. That may sound like just more of the same: The $14-billion-a-year company has long been renowned for product innovation. In its potpourri of 66,000 offerings, which include Post-It message pads, Scotch videotapes, and household sandpaper, 3M consistently met its goal: deriving 25% of its revenues from products introduced within the past five years. Three years ago, however, L.D. DeSimone, the stocky, cheerful CEO, became convinced that the standard had lost its magic. Revenues from new products were flat, and 3M was facing grim market conditions that seemed unlikely to improve: Adjusting for inflation, it hasn't been able to raise prices since 1988. Worse, 3M's overall growth had almost completely stalled. Sales rose less than 1% in 1993 and profits only 2.4%, far from the 10% annual increases that DeSimone wants to see by the late 1990s. In 1991, he unveiled a stretch target: Accelerate innovation to the point where 3M generates 30% of its sales from products introduced within the past four years, a 30-and-4 standard to replace the revered 25-and-5. Since new products grow far faster and generate higher margins than old ones, argued the CEO, the higher innovation rate would add the necessary octane to 3M's performance. Sporting a pink-and-gray paisley tie, his nickname DESI embossed on a laminated ID tag, DeSimone says simply, "We couldn't hit our growth targets without stepping up new products." The goal sent tremors through 3M's campuslike R&D headquarters in St. Paul. It quickly became evident that sales of new products were flat partly because developers were wasting energy on too many small products similar to ones already on the market. Products were taking too long to progress from prototype to full production -- as much as five years. DeSimone prodded his lieutenants to concentrate on products with the potential for big sales and launched a so-called pacing program to identify possible blockbusters in the lab and rush them to market. Pacing helped break the R&D traffic jam. In 1992 the $5.2-billion-a-year consumer and industrial goods business earmarked 50 new products for the program; more than half have already reached consumers. According to the sector's president, Ronald Mitsch, the eight biggest products alone will generate $1 billion in annual sales by the late Nineties. Meanwhile, 3M is closing in on its targets for corporate growth. So far this year, sales are up 6% and profits around 12% What was the hottest product to come out of the pacing program? Scotch-Brite Never Rust soap pads, which might be described as a quantum leap in the science of scrubbing. Ordinary scouring pads, like Brillo and SOS, are made of steel wool that can leave rust stains on the sink and tiny metal splinters in dishpan hands. Never Rust, by contrast, is made from recycled plastic beverage bottles. It's a web of plastic fibers coated with fine abrasives. It won't rust or splinter. Having test-marketed small numbers of pads from a pilot plant in 1991, 3M was encouraged by the results; in January 1992 it made Never Rust one of the first products in its pacing plan. The result was the fastest product introduction in company history. Almost immediately, division headquarters granted approval for a new plant, exclusively for Never Rust, in Prairie Du Chien, Wisconsin; construction crews broke ground in March 1992. The timetable was incredibly ambitious: It called for the plant to reach full production by November, and for 3M to launch the pads by January 1993. Workers started installing production equipment before the windows, bathrooms, or heating. When engineers hit a roadblock, they called on colleagues from all over 3M. Experts from the abrasives unit, for example, trekked to Prairie Du Chien to help with a crucial process: coating the pads with fine particles of abrasive. The plant hit full production right on schedule. Traditionally, 3M would have waited to make sure the machinery worked before developing a marketing plan; getting the product into stores nationwide would have taken another year. But Never Rust marketers had already been producing TV commercials and arranging promotions for months. The day the plant was finished, it started churning out pads and trucking them to supermarkets. The marketing campaign began with promotions in January and climaxed in a splashy nationwide launch in March. In the ensuing 18 months, Never Rust has made it onto a lot of sinks, capturing a stunning 22% of the $100-million-a-year U.S.market from Brillo and SOS.

LIKE CSX, Mead is clearing a path to profits in a bleak, capital-intensive business. When Mason, a 37-year Mead veteran, became CEO in 1992, the $4.8- billion-a-year papermaker was a cushy, complacent place to work. The boom- and-bust cyclicality of the paper industry offered easy rationalizations for poor performance. To many executives, paltry profits in the down part of the cycle were as natural as changing seasons. They simply waited for a windfall when business rebounded. Says Mason: "We accepted that our destiny was to be average." Anything average rankles Mason, 58, a wiry dynamo who retrieves balls tirelessly on the tennis court. Says one lieutenant: "Perfection isn't good enough for him." Mason was convinced Mead was in jeopardy. He argued that it was folly to bank on a cyclical upswing, because overcapacity and foreign competition would soak up any increase in demand and make it impossible to raise prices. Mead, which had difficulty earning its cost of capital even in peak markets, now faced the prospect of living permanently below the bar. That was the formula for doom: Chronically low profits would ultimately rob Mead of the ability to pay for expensive new equipment it would need to compete. It became clear, says Mason, "we need to prosper even if prices don't come back." He settled on using stretch targets after deciding that most management tools were either fads or too soft. In the 1980s and early 1990s he had watched Mead adopt and discard total quality improvement, management by objective, and continuous improvement. Of continuous improvement, he says, "we missed the point. We'd flutter up a little in performance and feel good about ourselves. But we really needed to jump to a new plateau." In 1991, Mason delivered a brash plan: to raise long-term return on capital from 5% in 1992 to between 10% and 12% by the late 1990s. From that goal he derived a gritty stretch target using a formula adopted from GE that measures productivity as a ratio of revenues to costs adjusted for selling prices, raw material expense, and salaries. Says Mason: "The beauty is that it concentrates on things managers can control." By the new yardstick, Mead's performance looked pitiful. Productivity had risen an average of 0.3% a year. Based on that analysis, Mason set a target he describes as "reasonably unreasonable": a tenfold stretch to 3% productivity improvement a year. The toughest challenge for Mason was to convince employees that they needed to act. "In a crisis, you can do anything. But we had no wolf at the door. People wanted to wait until prices came back," he says. Mason sent managers to visit GE's appliance and light bulb divisions, businesses that earn brisk profits in mature industries with flat prices by relentlessly honing productivity. Some of the managers were inspired. With those who weren't, Mason fell back on autocratic stubbornness. "People would say, 'This is a different business; we're already low cost,'" he recalls. "I'd say, 'That's interesting. Your goal is 3%.'" Since Mason tightened the screws, at least one-third of the 900 managers have left Mead. Those who remain have pushed the company to annual productivity gains of 2.6% over the past three years. Just as Mason had warned, a full-blown cyclical upswing has not materialized: Though paper industry sales are up, prices are barely rising. In the 12 months ended September 30, Mead has raised revenues 8% and earnings 24%. How does Mason react to the fact that Mead still hasn't hit its stretch target of 3%-a-year productivity gains? He has raised the target for 1995 to 4%.

BOEING has given itself stretch targets to overcome a major weakness few outsiders realize it has. In contrast to its sleek, sophisticated planes, Boeing's manufacturing process is shockingly primitive, cumbersome, and slow, a problem compounded, says CEO Shrontz, by veteran managers' mulish resistance to change. "We come from an incremental culture," says Shrontz. "I knew we could do far better in manufacturing if we really stretched, but managers kept saying 'Airplanes are different. They're far more difficult to make than other products.'" His lieutenants pointed out that Boeing was No. 1 in its industry and arguably far more efficient than other airplane makers. For Shrontz, being the best in an inefficient industry wasn't enough. Boeing's toughest competition, he argues, isn't just Airbus Industrie but old aircraft that are still in use. The prices of new planes have climbed so high that airlines put off buying and keep old aircraft flying. Merely convincing airlines to replace any plane that is at least 20 years old would swell overall demand in the late 1990s by one-third, or some $14 billion a year. Boeing needs a share of that business to stay healthy in the long term, warns Shrontz: "Replacement business means the difference between slow growth and the strong sales needed to finance future models." His straightforward solution: to lower prices so dramatically that it becomes cheaper for an airline to buy and operate new planes than to repair and operate old ones. In 1992, Shrontz set a stretch target: paring the cost of manufacturing a plane 25% by 1998. That would allow Boeing to raise the price of its 747-400, for example, currently about $150 million, far less than inflation, while adding new features, such as a new design on the wide-bodied 777 that makes it far easier than on previous models to reconfigure galleys and seats. Shrontz set a second stretch target that complements the first: to radically reduce the time needed to build a plane -- for the 747 and 767 -- from 18 months in 1992 to eight months in 1996. Speeding production helps cut costs by causing inventories to shrink. It's also a way to woo the airlines, by decreasing the risk involved in ordering a new plane. Up to now, doing so typically meant a two-year wait; with the short lead time, a United or Southwest can order as the business cycle takes off and roll the planes out while the recovery is in full flight. Managers ultimately found the CEO's logic hard to resist. But Shrontz still had to convince them to adopt lean manufacturing techniques from other industries. Benchmarking helped: Compared with GE, which turns its inventories in its heavy manufacturing operations at least seven times a year, Boeing's snail-like practice of turning its $8 billion inventory twice a year was embarrassing enough for the organization to strive for far better. To inspire his managers, Shrontz dispatched teams to study the world's best producers of everything from computers to ships. Progress is already remarkable. Boeing now designs and builds each order of wide-bodied 747s and 767s in just ten months; 737s and 757s require just 12. Rather than simply tweak old techniques, the company is installing new approaches in almost every production phase. The greatest gains came in the overhaul of Boeing's tortuous design process. Airplanes are the ultimate in customization. Each airline demands a unique configuration of galleys, lavatories, and seats, as well as engines, electronics, and landing gear. To avoid having to draw blueprints from scratch with each new order, engineers would scour cavernous archives for blueprints from past orders that matched portions of the new one. The engineers would then laboriously compile the new plans by copying parts of the old ones and adding their own modifications. The result was a thick sheaf of blueprints that included hundreds of thousands of part numbers transferred by hand from the fading old plans. As late as 1992, merely to design an order of 767s and procure the parts took a year of full-time toil by more than 1,000 engineers. To ensure that pipes and wiring ran properly through sections of the plane, Boeing would often create life-size mockups in plywood and plastic. When the planes finally went into production, mistakes would emerge by the dozen: Engineers frequently copied incorrect component numbers onto the new blueprints, causing slews of wrong parts to arrive at the plant. The company is replacing that lumbering process with a huge computer library of parts and configurations for engineers to mix and match. Reconfiguring, say, the forward fuselage of a 767 takes weeks less than before. Production is faster too: The assembly plant receives the right parts; wiring and piping fit perfectly.

BOEING is making comparable leaps in other phases of manufacturing, such as parts production and final assembly -- building big sections of the plane in parallel rather than in sequence, for example. To hit its stretch target in inventory turns, the company is bubbling with new ideas. One source of innovation is the Sheet Metal Center, a unit that makes 100,000 different parts a year, including the skins and frames that form the shell of an aircraft. Director Joe Peritore and his staff have invented a new way of producing doors. Until recently, Sheet Metal supplied door parts in big bins to assembly division warehouses. Workers would spend hours sorting through the bins before the doors could be built. Now Peritore's team delivers door parts in ready-to-assemble kits directly to the assembly bays, totally bypassing the warehouses. "I no longer have to rummage through boxes to find the right part," says Doug Russo, who assembles doors in Renton, Washington. Producing kits for entire modules shows how stretch targets draw the best ideas from the plant floor. Since 1993, Sheet Metal has cut stocks awaiting assembly from $270 million to $130 million. Now, Boeing is extending the kit concept to wing assemblies and other sections. For many companies, reaching for heroic goals isn't an option, but a necessity. It takes a visionary boss like Shrontz or Mason to see the threat over the horizon and act before it's too late. And it takes adroit management to enlist the people on the factory floor who fashion airplane components, man the paper mills, hatch new products, and move the rubber tree plants.

BOX: HOW TO STRETCH

-- Set a clear, convincing, long-term corporate goal. Example: earning the full cost of capital. -- Translate it into one or two specific stretch targets for managers, such as doubling inventory turns. -- Use benchmarking to prove that the goal -- though tough -- isn't impossible, and to enlist employees in the crusade. -- Get out of the way: Let the people in the plants and labs find ways to meet the goals.