The Surprising Success of Hands-On Leaders

When we talk with business leaders, one question we seek to understand is this: With all the tasks you could prioritize and the limited hours available each day, how do you choose what to work on?

Among most CEOs, there’s broad consensus: Senior leaders should focus on the “what”—purpose, vision, strategy, goals, resource allocation, and assembling a capable team. To protect their time for those high-level priorities, they must delegate the day-to-day operating decisions (the “how”) to subordinates. Peter Drucker, arguably the most influential thinker on the manager’s role, wrote, “The executive is not supposed to be a handyman. He is supposed to be a builder.” In the New CEO Workshop program that one of us (Nitin) coleads at Harvard Business School, the faculty admonish new chief executives to move beyond execution and focus on the big picture. They say, “To truly be the chief executive officer, you need to stop behaving like the chief operating officer.” Leaders who get too deep into the details of execution are typically criticized as being micromanagers.

Yet when we study some of the world’s top-performing firms, we see a contradictory set of behaviors. These are companies where the CEOs and senior leaders care deeply about the “how”—that is, how people do their work—spending significant amounts of time observing the way middle managers and frontline employees work and designing systems and modeling behaviors that guide the methods people use to perform their tasks.

For this article we’ve conducted intensive research on four extraordinarily high-performing firms: Amazon, Danaher, RELX, and Toyota. We’ve visited their facilities, interviewed more than 20 executives, and read much of what’s been written about the factors that have led to their success. On the surface these four organizations have few attributes in common. They operate in different industries on three different continents and have diverse histories spanning family-run, founder-run, and professionally managed structures. One was born in the internet age, one in the early 1980s, and two are more than a century old.

The commonality is that each company has leaders who embrace a contrarian view of what a CEO should prioritize. Our study put a particular focus on Jeff Bezos, who founded and led Amazon for its first 27 years; Larry Culp, who led Danaher from 2000 to 2014, went on to steer a remarkable turnaround at General Electric from 2017 to 2024, and remains the CEO of GE Aerospace; Erik Engstrom, who led RELX’s then-largest division starting in 2004 and has been the CEO of RELX since 2009; and Eiji Toyoda, who led Toyota from the 1960s into the 1990s. (He died in 2013.) Leaders like these reject the model in which a CEO is primarily a delegator; instead, they see the chief executive as a vital participant in shaping how work gets done.

To be sure, many CEOs would claim they care about execution. What distinguishes the leaders of these high-performing firms is their sustained and close attention to behaviors and systems. They are not inserting themselves into every decision or displacing their teams. Instead, they act as teachers and system builders: They’re present in the work not to control it or make every decision themselves but to model standards, sharpen problem-solving, and establish behavioral norms that enable others to act with autonomy and discipline. They don’t meddle—they coach. They don’t override—they elevate. They don’t hoard decision rights—they teach others how to make sound decisions on their own. Their involvement is not disempowering—it is energizing. And it is purposeful: to build a system that performs reliably even when they’re not in the room.

In this article we will explore how these company leaders created or sustained systems and cultures in which leaders prioritize managing the how alongside the what. We’ll showcase the techniques they’ve used to do this. We’ll parse the difference between obsessive attention to detail (a positive) and micromanaging (a negative). And we’ll examine how creating an operating culture that’s passionate about the how can be an underappreciated source of durable competitive advantage.

In the course of our study, we found five core principles that guide hands-on leaders. Let’s look at each one in turn.

They Obsess Over the Metrics That Customers Value

RELX is a 200-year-old London-based company formerly known as Reed Elsevier. In 1995 Forbes magazine predicted the company’s business model would make it “the first casualty of the internet.” In fact, since becoming its CEO, Engstrom has transformed RELX (its name since 2015) into a technology-driven information and analytics powerhouse—one that the London Stock Exchange last year named as the best-performing stock in the 40-year history of the FTSE 100 Index.

Engstrom, who was a McKinsey consultant and an executive at two publishing houses before assuming leadership of RELX, did that by coaching every RELX employee to obsess over “customer value”—the benefit the customer realizes from using a product. He has been repeating the same questions for 20 years: How does the customer measure value? How do we know? How do we measure that? How does using this product improve the customer’s economics? How do we know how much better off the customer is with our product—and how do we know that it is better on that metric than the alternatives are?

All too often when companies talk about customer metrics, they look at numbers such as customer acquisition cost, customer retention, lifetime value, and average transaction size—metrics that reflect how a customer benefits the company. In contrast, the how-obsessed CEOs we’ve studied focus on metrics that reflect how the company benefits the customer. Consider two examples at Amazon. Many companies say they offer low prices. But Bezos has held his group accountable for actually delivering the lowest prices and had a team build a bot to continually track Amazon’s and rivals’ prices on 1,000 items. The bot reprices a product if a competitor’s price drops. This enables Amazon to consistently offer customers what they want—the lowest prices on the widest assortment of goods. Bezos also realized that fast, reliable delivery is important to online shoppers. So he directed a team to build systems to measure precisely how long each package takes to get from order to delivery—then held the order fulfillment teams accountable. This approach has helped Amazon make deliveries progressively faster: reducing a week to two days to, in some cases, hours.

Focusing deeply on how the company creates and delivers value for its customers was a cornerstone of the practices of the leaders we studied. According to our interviews, people in the organization don’t experience this specific kind of detail orientation as micromanaging. Rather, it creates mission clarity. When leaders show how much they personally care about what matters most to customers, attending to details becomes a shared norm for every employee—which expands the decision rights of those close to the front lines.

They Architect the Way Work Gets Done

Amazon was founded in 1995, and by 2003 it was a large company—a survivor of the dot-com bust and a darling of the internet age. Yet inside the company, employees were already seeing alarming signs that it was becoming slow and bureaucratic. “To many of us, Amazon feels more like a tectonic plate than an F-16,” one engineering leader wrote in a memo that year. That frank admission, along with other signs of sluggish decision-making, led Bezos and his lieutenants Rick Dalzell and Jeff Wilke to rethink the way product teams worked—putting a relentless focus on speed. (Disclosure: One of us, Scott, served on Amazon’s board from 1997 to 2002 and remains a shareholder.)

Bezos, one of the most analyzed leaders of his generation, brought a unique background to his work. Raised by a family of tinkerers, he studied engineering and worked at a detail-oriented New York City hedge fund before driving west to found his online bookstore. In the early days he built desks and packed boxes himself—and even as Amazon grew he was determined to keep a hand in the work, identifying and removing barriers that slowed teams down. He redesigned work to create a system of independent, decoupled teams who were guided by a set of hows dramatically different from prior practices at Amazon and other large tech companies. For example, most software companies require teams to utilize one another’s code to increase efficiency. To increase velocity and reduce dependency, Amazon does the opposite: Its software teams are mostly free to choose whether they use common services or create their own building blocks of code. This results in duplicative work and some inconsistencies in the user experience, but Bezos justifies it by saying he’d “rather have two than none.”

Some of Amazon’s managerial innovations have become celebrated. For example, teams adhere to the “two-pizza rule,” limiting team size to only as many people as can be fed by two pizzas (typically eight or fewer). PowerPoint presentations are forbidden because they lull participants into passivity; instead, Amazon requires every proposal to be written in a narrative memo of six pages or fewer, to deliver more detailed thinking than is typically found in slide presentations. After all the meeting attendees read the memo closely, they have a nonhierarchical, no-holds-barred debate. The goal is to interrogate everything thoroughly, and vigorous dissent is encouraged. Before stepping down as CEO, Bezos participated in hundreds of these meetings himself. Instead of sitting back and waiting for the CEO to make the decision, everyone at Amazon is expected to speak up and disagree—even with Bezos or his successor, Andy Jassy. Once a determination is made, the CEO supports the path forward, knowing that most conclusions are provisional and can be revisited as new information comes in—what Amazon calls “two-way-door decisions.”

This is what we mean when we say leaders architect the way work gets done. It’s not about changing org charts. It’s about shifting decision rights closer to the front lines—and equipping those teams with the tools and frameworks that allow them to act. Those include customer-value metrics, structured memos, and experimentation protocols. It’s difficult work because it often requires leaders to give up control—to redistribute authority, remove approvals, and reduce the frictions that slow teams down. When done well, it promotes speed, clarity, and autonomy—which is why Amazon, a giant company, continues to launch innovative products and services quickly.

All four leaders we studied recognized that frontline teams are key to creating and delivering value for customers, whether they are internal or external. They made it their personal mission to design work processes so that employees are empowered with the tools and support they need and don’t hit obstacles that slow them down.

They Use Experiments to Make Decisions

Toyota may be best understood as a system of nested experiments—from the assembly line to the CEO’s office. This approach stems from the Toyota Production System (TPS), developed after World War II by Eiji Toyoda and Taiichi Ohno. Their system is built upon two pillars: continuous improvement (kaizen), which drives relentless pursuit of better ways of working, and respect for people, which empowers employees and suppliers to contribute their creativity and judgment. The resulting system is both highly efficient and profoundly human centered. One principle is this: Decisions are not made by rank or hunch—they are tested. That makes for better decisions, and it commits leaders at every level to humble learning. Everyone—from frontline workers to executives—is expected to propose ideas and test them through structured experiments.

At Toyota frontline workers are coached to redesign their own workflows—a responsibility that in other companies would be reserved for specialists or managers. That’s why visitors to Toyota’s production lines are often surprised to find racks and jigs made of light plastic piping rather than heavy steel. The point is not permanence. It’s adaptability. Toyota wants it to be easy for workers to modify the equipment they use.

Plant managers test their own proposals against alternatives from subordinates to see what actually works best. Even Toyota’s CEOs test their ideas when possible. In the 1980s, when Toyoda wanted to evaluate whether Toyota could manufacture cars in the United States, he didn’t make a strategic declaration—he ran an experiment. He persuaded General Motors to let him reopen a shuttered GM plant in California under Toyota management. The experiment was a success, and that led Toyota to begin manufacturing in Kentucky.

This commitment to experimentation shapes the culture. In their 1999 HBR article on the Toyota Production System, Steven Spear and H. Kent Bowen observed that “the scientific method is so ingrained at Toyota” that it prevents command-and-control dynamics and instead “stimulates workers and managers to engage in the kind of experimentation that is widely recognized as the cornerstone of a learning organization.”

That helps explain why hands-on leaders aren’t seen as micromanagers. When CEOs join in testing an idea—demanding that data, not hierarchy, decides—they elevate teams rather than override them. Authority comes not from opinion but from evidence.

They Lead by Teaching the Tool Kit

On several Sundays each year, 100 Danaher executives are flown from around the world to a single Danaher location. These leaders hail from different business units and functional roles, but for the following week they focus on a single set of tasks.

Is this a secret project? An emergency response to a crisis? No—this is a standard training practice at Danaher. The executives are divided into 10 teams of 10, each assigned to address a real business problem using Danaher’s kaizen method. Starting on Monday, each team follows the structured steps of kaizen: Know the customer and what they want, observe to understand the problem, analyze root causes via data, brainstorm countermeasures, and rapidly prototype. By Friday morning, each team will have implemented solutions and measured results. These weeks are not just about solving problems—they’re about learning, and teaching, Danaher’s system of hows.

When executives are hired from the outside, they aren’t given the roles they were hired for right away. Instead they spend two months in an immersive boot camp that inculcates them in Danaher’s tool kit, including concepts such as the voice of the customer, value stream mapping, standard work, policy deployment, and kaizen problem-solving. As Culp puts it: “We force division presidents to develop a command of the how so that they can teach the how. They shouldn’t say, ‘Go do that’ but instead, ‘Come do it with me.’”

You can feel the difference in meetings. Jim Lico, the CEO of Danaher’s spin-off Fortive, describes it this way: “It’s easy to set goals, to say the sales goal should be 9%, not 6%. What’s hard is teaching and instilling the tools to get there.” If Lico spends six hours in a business review, just 30 minutes might be devoted to reviewing financials. “The rest is applying the tools to improving the business,” he says.

In interviews for this article, Culp described using the same approach at GE. He believes hands-on, in-the-trenches work isn’t a phase to outgrow. “Too many leaders treat this work as developmental,” he says. “The best leaders don’t. They know the real fun is getting in with the team and getting their hands dirty. As a CEO you choose how you spend your time. There are many constituencies that make shrill demands for your time. You must attend to them, but you can’t allow yourself to get distracted from staying rooted in the real daily work of the organization.”

They Strive to Be Better, Faster, Cheaper—Every Year, Forever

Continuous improvement—driven and modeled by the CEO—is a defining feature of the companies we studied. At RELX, Engstrom distilled this ethos into a deceptively simple mantra: “Better, faster, cheaper—every year, forever.” But what sets RELX apart is not the language, it’s the discipline. Engstrom expects every team to translate that phrase into measurable outcomes, continually. Whether they’re improving fraud detection rates for insurers or shortening the time to legal insight for lawyers, teams are expected to show how each iteration delivers more customer value than the last. Because the bar never stops rising, each success becomes the baseline for the next cycle of improvement.

These leaders reject the logic of transformation—the idea that performance improves through occasional, heroic interventions. They don’t aim for one-shot breakthroughs. They build systems, habits, and norms that make improvement the standard business, practiced every day.

Toyota won’t use the word “solution.” Problems don’t get solved—they get managed with successive countermeasures and interventions, which are always open to refinement. The mindset is that any improvement can be made better. The danger of thinking in terms of transformation is that it assumes the job will one day be completed.

A deep belief these leaders share is that if you don’t commit to continuous improvement, you eventually set yourself up to need transformation. You get out of shape. And when you do, you can’t just go to the gym for a few weeks and expect to be fit. These companies work the improvement muscle every day. They stay in shape.

At Danaher, Culp embedded this ethic throughout the architecture of the Danaher Business System. Kaizen events, structured problem-solving, operational reviews, and leadership rotations became integral to the operating rhythm. Learning wasn’t an initiative. It was built into the way work got done. The result: a culture in which improvement isn’t episodic but is systematic—and self-sustaining.

Toyota exemplifies this approach with its practices of kaizen and hansei (Japanese for “self-reflection”). Leaders model both, surfacing small failures, asking probing questions, and using every issue to try to improve the system. Vulnerable reflection isn’t considered a weakness. It’s considered a strength that makes the company sharper over time.

At Amazon, Bezos built infrastructure that enabled constant learning—creating systems that were fast, cheap, and scalable. His mantra that “failure and invention are inseparable twins” was institutionalized in A/B testing protocols and real-time dashboards. The result: improvements that compound.

In these companies continuous improvement is not something to get through. It’s the way they work. And it starts at the top.

Why Is This Approach So Hard to Copy?

For decades, Toyota has welcomed outsiders into its factories to observe its production systems firsthand. Bezos has articulated Amazon’s management philosophy with striking transparency in his shareholder letters. Culp, after stepping down from Danaher and before becoming the CEO of General Electric, taught at Harvard Business School—sharing the leadership principles that underpinned Danaher’s performance. These leaders haven’t kept their methods secret; in fact, they’ve gone out of their way to make them visible.

And yet few companies have succeeded in replicating their approach. Why is that?

Part of the answer lies in how we continue to define the CEO’s role. In most organizations the CEO is still expected to articulate the vision, shape the strategy, allocate resources, and assemble a capable team. These responsibilities matter. But they often rest on an implicit assumption: that a CEO adds the most value by staying above the operational fray. Even when leaders see the merits of being more engaged, they are often constrained—by organizational norms, by board expectations, or by their own reluctance to let go of a familiar playbook.

Moving from this conventional model to one where the CEO is the chief architect and role model of the systems of execution requires more than a behavioral change. It requires a redefinition of leadership itself—one that challenges long-held assumptions. That kind of identity shift is difficult, especially for leaders who rose through the ranks by mastering the traditional model.

A second challenge is the rarity of a particular kind of discipline: the ability to move fluidly between altitude and detail. The leaders we studied know how to dive into specifics without getting lost in them. They know how to engage on the front lines without undercutting local initiative. They have trained themselves to see the link between seemingly small execution methods and large-scale outcomes. They are not dabblers. When they get into the details, they do so with intent—using their presence to model, teach, and reinforce the behaviors they want to spread. This kind of situational judgment—when to zoom in, when to step back, and how to do both without disrupting the rhythm of the organization—is not easily taught. And it cannot be faked.

But perhaps most fundamentally, this is not a leadership style defined by isolated behaviors that can be grafted onto an existing model. It is a system—a tightly woven set of practices, norms, mechanisms, and beliefs. The habits we’ve described—an obsession with customer-defined value, the rigorous use of data, distributed decision-making, structured experimentation and reflection, relentless improvement, and leading by teaching—are mutually reinforcing. They gain power not on their own but through integration. You cannot mandate six-page memos and expect to get Amazon’s innovation engine. You cannot simply borrow kaizen and hope to replicate Danaher’s operating consistency. The system matters—and the CEO’s role in designing, modeling, and reinforcing that system is essential.

From the outside these practices may appear simple. But sustaining them requires uncommon depth of care, of consistency, and of lived commitment. That is what sets these leaders apart—not just that they care about execution but that they embed that care into the very fabric of how their organizations operate.

. . .

The leader’s task is often framed as setting the vision and aligning people around it. The leaders of Amazon, Danaher, RELX, and Toyota perform that work differently. They spend an inordinate amount time alongside frontline workers, architecting the day-to-day methods of execution in ways that set the standard and teach others to do work well—even when the leader leaves the room. These leaders are just as invested in the how as they are in the what, and they’ve built systems and cultures that drive the entire organization to adopt that mindset. The superior performance that has resulted from this way of leading is lasting proof of its power.

Copyright 2025 Harvard Business School Publishing Corporation. Distributed by The New York Times Syndicate.

Explore AAPL Membership benefits.