Former Honeywell CEO David Cote just wrote one of the best guides ever on how to lead a company

It’s fulfilling these “we can do both” imperatives that lifts enterprises to their full potential. And the most important is simultaneously achieving what many managers find vexing if not impossible, “making the numbers” while at the same time making the daring bets to ensure those numbers will be far higher five years from now. As his book’s title suggests, Cote swears the idea that CEOs must choose between winning now and winning later is wrong: They must find a way to do both. Thinking short- and long-term works in concert since you need today’s profits to fund tomorrow’s hits. He shows how Honeywell succeeded in constantly improving its most lucrative existing products, from airline components to gas detection devices, to increase quarterly earnings, while simultaneously plowing billions into next-generation projects that took five or six years to harvest, then paid off big.

The book is packed with other examples of taking two actions that seem to butt heads and getting them to march arm in arm. While many companies think they need big inventories to ensure fast delivery to customers, Cote shows how it’s possible to both reduce stocks and lower shipment times by expertly managing the supply chain.

Cote also believes that companies should advance not in giant leaps but small agile steps. “It’s survival of the most flexible,” he writes. “Revolutionary change is not the key to short- or long-term performance. It’s a huge risk. Revolutions move in unintended directions. The best change is gradual change.” His goal is to stay just ahead of the market as opposed to betting heavily on future trends that may not play out, or falling behind so that only demoralizing, disruptive restructurings can save the company.

For Cote, it’s best to spread plant closures and layoffs over a number of years. Another tenet: Making lots of small targeted acquisitions is more profitable, and a lot less dangerous, than chasing the transformational whales. His contrarian thinking extends to personnel matters. Forget the “praise in public, criticize in private” maxim, says Cote. When a business head makes an inadequate plan for building his or her business, the CEO should uncork the likes of “This is unacceptable, go back and fix it”––and explain why––in front of the assembled peers to demonstrate that he’s setting a high standard.

The official publication date for Winning Now, Winning Later is June 30. But Cote gave Fortune an exclusive first look. I also interviewed him extensively to glean further insights. The overriding theme from both the book and these conversations is that this son of a rural New Hampshire gas station owner regards leadership as both highly intellectual, and highly hands-on. How many CEOs take full days, sans appointments, to think big and hatch bold new initiatives?

Courtesy of HarperCollins Leadership

Yet Cote insists that “the idea you focus on strategy and outsource implementation to great people is wrong.” At Honeywell, he personally interviewed all candidates for the top 300 jobs, and for the first 90 days after making every one of its 100 acquisitions held monthly meetings to review their progress and quarterly reviews for the next year. At the same time, as one lieutenant put it, he showed the tenacity of “a big annoying bear threatening the livestock” in constantly monitoring the numbers to ensure that each business was adding sales faster than people, investing sufficiently while reaching the right balance in product enhancements and long-horizon projects, and generating strong and growing cash flow.

Winning Now, Winning Later opens by describing the deep-seated problems at Honeywell. According to Cote, the “short-termism” that plagued Honeywell is an issue many companies grapple with. The best argument that Cote’s formula can achieve the dual objectives in the book’s title is its spectacular success at Honeywell. From 2002 to 2018, Cote lifted Honeywell’s sales from $22 billion to $40 billion, took operating margins from 8% to 16%, and hiked the stock price fourfold, swelling the market cap from $20 billion to $120 billion, while spending $10 billion resolving environmental issues, including asbestos claims, and fully funding the underwater pension plan. Here are five original concepts and strategies that formed the Cote blueprint.

Defeating short-termism

In 1999, Cote left GE after a 25-year career, then served for two years as chief of TRW before going to Honeywell in February of 2002. Cote coveted the ultimate prize of running GE upon Jack Welch’s retirement, but departed when the legendary CEO said that Cote––then heading the appliance division––wasn’t a finalist. Had Cote gotten the nod instead of Jeff Immelt, it’s likely that GE would be a far more successful company than what it became by overpaying for acquisitions in power generation, oil and gas, and security systems, and using commercial paper at GE Capital to fund long-term loans, a practice that backfired during the financial crisis. The irony is that at Honeywell, Cote found some of the poor practices that he yearned to fix at GE. And it was by overhauling those crippling methods that Cote got Honeywell roaring while his alma mater sputtered.

Cote recalls that at GE, success was all about “making the quarterly numbers” at all costs. Any given year, the outfit where Cote was working would hire 1,000 people to grow sales. Then in October or November, the managers would discover that they couldn’t hit the targets for the following year without laying off 800 of the folks they just added. “Why, I wondered, hadn’t they thought ahead and only hired 200 people instead of 1,000? We didn’t think we had a choice,” he writes. The layoffs were highly disruptive and required big restructuring costs. For Cote, the better practice was what he did at Honeywell: getting his business heads and staff people to plan for the following year’s personnel and other financial commitments early in the current year, avoiding the need for this wrenching cycle of layoffs and rehiring.

At Honeywell, Cote found a mindset of what he brands “short-termism run horribly amuck.” Managers at the specialty chemicals, aerospace, automotive, and controls conglomerate would offer distributors special discounts at the end of a quarter to boost sales, a practice known as “distribution loading,” that enabled them to hit their targets. The businesses had to quickly inflate inventories to make the deliveries, but the shipping department often couldn’t keep up, creating bottlenecks and delays that antagonized customers. And businesses, wrongly guessing in advance what distributors would want, often filled warehouses with product their customers didn’t want.

Often, 25% of the revenues from aerospace, controls, and specialty chemicals businesses would come in the last week of the quarter. Instead of buying at regular prices in the early weeks, customers waited Honeywell out until the last few days, and in managers’ desperation to ramp up sales, the deals would get better as the quarter drew to a close.

To keep hitting the mark, the business heads would resort to more gimmicks such as selling good businesses to book quick profits, or securing one-time cash payments from suppliers in exchange for guaranteed future business. That locked Honeywell into deals with single vendors that forced them to overpay for components or services over long periods. To his horror, Cote learned as much as 20% of Honeywell’s reported earnings were coming from those one-time items known as “specials”: One plant manager in Louisiana even cut down trees in the forest adjoining the factory and sold the lumber to make the numbers––and received a performance reward for getting it done.

In aerospace, managers would compete for new contracts for wheels and brakes by providing the first hundred “ship sets” of those components for free. That’s standard practice in the industry. The difference was that Honeywell was capitalizing the cost of the ship sets and amortizing it over 20 years. “They did it,” says Cote, “because investors are much less alarmed by things that don’t go in the income statement.”

That wasn’t the only case of super-aggressive bookkeeping. Honeywell would capitalize R&D so that it spent the research dollars now, but spread the reported costs into the future. As a result, Honeywell’s reported profits kept rising faster than its cash flow, signaling that its facade of profitability was about to crumble. For the decade prior to Cote’s arrival, Honeywell was generating just 69¢ for every dollar in earnings, a sure sign of trouble.

It didn’t take long for trouble to strike. After Cote had served just a few months, the finance department revealed that earnings for the second half of 2002 would fall 20% below Honeywell’s forecast to Wall Street. “‘What the hell was going on?’ I asked,” writes Cote. “Their response, ‘Well, the financial goals we were trying to meet were never realistic to begin with.’” The finance chiefs brushed off managers’ complaints, telling them to “Just get it done.” Honeywell’s stock price dropped 25%, and an enraged Cote told everyone to drop the tricks––no more aggressive accounting, no more capitalizing R&D and ship sets, no more distribution loading, and no more “make the numbers” meetings hosted by the finance staff that drove all of those dysfunctional decisions.

Producing both long- and short-term profits

Although the “make the numbers or bust” mindset was hammering Honeywell, Cote still believed that achieving regular quarterly gains in earnings––reaping a healthy crop today, no gimmicks allowed––was essential to success. At the same time, Honeywell would need to start investing now in new products, services, and processes, as well as international expansion, to secure its future, by planting seeds for tomorrow.

“We would have to win today and set ourselves up for tomorrow,” he says. “I realized we could do both at the same time. Short- and long-term goals are more tightly intertwined than they appeared. Short-term results would validate that we were on the right long-term path.” What’s more, he added to me in an interview, each goal reinforces rather than contradicts the other.

To get there, Cote vowed to follow three principles. The first was “Scrub accounting and business practices down to what’s real.” Second came “Invest in the future but not excessively” followed by “Grow sales while keeping fixed costs constant.” The theme was to plow current earnings into the long-horizon projects such as next-gen aircraft cockpits and new molecules for refrigeration that Honeywell was shortchanging, but not invest too much.

Honeywell would maintain sufficient quarterly earnings to demonstrate consistent progress. By expanding sales while holding the fixed costs, consisting primarily of labor, constant, Honeywell could generate the savings to keep raising its spending on both big, multiyear systems and enhancements to existing products and services and still generate the short-term returns investors expected. Those practices would get the flywheel spinning, and a “virtuous cycle” would take hold, which would allow Honeywell to generate even more cash to invest, which would lead to further performance gains.

Growing sales without growing fixed costs

Cote’s plan for restraining costs came in two parts. First, he immediately demanded that the four big overhead functions, finance, human resources, legal, and IT, representing about one-fifth of fixed expenses, hold their annual dollar outlays at 2003 levels, forever. “The goal was to double sales, so that overhead would fall by 50% as a share of revenues,” he told me in an interview. As it turned out, the costs of the four functions actually fell by 30% or $1 billion over 15 years, so that spending rose more like 30¢ for each dollar gain in revenues. “We used part of the savings to improve IT for all four functions, so that we needed fewer managers,” says Cote. “For every 10 managers who retired or left, we only had to replace about seven.”

In line with his conviction to achieve two seemingly conflicting things at the same time, he also demanded that the four functions demonstrate better service to their internal Honeywell customers, as determined by anonymous surveys. Cote also taught the Honeywell brass to take on broader roles, so that he could run the company with far fewer leaders. The leadership ranks over 16 years shrank from 740 to 650 even as sales almost doubled. “I did it for two reasons,” says Cote, “to save costs, but most important, because leaders create work for other leaders, and instead of focusing on markets and customers, focus on satisfying each other.”

The second initiative was an epic campaign to boost productivity in manufacturing. “If you can grow output much faster than you add costs, including for payroll and floor space,” says Cote, “you’re bound to generate the savings that both fuel new investment, and increase quarterly earnings.” In 2005, Cote dispatched a team to study Toyota’s celebrated Toyota Production System, at the automaker’s plant in Georgetown, Ky. Over the next decade, Honeywell gradually rolled out its Honeywell Operating System or HOS, largely based on the TPS practices of engaging workers to recommend and implement improvements in their factory-floor tasks that save costs and improve quality.

The initiative was so successful that Honeywell over that period increased its sales per employee by two-thirds. HOS also enabled Honeywell to keep shrinking factory space as it made more and more product. By 2018, Honeywell was producing far bigger volumes in only 70% of the plant footprint it was using in 2002.

Early in Cote’s tenure, Honeywell was manufacturing sensors in 37 small plants. Cote asked the managers to do a “white paper” exercise imagining the ideal footprint if manufacturing could be redesigned from scratch. The answer was more like 12 factories that each produced much bigger volumes. In a classic Cote “go slow to go fast” maneuver, the sensors unit spent 10 years making that blueprint a reality by packing all production into a dozen plants. “We did it laying off just a few workers a year so that we didn’t disrupt customer service,” writes Cote. “It was part of our philosophy of constant restructuring.”

Cote notes he reached his expense targets while still increasing the number of workers in manufacturing. “We started with around 60,000 in 2002,” he said in an interview. “And we hired about 20,000 or 30%-plus over 15 years. The key was that we doubled the size of the company, so the revenues grew much faster than payroll.”

Those cost-savings plans freed the cash for both short- and long-cycle investments. For Cote, the bigger challenge by far was finding the right mega-bets that would swell earnings anywhere from two to six years hence. “We were already getting lots of enhancements on existing products because people wanted to make their numbers for the quarter,” he says. “HOS enabled us to make those improvements a lot faster, but the big one was the seed planting.” He writes that prior to his arrival, Honeywell was spreading investment dollars across a broad spectrum of projects without carefully assessing their potential profitability, hoping that the more bets you made, the more would win. Cote moved to advancing fewer huge platforms that played to Honeywell’s greatest strengths, especially in aerospace, and that promised the biggest payoff.

One major hit was HFO, a new form of fluorine for industrial refrigeration. HFO was a breakthrough molecule that Honeywell developed in a quest to reduce the global warming caused by previous fluorine molecules. “That was really touch and go,” Cote told me. “It took five years, and there were two or three times when we thought it wouldn’t work.” HFO raises global warming 20% less than carbon dioxide and 1,500 times less than the previous molecule, and it has mushroomed into a highly profitable $1 billion business.

Another haymaker: Experion Orion Console, a monitoring system for oil refineries. “It’s like a jetliner cockpit for refineries,” says Cote. “It collects and processes huge amounts of data from the facility, and controls all the flows to ensure the highest possible productivity and safety.” It was Darius Adamczyk, whom Cote mentored and succeeded him as CEO, who shepherded the Experion Orion. Overall, says Cote, Honeywell now captures three-quarters of the big aerospace projects it competes for, up from half in 2002. And by the way, today’s hits are much bigger than yesterday’s.

Overall, says Cote, Honeywell’s strongest growth engine was the payoff from long-horizon projects. Of the extra $18 billion in sales generated from 2002 to 2018, $6.5 billion came from acquisitions, and the remaining $11.5 from its own products. And three-quarters of that organic growth flowed from Cote’s seed planting.

Why it’s good to criticize in public

In Winning Now, Winning Later, Cote relates that in large meetings, he’d listen to the presentations and encourage everyone to talk. And when the open discussion ended, he’d request opinions on how to proceed from everyone in the room. He always asked the lowest level managers to speak first, and as they talked, he’d show no hint of agreeing or disagreeing. “If their bosses spoke first, the lower level folks would be tempted to just parrot what their bosses said,” Cote told me. “I’d never interrupt and wouldn’t express my opinion until the very end.”

For Cote, teamwork consisted not of the usual groupthink but ensuring that everyone in the room cited facts and expressed opinions. Then Cote, as leader, would make a decision and explain his reasoning. Cote wasn’t looking for widespread buy-in. “Consensus was not the goal. A good decision was the goal,” he says. However, Cote always explained his rationale, to guide the organization on how the leader wanted them to think about problems and let the people who disagree understand that he weighed their arguments and respected their viewpoint. Cote says his goal was to be right at the end of the meeting, which might be different from what he thought at the beginning. And the only way to get as close as possible to “right” was to keep participants from knowing what he thought and getting facts and opinions from everybody.

Once again, Cote thinks it’s constructive to call out a leader who presents a weak business plan so that all present understands what the boss accepts. “Organizations need to learn what is acceptable and what is not,” he told me. “If a business head presents a plan that’s not well thought through, or gives me a great story instead of results, I’ll call them to task. Everyone around the table needs to hear that the work product is not acceptable.”

His technique isn’t to personalize the problem. “I won’t call them an idiot,” says Cote. Instead, he’ll ask the executive to start over and come back with a much better proposal. “The leader will walk out thinking, ‘That wasn’t a good meeting,’” says Cote. “But it creates a good dynamic. It builds respect for you when you tell them what’s wrong. They may panic, but they’ll come back with something a hell of a lot better.” Cote adds that he would always suggest ways the executive could improve his or her plan. “I’d always say, ‘Here are four or five things you need to think about,’” he recalls. “I worked for bosses who’d give you no idea what was wrong and just say, ‘Do better.’ When your people respond to positive suggestions, their business gets better.”

In managing through a downturn, plan how you’ll spring back stronger

During the financial crisis, Cote focused just as much on ensuring that Honeywell could roar back in the recovery as on lowering costs to preserve investment and profits. “An economist said that you’ll most likely come out of a downturn the inverse of the way you went in,” he relates in the book. “That made a tremendous amount of sense to me. If your sales fall 20% in the first six months of the recession, they’ll probably jump by 20%-plus to the same level when the economy bounces back.” Cote orchestrated his response to ensure that Honeywell could handle a huge surge in orders to capitalize during the inevitable recovery.

Cote laid off 3,000 workers, moving up reductions already planned. But for the rest of the 125,000-person workforce, he deployed furloughs to keep everyone employed at the same time his rivals were axing a big chunk of their workforces. Employees were furloughed without pay for four weeks a year. That move saved $200 million, but it wasn’t enough. Cote also reduced the 401(k) match by 50%, used HOS to keep trimming operating costs, and forced management to take some of the pain by scrapping bonuses for 2009. “We performed in line with our peers, with two big differences,” he writes. “We didn’t cut our long-term investment plans, and for the most part, we didn’t cut people. Recessions don’t have to destroy your foundation. It’s best to keep calm while everyone else is panicking.”

Cote pushed ahead with two key acquisitions: One of them, the purchase of bar code scanning specialist Metrologic, landed its leader Adamczyk. He also made special deals with suppliers to ensure that Honeywell got first priority on supplies in the recovery. “If airline hours decline 6%, you go down three levels in the supply chain for parts, and the suppliers cut their capacity by 50%,” he said in an interview. “Then when flights pick up, this avalanche of orders pours in, and you can’t get supplies.” Cote’s solution: Make advance payments to vendors in exchange for a commitment to serve Honeywell first when good times return, and guarantee big future volumes providing the suppliers give Honeywell first dibs.

The strategy proved a winner when sales took off in 2011 and 2012. His workers were ready to go, while rivals struggled to quickly rebuild their workforces and keep pace with orders, and stood in line behind Honeywell for supplies. Over those two years, its sales of aerospace parts soared 30% above pre-crisis levels, largely because the ever-agile Honeywell stole orders from its flat-footed competitors.

Believe it or not, this review only scratches the surface of this book, brimming with original strategies that actually worked. Cote swears that part of being a successful CEO is being a good teacher. Now, readers from around the globe can absorb his lessons.

More must-read finance coverage from Fortune:

• Why black-owned businesses were hit the hardest by the pandemic
• George Floyd protests force Britain to reckon with its role in slavery, leading some companies to pay reparations
• This influential crypto CEO warns that hyperinflation will be “the next big problem”
• Looking to invest in companies that care about equality? This NAACP-backed ETF may be the answer
• 6 reasons Boeing’s financial picture may be brighter than most assume