All In: Managing Professionals

The man with the stopwatch

Where our management instincts were born, and why they worked.

In 1899, Frederick Taylor stood over a man loading pig iron at Bethlehem Steel with a stopwatch and a clipboard. He timed every motion, removed the wasted ones, prescribed the exact rest intervals, and handed the worker a script. Then he did the same with shoveling, deciding that the one best shovel load was 21 pounds. Every time, the output climbed steeply.

This is the birth of modern management, and it genuinely worked. It worked for one reason: the task was simple, repeated, and fully visible to a stranger with a stopwatch. When all the value lives in observable motion, the boss really can know the one best way, and the worker's job is to follow it.

The worker who owns the factory

Why the stopwatch breaks the moment the value moves inside someone's head.

Now picture your best person. The thing that makes them valuable is not a motion you can time. It is a judgment that happens somewhere behind their eyes, and when they go home at night, they take the entire factory with them. Peter Drucker saw this coming decades ago, when he coined the term knowledge worker.

Knowledge workers own the means of production. It is the knowledge between their ears, and it is a totally portable and enormous capital asset.Peter Drucker

Two things follow, and Taylor's system survives neither. First, you usually cannot see the work, so you cannot prescribe the one best way. Second, past the apprentice stage, a good professional knows their job better than you do. The old deal, where the manager does the thinking and the worker does the executing, runs backward. So you stop managing professionals as subordinates and start treating them as partners, even as volunteers. And partners cannot be ordered. They have to be persuaded.

Standard work still has a job. In a warehouse, a lockout procedure, hazardous-material rule, or receiving scan may need one safe method every time. Standardize the safety boundary and the repeated handoff. Leave room inside it for people to diagnose, prioritize, and improve. Autonomy without a guardrail is negligence. A guardrail around every judgment is bureaucracy.

The boss who gets what he expects

Your private theory of people is not private. It comes true.

In 1960, Douglas McGregor noticed that every manager quietly runs on one of two assumptions. Theory X says people dislike work and need to be pushed, watched, and controlled. Theory Y says effort is as natural as rest, and people will direct themselves toward goals they actually believe in. McGregor's unsettling point was that both are self-fulfilling. Watch people like they cannot be trusted, and they stop showing you the judgment you are not looking for. You manufacture the very passivity you then cite as proof you were right.

With laborers on a line, a Theory X system can still produce tons. With professionals, it is corrosive, because the thing you are suppressing, their initiative, is the entire reason you hired them.

The gap

There is a ceiling on every professional that no authority can reach. Only you can.

For a person on Taylor's line, the distance between the least and the most they can give is small, bounded by physical pace. For a professional it is vast. A disengaged engineer ships code that works. An engaged one solves the problem you did not know you had. That distance is discretionary effort, and whether it gets spent is what we call engagement.

And the single biggest factor deciding whether it gets volunteered is not pay, perks, or the company logo. It is the person they report to.

Engagement is a condition, not the score. A committed team can still pursue the wrong priorities, miss the quality bar, or lose the customer. Management also chooses direction, sets standards, makes tradeoffs, and acts when results fall short. The point of earning discretionary effort is to spend it on work that matters.

Two ledgers

Why a raise buys silence, not fire.

In his study of motivation at work, Frederick Herzberg found that what made people love their work and what made them hate it were not opposites. They were different ledgers entirely. Hygiene factors, like salary, policy, conditions, and a decent boss, only prevent dissatisfaction when they are right. Get them wrong and people revolt. Get them right and people feel, at best, nothing. Motivators, like achievement, the work itself, responsibility, recognition, and growth, are the only things that actually light people up.

Herzberg watched this run as an experiment at AT&T. A group of clerks who answered letters from shareholders were handed real ownership of the work: the authority to sign their own names, and personal accountability for getting the answer right. Their pay did not change. The reported service index climbed from the 20s into the 90s within months. The work itself, made whole, did the motivating.

I can charge a man's battery, and recharge it, and recharge it again. But it is only when he has his own generator that we can talk about motivation.Frederick Herzberg

Autonomy, mastery, purpose

For thinking work, the carrot can make performance worse. This is not opinion. It is repeatable.

In 1962, Sam Glucksberg gave people a tricky puzzle and offered one group a cash prize for speed. The prize made them slower. The paid group did worse, and it is one of the most counterintuitive findings in the field. A meta-analysis of 128 studies later found that expected tangible rewards tend to weaken intrinsic motivation, especially when the work is already interesting.

Daniel Pink gathered this into three drivers that actually move professionals: autonomy (control over their work), mastery (getting visibly better at something that matters), and purpose (a reason beyond the paycheck). The carrot-and-stick was built for routine work. On thinking work it backfires. Try it:

The upside of the opposite approach is easy to find in the wild. 3M has long encouraged employees to spend around 15 percent of their time exploring their own ideas; that culture helped produce the Post-it Note. Google adopted "20 percent time," and its 2004 filing named Google News and AdSense among the products prototyped through independent projects. Nobody ordered those things built. People were trusted with their own attention, and they over-delivered.

Small wins

The most powerful daily motivator is also the easiest one to step on.

Teresa Amabile collected nearly 12,000 daily diary entries from professionals and asked a simple question: on the days people felt most motivated, what had actually happened? The answer was not a bonus or a pep talk. It was progress in meaningful work. The single most powerful thing inside a good workday is the feeling of moving something forward. The corollary is brutal: small setbacks and petty interference hurt more than small wins help.

Dan Ariely showed the dark edge of this. When people's finished work was ignored, not shredded, just ignored and set aside unread, they gave up almost as fast as people whose work was destroyed in front of them. To a professional, being unseen is nearly as demotivating as being torn down.

Match your grip to the person

The same management that grows one person smothers another. The variable is the task, not the title.

Andy Grove, who built Intel, taught that there is no single correct management style, only a correct fit for a person's maturity on a specific task. He called it task-relevant maturity. New to a task, a person needs clear direction. As they grow on it, they need coaching and support. Once they own it, they need you to get out of the way. The catch that trips up every new manager: maturity does not transfer. Promote your best engineer into management and they are a beginner again, even though they were an expert an hour ago.

Grove also named the room where all of this gets read: the one-on-one, a standing meeting that belongs to the report, not to you. Their agenda, your questions. It is where you learn which stage someone is at on this month's work, where problems surface while they are still small, and where finished work gets seen. His arithmetic for it was blunt:

Ninety minutes of your time can enhance the quality of your subordinate's work for two weeks, or for some eighty-plus hours.Andy Grove, High Output Management

No other meeting on your calendar pays anything close to that, which makes the weekly one-on-one the last meeting you should cancel, not the first. Skipping it never reads as "we are busy." It reads as "you are the skippable part of my week."

Ask, do not tell

When Google went looking for what makes a great manager, the answer surprised even Google.

Google's engineers suspected managers were overhead. So Google studied it, hard, across thousands of data points, expecting to find that technical brilliance was what mattered. Technical skill came in last. The number one behavior of their best managers was the same one Bill Campbell used to coach the founders of Google, Apple, and Intuit: being a good coach. Not having the answers. Drawing them out. From 2010 through 2012, Google reported that median manager-favorability scores climbed from 83 percent to 88 percent, with the lowest-scoring managers improving the most.

Michael Bungay Stanier calls the instinct to jump in with answers the "advice monster," and it has a hidden cost: every time you solve it for them, they bring you the next one, and they never build the muscle. The repair is almost embarrassingly simple. Open the conversation with "What's on your mind?", then resist filling the silence. Ask "And what else?" The real issue is rarely the first one named.

Coaching is no longer a specialty. You cannot be a good manager without being a good coach.The leadership handbook of Bill Campbell

Care personally, challenge directly

The kindest thing you can do for a professional is also the hardest: tell them the truth.

Kim Scott maps feedback on two axes: how much you care about the person, and how directly you challenge them. Get both high and you have what she calls radical candor. The failure she sees most often is not cruelty. It is the opposite: ruinous empathy, where you care so much about the person's feelings in the moment that you soften the message until it carries no information, and they walk into a wall you saw coming. To a professional, that is not kindness. It is being denied the one thing they need to get better.

And when you do praise, be specific, because there is a deeper truth in the research, the one Marcus Buckingham and Ashley Goodall call the feedback fallacy: people grow far more from understanding what they did well than from a list of what they did poorly. Excellence is idiosyncratic. It shows up differently in every person, so your job is to notice what each one does especially well, and tell them exactly what you saw.

Multiply, do not diminish

Two leaders, the same team, double the output. The difference is who has to be the genius.

Liz Wiseman studied leaders and found two types. Diminishers are the smartest person in the room, and they prove it, so the room goes quiet. Multipliers treat their people as smart and get more out of them by asking hard questions and handing over real ownership. In Wiseman's study, team members reported giving almost twice as much of their capability to Multipliers as they gave to Diminishers. The most humbling part of her work: most diminishing is accidental, done by well-meaning managers who jump in to rescue, set a blistering pace, or fire off ideas faster than anyone can act on them.

It isn't how much you know that matters. What matters is how much access you have to what other people know.Liz Wiseman

Make it safe to say the hard thing

The best teams do not make fewer mistakes. They are simply not afraid to tell you about them.

Amy Edmondson went into hospitals expecting the best nursing teams to report the fewest medication errors. She found the opposite, and it puzzled her until she saw what was really going on. The best teams reported more errors not because they made more, but because they felt safe enough to admit them. On the weak teams, people quietly hid their mistakes, so nobody could fix them or learn from them. The point is counterintuitive: a team that surfaces lots of problems is usually the healthy one, and a silent team is the worrying one. That sense of safety is what she named psychological safety, the shared belief that you can speak up, with a worry, a mistake, or a half-formed idea, without being punished or humiliated. Years later, Google's team-effectiveness study also put psychological safety first.

For professionals, this is not a nice-to-have. Their real work is thinking, and good thinking needs people who will disagree out loud, float an idea before it is polished, and admit when they are not sure. None of that happens on a team where speaking up feels risky. A team that cannot say "I think this is wrong" cannot do its best work.

Give context, not control

The more you have to approve, the less they can think. Trade approvals for understanding.

The instinct of an anxious manager is to add a checkpoint: route the decision up, require a sign-off, keep control. Netflix built one of the most productive cultures on earth on the opposite reflex, which they call context, not control. Instead of approving a professional's decisions, you flood them with the context, the strategy, the constraints, the why, until they can make the call you would have made, without asking. Every approval you remove is a vote of confidence the professional can feel. Every one you add says, quietly, "I do not trust your judgment."

The worry, of course, is that all this freedom turns into a free-for-all. But the chaos managers dread does not come from people having room to decide. It comes from nobody being sure who decides. So taking the approval gates away is only half the job. The other half is clarity: each person needs to know exactly which calls are theirs to own. That is not a new leash. It is the opposite, the thing that makes the freedom safe, because you are handing people their own judgment and cutting the hesitation of "wait, is this mine to call?" Remove the approvals, name the owners, and you have done one coherent thing, not two opposite ones.

If you dummy-proof the system, only dummies will want to work there.Reed Hastings, Netflix

Specific and hard

Autonomy says the how is yours. A real goal says what winning means. Freedom in the how only works next to a bar you can score.

In the mid-1970s a logging company was quietly wasting a fortune on its own trucks: drivers hauled timber out of the woods loaded to about 60 percent of the legal weight limit, trip after trip, because the standing instruction was the one every professional still hears, do your best. With the drivers and their union on board, the company replaced it with one specific, difficult goal: load to 94 percent. Within a few months the fleet was running near 90 percent, and it stayed there. No bonus, no training, no new equipment. The company's cost accounting priced the change at a quarter of a million dollars in trucks it no longer had to buy.

Gary Latham ran that study, and with Edwin Locke he spent the next three decades replicating the pattern across hundreds of studies and tens of thousands of people, until it stood as one of the most replicated findings in the whole field: specific, difficult goals beat vague or easy ones on almost any work you can measure. "Do your best" fails for a humbling reason. It lets everyone define best downward without ever noticing, while a concrete, hard target turns the work into a game you can tell you are winning, which the small-wins research above says is exactly the fuel professionals run on.

Two lines of fine print. A goal narrows attention, the same force the cash bonus applied in the lab above, so hang hard numbers on execution work and give creative work learning goals (surface three viable approaches) rather than output quotas. And a number handed down without context is not a goal, it is a quota; set it with the person, and connect it to why it matters. Do that and this section snaps together with the last one: a clear bar, a named owner, and freedom in between.

The proof, at scale

This is not a theory that only survives in a lab. Whole companies are built on it.

If trusting professionals were naive, the companies that lean hardest into it would be the ones bleeding out. Instead, they include some of the most valuable companies on earth and the quiet giants of their industries. Four of them:

One honest caveat. Copying the trappings does not work. Plenty of companies have borrowed the open offices and the flat titles without the underlying trust, and ended up with the vocabulary and none of the results. Autonomy has to be real authority, not theater, and professionals can tell the difference in a week.

No seat fillers

You can do every one of these things right and still get stuck with someone who won't rise. The failure isn't them. It's the seat you let them fill forever.

There's an old reading of the parable of the sower: the story is less about how well you scatter the seed than about how fertile the soil is. Even in a room full of professionals, some of the ground is rock. You will manage people who coast, who quietly settle, who fill a seat without ever filling the role. And the costliest one is rarely the obvious flop. It's the capable person who stopped growing years ago and now collects a check while the team carries them. Your best people notice that long before you do, and they draw the obvious conclusion: if the standard is optional for the coaster, it's optional. The message lands on exactly the people you most need to keep.

Start, though, with the humbling half. A manager's ceiling is partly set by the soil: who is in the seat, and whether it is the right one. People do not change all that much once grown; you draw out what is in them, you cannot pour in what is not. So before you decide someone is the wrong person, ask the cheaper question first, are they simply in the wrong seat? A struggling performer is very often a strong one aimed at the wrong work.

The sharpest proof is a car factory. In 1982 General Motors shut its Fremont plant after absenteeism had regularly exceeded 20 percent. Two years later Toyota reopened the same building and rehired about 85 percent of the former workforce. The plant's performance changed with the system around the same people. W. Edwards Deming, whose thinking shaped that system, put it in ten words: "A bad system will beat a good person every time." What looks like a bad person is often just a bad fit.

That middle question is the conscience of the entire thing, because managers manufacture their own poor performers more often than they will admit. Doubt a person and you start watching them closer, handing them safer work, trusting them less, and they feel every bit of it and shrink to fit the box you built. The same force runs the other way too: when a manager genuinely expects more, people reliably rise to it. One insurance branch once lifted a group of strictly average agents to top performance for no reason other than a manager who refused to believe they couldn't. So the honest test before you write anyone off is not "did this person perform poorly?" It is "did I give them the conditions to perform at all?"

When you have asked all of that in good faith and the answer is still no, then comes the real discipline, and it is simpler than it sounds. There are only two acceptable jobs: make the person genuinely good, or help them leave well. The cleanest test in the business is Netflix's keeper test, if this person told you they were leaving for a similar job elsewhere, would you fight to keep them? If the honest answer is no, you already know, and every month you wait is a month you have chosen the comfortable lie over the kind truth. Letting someone go from a job they are failing is not the cruelty. Leaving them stranded in it is.

There is a brutal counterfeit of this idea, though, and it is worth naming so you never reach for it. The version where you rank everyone and cut a fixed group every year, on a schedule, is sorting dressed up as rigor. Microsoft eventually abandoned stack ranking after years of internal competition, and General Electric moved away from the forced curve it had helped popularize. Take the candor, not the curve. This is one honest judgment about one real person, never a quota you fill.

And the hardest seat-filler of all to remove is the one posting the best numbers. A brilliant jerk is not dead weight in any obvious way; they are filling their seat with output. But the output is a trick of the eye. Men's Wearhouse once fired its single best salesman, a man who hit every target and poisoned everyone near him, hoarding customers and helping no one. Afterward, no remaining person matched his individual sales, and the store's total sales went up. His good numbers had been hiding the damage he did to everyone else's. In a large study of hourly workers, avoiding one toxic worker was worth more than twice as much as hiring a superstar, because the star lifts their own work while the toxic worker drags down everybody's.

It comes back to the bargain this whole essay rests on. A professional's side of it is judgment: you hand over the decisions, and they bring the thinking to make them well. Someone who refuses, who won't own a call or risk an opinion and only wants to be told exactly what to do, has quietly taken the laborer's deal instead. Fine. Give it to them honestly: a clear bar, close direction, and no mystery about what good looks like. Some people wake up the moment the deal is said out loud. And the ones who don't? They have just made your decision for you.

Your unit of work becomes the system

The first promotion changes whose work you do. The next changes how far your decisions travel.

A new manager gets results through other people. A director or manager of managers gets results through several teams, the managers who lead them, and peers they cannot command. You can no longer inspect every decision or coach every person. If you try, work queues behind you and your managers learn to wait. The habit that made you a helpful manager turns you into the constraint.

Managing through managers starts with restraint. Give each manager an outcome, the decisions that are theirs, the boundaries that are fixed, and the resources they can move. Review how their team operates and how they are developing people. Do not quietly become the real manager of their reports. A skip-level conversation should give you signal, not become a second chain of command.

Strategy becomes concrete at this layer. A company goal usually arrives too broad to run: improve retention, ship faster, grow efficiently. Your job is to turn it into choices. Which customers matter most? Which work stops? Where will quality beat speed? What risk will you accept? A priority that costs nothing else is only a wish.

Most director-level results cross an org chart. Product needs Sales. Customer Success needs Support. Operations needs Finance. Authority will not carry you across those lines, so the work becomes influence: bring the evidence, name the shared outcome, make the tradeoff visible, and put one owner and one date on the decision. Escalation is useful after the people closest to the work have made the disagreement clear. Before that, it is usually borrowed authority covering for an unfinished conversation.

Resources are strategy with names attached. Every new priority consumes a person's attention, a budget, or both. Rebalancing those resources is part of the job, including the uncomfortable act of stopping good work so the most important work can finish. The same is true of communication. Tell executives what is at risk before the miss. Tell teams what changed, why it changed, and who now decides. Repeating a clear message is cheaper than letting five groups invent their own version.

Succession belongs inside delivery. If one manager, architect, or operator cannot leave for a week without the system bending around the absence, you have found a leadership risk. Build judgment in more than one person. Give the next leader real decisions while the current one is still there to coach them. Senior leadership shows up in the number of good decisions the system can make without waiting for you.