I learned leadership mainly at Mars and LEGO, from managers who considered leadership a craft with rules, not just a lucky character trait you either have or don't.
I learned leadership mainly at Mars and LEGO, from managers who considered it a craft with rules, not a lucky character trait you either have or don't.
And almost every manager I've met since has underestimated how much conscious work goes into a good team. Usually the bet is on the power of the imperative: formulate clearly, repeat often, expect results.
Unfortunately, it doesn't work that simply. A dream team doesn't assemble itself. And even if you already have a team of superstars, it requires even more attention and work. Thankless, invisible work, and precisely for that reason most often skipped. This whole text is about that.
What follows are 5 situations I observed firsthand in different companies around the world. Each is a fact of managerial negligence, and I'm sharing them as a reminder to everyone trying to make a business efficient or scale: the main component of that success is the team that moves that very business. So, further in the article I'll talk about what markers show what kind of manager you are right now and write a checklist for building a dream team. No lead magnets and no "write I love lead magnets"
If you have direct reports? - read the markers section slowly and carefully. If you're just about to start managing - save it for later.
In Brief (as always)
• The five cases below go in sequence from a manager who confused their role, did the team's work over the weekend and then blamed the team for it; a CEO who asked a newcomer to publicly pass judgment on their colleague; a culture that asks for loyalty instead of competence; a fake CFO who stole 5 million dollars from a Fortune 500 company in a few months.
• Toxic culture is 10 times stronger than pay in predicting turnover. And TOP companies by employee engagement get +23% to profitability.
• 82% of people entering management received no formal leadership training. Gallup says 70% of engagement problems between departments fall on the manager.
• Fraudsters pass all HR checks 88% of the time, posing as specialists they are not.
FIVE CASES: WHAT BREAKS A TEAM
For safety: everything is anonymized and generalized. Not one case describes a specific company, but someone will recognize themselves in each. They go in ascending order of scale. The first costs one weekend. The last costs five million dollars.
What happened? At one company, a manager opened Monday's team meeting with an announcement: "Over the weekend I conducted thorough research." He went through his findings. It would seem you could then discuss conclusions and distribute new tasks. But he ended differently: actually, the team should have done this work. And if he has to keep doing their work on weekends, people face dismissal. At first glance it sounds fair. Should a manager do subordinates' work on their weekends? No.
Only the problem was different: the team didn't know this research needed to be done at all. They weren't given the task, weren't given a deadline, and weren't offered to participate. First people were deprived of the opportunity to do the work, then blamed for it not being done.
The research itself, as is now common, the manager assembled with AI help, so without business context and practical verification its value had yet to be proven.
What does this mean? In that hour the team learned one thing, and it wasn't the research. You won't be told about a task, you won't be offered a task, you won't be given the opportunity to participate, and you'll be the one to blame. The safest move from now on is to think not about work and projects, but about your alibi for the next meeting. The manager can do the team's work himself and will continue to do his team's work, but this work is exactly the opposite of his position. Managing a team is much more complex and important than weekend research and it's hard to offload to an AI agent. Not surprisingly, turnover in this team is 100% per year - a policy costing the business millions of dollars annually, and tens of times more than that manager's salary.
What do studies say? Google in Project Oxygen, a multi-year study of what distinguishes its best managers from the rest, puts "gives the team freedom and doesn't micromanage" in second place out of ten characteristics, above technical expertise (Google re:Work). Now the cost of turnover. Gallup estimates replacing one person at half to two times their annual salary, meaning a team that completely renews in a year pays its entire payroll a second time, only now for replacing people (Gallup, 2019), and the SHRM 2026 benchmark across 4600 organizations gives a median 39 calendar days to close a non-management position, so each spot sits empty more than a month per cycle (SHRM Recruiting Benchmarking, 2026).
Case 2. Leadership Issued as a Title
What happened? At another company, the team's strongest performer was put in charge of it. No one taught him: the promotion was the reward, and training would hint the reward was incomplete. His working management model was the only one he'd observed up close: take the plan, push it down, ask for the number, demand the number. Within a year half the team left, and the other half burned out. A good performer suddenly turned into a nightmare manager: orders instead of vision, goals and support, absence of development and discussion of career prospects, and those who stayed quietly "quit" leaving only the minimum tasks so they wouldn't be fired.
What does this mean? Management here was perceived as a position, not as a separate profession with its own skills: leading a team, reading motivation, understanding what a group of people does under pressure. Some organizations spend serious money precisely on this - multi-day programs, coaching, mentoring, turning the best specialists into trained managers. Others issue a title and a set of imperatives, then each time encounter the same three problems: turnover, low motivation, and quiet quitting, when a person continues coming to work as if nothing connects them to the organization anymore.
What do studies say? Chartered Management Institute together with YouGov surveyed more than 4500 British employees and managers in 2023 and found that 82% of people entering management received no formal management training, and every third employee left work because of toxic culture (CMI, October 2023). Money, when spent, works: Lacerenza with colleagues combined 335 samples totaling 26,573 people and showed that leadership training noticeably improves both skill application at work and organizational results, with multi-session programs with intervals working noticeably better than one-time workshops (Lacerenza et al., *Journal of Applied Psychology*, 2017). And quiet quitting has a name in research too: Gallup defines disengaged employees as quietly quitting, psychologically attached neither to work nor to company, and says global engagement is at 20% (Gallup, State of the Global Workplace 2026).
Case 3. The Question with Lowered Thumb
What happened? A newly appointed manager sits in one of his first management team meetings. In front of the whole team the CEO asks him what he would do with one of the country managers, whose team several minutes earlier presented results, and the results were unsatisfactory. Would you fire them? What would you do in my place? There's no good answer, and the answer was never the goal. The goal was demonstration, an immersive performance: "judgment on an equal, delivered publicly and on demand." And everyone in the room calculated the consequences in real time. The question about the person next to me will one day turn into a question about me, and they'll call on those sitting nearby to answer it.
What does this mean? Such a company starts paying for loyalty instead of competence. The problem is that loyalty itself doesn't convert into professional growth, strong results, or profit. The most capable notice this first, because they're usually the ones who have somewhere to go.
If you need diagnostics faster than a full engagement survey, there are three questions from Jake Breeden that take a minute: - What happens when someone disagrees? - Can you explain your work to an outsider without internal jargon? - If you leave tomorrow, what exactly will you lose?
The first question is especially telling. Culture responds to disagreement with conversation and problem-solving. A cult responds with consequences.
What do studies say? Google in Project Aristotle across 180 teams put psychological safety, confidence that you won't be punished or put in an awkward position for speaking up, in first place among five effectiveness factors, above reliability, structure, meaning, and environmental impact (Google re:Work).
Case 4. Culture That Asks for Loyalty Instead of Competence
What happened? I've observed this case in so many teams and organizations it would be hard to count and it always unfolds in the same five steps. The manager becomes unchallengeable, and disagreement is read as disloyalty, not as information. The word "team" surfaces exactly in moments when the team has nothing to do with it: salary, hours, personal boundaries, requests for unpaid additional contribution. Curiosity is considered a breach of discipline. Self-immolation is praised as good practice. "Cult of submission" - what up to 85% of people who've ever worked as employees have gone through.
What does this mean? Such a company pays with loyalty instead of competence, and loyalty doesn't convert into good experience, career success, profit. The most capable notice this first, because they have somewhere to go. If you need diagnostics faster than an engagement survey, here are three questions from Jake Breeden that take a minute: what happens when someone disagrees?; can you explain your work to an outsider without internal jargon?; and if you leave tomorrow, what exactly will you lose? To the first question culture responds with conversation and problem-solving, and a cult responds with consequences. (Psychology Today, 2026).
What do studies say? The cost is measurable. Analyzing 34 million employee profiles, Donald Sull, Charles Sull and Ben Zweig found that toxic culture is ten times stronger than pay in predicting turnover (MIT Sloan Management Review, 2022), and in a related study named its five attributes: disrespect, exclusionary behavior, unethical conduct, cutthroat competition, abuse (MIT SMR, March 2022).
CASE 5. THE CFO WHOSE MUSTACHE FELL OFF
What happened? A new CFO joins the team and something doesn't add up right away. His psychological profile doesn't read as someone who spent years inside spreadsheets, pivots and reconciliation scripts. He gives the impression of a hardened salesman who sold everything, even semi-legal goods to semi-legal buyers. But the HR department conducted a series of interviews, security check, so the only evidence was the feeling that something was wrong. Several months later it turned out he forged everything and passed all filters into a Fortune 500 company. He performed normal role functions, gradually transferring client payments to shell accounts. And when the fraud was uncovered - he disappeared leaving a five million dollar hole.
What does this mean? The loud version of company failure, fraud, corruption, money leakage. The quiet version can look much more mundane: the company can't close a key position for months, hires the wrong person, loses them after a few months and launches the process again. The cost of such a mistake is rarely visible in one financial indicator. It's distributed among recruitment costs, manager time, lost productivity, team burden, training, turnover and re-hiring.
Studies show that the cost of a bad hire can be a significant portion of the employee's annual salary, and the cost of replacing an employee can reach 0.5-2 annual salaries. Meanwhile, according to SHRM, only 20% of organizations systematically track Quality of Hire.
And here an interesting paradox arises. Business has learned quite well to count cost of hire, but much worse at counting cost of wrong hire. And even worse at understanding why this hire was wrong: made a mistake in selection, poorly defined the role, or simply abandoned the person after their first day.
What do studies say? Robert Half research. It showed that a bad hire costs the company approximately 15-21% of annual salary, depending on position level. Meanwhile consequences include not only salary, but also lost productivity, time for re-hiring and training, decreased team morale and additional burden on managers, but don't include the repeating scenario. Robert Half - The rising cost of a bad hire Gallup notes significant turnover among new employees. Their research mentions an estimate that a third of new employees may not even make it to 90 days, and a significant portion of new employees leave the company within the first year. (https://www.gallup.com/workplace/246242/essential-ingredients-effective-onboarding-program.aspx?utm_source=chatgpt.com)
BAD ONBOARDING TURNS A HIRING MISTAKE INTO A REPEATING CYCLE OF COSTS.
Does your leadership demands blind obedience.jpg
SO WHY DOESN'T A DREAM TEAM ASSEMBLE ITSELF?
At first glance these are five completely different stories. Micromanagement. Untrained manager. Public humiliation. Culture of submission. Hiring failure. But they have one mechanism: the company left team quality to chance.
Somewhere no one defined management rules. Somewhere they didn't teach a person to manage. Somewhere they didn't create a safe environment. Somewhere they didn't build proper hiring and control. Results are later called different things: turnover, low engagement, bad hire, burnout, lost productivity, sometimes five million dollars.
So a dream team is a two-way story. The team should be a dream for the manager, and the manager, for the team. And neither arises on its own.
The main lever here is the manager. Not because they're personally to blame for everything happening, but because they have the most tools to change the system. And from here follows the next question: what specifically should they do?
You can start with a rather boring thing that's almost 100 years old - the management cycle.
Five steps you need to do and repeat
1. Plan. What is your team's goal and purpose, how do they break down into separate functions and tasks, what specialist profile can close them. Write roles and responsibilities, a 6-12 month plan, job posting, interview questions and onboarding plan for at least the first 3 months. For an already filled position you can skip hiring documents, but not planning itself, even if the position is filled.
2. Organize. Make sure people have all resources and tools to do the work well, and that hygiene factors are covered. Break big work into tasks with goals and KPIs and give each task to whoever will handle it best.
3. Teach. Make sure the team knows how to do the work. A good manager explains clearly so no one feels lost, teaches new skills and patiently answers questions.
4. Control. Control is checking progress so everything goes according to plan. It's not about being bossy, but about safety and alignment: track deadlines and budget, fix mistakes early, check that work meets quality standards.
5. Motivate. Keep the team engaged and happy about good work. People work better when they feel valued: praise the team for successes, listen to their feelings and ideas, remind everyone why their work matters.
Apply the right tools to the right level. You need to approach a specialist and a senior manager differently, and the exact difference is in the table below.
How much does absence of a dream team cost?
Here's a simple model for one strong specialist with salary A$120,000, my assumptions marked "[estimate]".
That's why it's hard to notice in time. Disengagement never announces itself. First to go is what the person did beyond duties: an idea they didn't voice, a problem they noticed and didn't highlight, Sunday thoughts because they cared. By the time of resignation these losses have been going for six months to a year and a half.
Am I a good manager? Markers that answer this
Self-assessment in leadership is almost useless: the question "am I a good manager" is answered by intention, not facts. So here are observable signs. Without adjectives and without needing to assess your own empathy.
Good markers
1. People bring you problems while they're small, including ones that make them look bad.
2. Someone on the team publicly disagreed with you in the last month, and nothing happened to them.
3. Reversible decisions within their zone are made without you, and you learn about them from an update.
4. You can name what specifically each subordinate is better at than you, without looking at notes.
5. The last three priority changes you announced with explicit cancellation of the previous one.
6. They show you draft ideas while no one has had to defend them yet.
7. You know your team's engagement score, not just the company average.
8. When a person leaves the team for a promotion, they recommend you to whoever comes to their place.
Bad markers
1. You learn about problems approximately simultaneously with the client.
2. Two of your instructions live simultaneously, and no one told you which one to follow.
3. You did a subordinate's work yourself "because it's faster" more than once per quarter.
4. In meetings with strong people you ask questions whose answers you already know.
5. Your one-on-ones are status updates with friendlier intonation.
6. No one has argued with you in a meeting for a month, and you took it as agreement.
7. You can't remember when a decision in your team last got a date.
8. The last resignation was a surprise to you.
Count honestly. Six or more good markers - the system works. Three or more bad ones - you have a specific fixable problem, not a character defect. Each item on this list has a pair in the checklist below.
My checklist for building a dream team
A checklist is only useful if each line can be marked "done" or "not done" by someone other than you. So each item below has an action, an artifact that remains after it, a test for completion, and a case or study it grew from. Thirteen items, in the order I would do them. Do just the first five - you'll already be ahead of every manager from the cases above.
In the first week with the team
1. Start with yourself: confidential feedback from the team. Before evaluating anyone, let them evaluate you. Take the 13 published Google items or your own, and completely remove yourself from collection: let the team itself choose a person they trust to compile answers, or run them through an AI interface that removes names and rewrites formulations before you see a word. Only if at least three people report to you; below that anonymity is fiction. Artifact: scores for each item and two open responses verbatim. Test: within 30 days you tell the team two things you'll change, and one you won't change, with reason. Source: Google's semi-annual manager survey; FAQ above.
2. Know the team as people, not as functions. One page per person, in your hand, with four fields: what they're strong at in their own words; what's important to them outside work - interests, family, habits, the hour you can never schedule a meeting; where they want to be in 24 months; and what they stopped raising. Artifact: registry, corrected by each in their own one-on-one. Test: about each you name without notes one thing outside work that's important to them, and their 24-month goal. Source: Gallup Q12 fifth question, "my supervisor or someone at work seems to care about me as a person"; Case 2.
3. Conduct engagement review and agree with team on six-month plan. Take the team engagement measurement result, twelve questions or your short pulse, sit with the team and agree on a list of items you'll work on the next six months. Responsibility for engagement is shared within the team, and it's divided 60-40, where the bigger piece of pie is always the manager's. Artifact: document with measurement result and six-month plan, each item has owner and date. Test: everyone on the team can name items they're responsible for, and the plan is reviewed at one regular meeting. Source: Gallup Q12; conclusion about variance at manager level; Case 2.
4. Fix the starting point before changing anything. Three numbers: your team's unwanted turnover for 12 months versus company figure; last team engagement score versus company average; portion of team that without prompting names the quarter's three main priorities. Artifact: one line, three numbers, date. Test: you name all three from memory. Source: Gallup estimate that manager accounts for no less than 70% of engagement variance; Watkins on diagnosis before action.
5. Ask about stuck topics and close one. Ask each: what did you stop raising because nothing happened last time? Write down each answer with owner and date. Artifact: list of stuck topics. Test: at least one topic resolved in 30 days, and this is communicated to whoever raised it. Source: Case 3; Project Aristotle on psychological safety.
Every week
6. Praise each person, specifically, every week. No "good job" and "great work." Your job is to find what to praise, and describe it precisely: what the person did, what it changed, why it was harder than it looked. If this week you can't praise someone for anything - that's information about your attention, not their work. Artifact: one written praise per person per week, where they can save it. Test: each names specific action and its effect. Source: Gallup Q12 fourth question, "in the last seven days I have been recognized or praised for good work" - that's why the rhythm is weekly, not monthly.
7. Brief before doing. Any work you're going to do inside someone else's zone gets a written note by Thursday: what you're looking at, why, and what part is theirs. Artifact: note sent before starting work. Test: zero finished artifacts shown to team without warning, for a month. Source: Case 1; Project Oxygen second behavior, don't micromanage.
8. Conduct one-on-ones with three fixed questions. Thirty minutes, their agenda, three questions you always ask: what's stuck, what do you want next, what should I stop doing. Status - in shared document, not in meeting. Artifact: current notes with "stop doing" column. Test: one "stop doing" answer per month that you actually stopped doing. Source: Google manager survey; Gallup Q12 eleventh question about conversation about progress.
Every quarter
9. Make the career deal explicit. Return to each person's two-year plan and write it as a deal in two columns: what you'll do so the person gets there, and what you need from them and from the team in return. Say the second column out loud; people accept deals much more readily than favors. Artifact: two-column plan, in registry, with date. Test: each names their next step and what they owe the team for it. Source: Gallup Q12 twelfth question, opportunities to learn and grow; Google survey item about career conversation in six months; Case 2.
10. Let them try your work. Everyone who wants your position gets a real piece of it: "Go to the meeting with my leadership instead of me. I'll help prepare." Budget defense, contractor negotiations, slide for board of directors. You prepare the person, don't enter the room, debrief after. Artifact: calendar invitation with their name, not yours. Test: once per quarter for each who said they want to grow, and you weren't in the room. Source: Project Oxygen first and sixth behaviors, coaching and supporting career development; Lacerenza that practice on real task gives more than classroom training.
11. Measure engagement pulse and compare with company. Twelve questions or short pulse, by team, versus company average and last quarter. Artifact: one chart, two lines. Test: results and written "what's changing" reach team in 30 days. Source: Gallup Q12 meta-analysis; conclusion about variance at manager level.
12. Check six-month plan at midpoint. Go through each item with team: done, on track, stuck. Reassign stuck items, remove what stopped being important, record changes. Artifact: plan with status for each item, with date. Test: not one item stuck without named owner and new date, and team engagement score moved toward plan. Source: item 3 above; Gallup Q12 meta-analysis.
13. Check hiring kit before opening any role. Seven documents exist before posting vacancy: position description, vacancy text, roles and responsibilities, offer template, onboarding plan, probation review and written identity and document verification step. Artifact: kit folder. Test: not one role opens without all seven; every offer has verification line signed by someone other than hiring manager. Source: Case 5; HireRight 2025; ACFE 2026.
Where we stand on this
Most owners we talk to can fairly quickly name an employee who is "supposedly strong, but no results." And then the most interesting begins. In five cases out of six the result assessment is held mainly on the manager's opinion. And in four of them there's nothing you can look at from the side and say: here are expectations, here are facts, here's the result. So our position is quite simple. In a company up to 200 people you don't necessarily need to start with a big transformation, new system or another hire. Often much more useful are three rather boring things: proper engagement review with six-month action plan, unified hiring document kit and short quarterly employee survey. This doesn't look like big strategy. But it quickly shows where the company is actually losing people and money. Who doesn't understand what's expected of them. Where the manager considers the employee weak, but the problem isn't in the employee at all. Why a good candidate doesn't reach the offer. Why a new person comes to work and the first two months just tries to understand what's happening here. Yes, process won't save every unsuccessful hire. And a person who looks great in interview sometimes turns out completely different at work. But that's exactly why you need not feelings, but data. A good HR system doesn't promise there won't be mistakes. It makes the company notice them earlier, understand causes and not repeat the same thing again.
What will change in 30 days, a quarter and a year
30 days
First engagement review exists, six-month improvement plan is clearly defined and going on schedule.
Hiring kit for next open role is written before search starts: position description, roles and responsibilities, offer template, onboarding plan, probation review.
Contradicting instructions surface within a week, because the old one lies on the page next to the new one.
Quarter
Engagement plan is half complete, you can already check progress and see first fruits of your efforts.
Everyone on the team has their own career development plan, and the team clearly sees how their results will affect future career opportunities.
Team takes more responsibility and shows real interest in solving team tasks together.
Year
Your newborn dream team not only moves the business at least 30% more efficiently, it also talks about their workplace, so you start hiring stronger people, and they move the business even better.
Voluntary turnover among strong employees is a number to watch: it's falling, and the budget saves from 0.5 to 2 salaries on each departure, and that's where the money lies.
First quality promotions appear in the team: strong specialists don't leave the company but grow inside, and you save on the whole cycle. Matthew Bidwell from Wharton showed that external hires are paid 18-20% more than those promoted to the same position, and meanwhile the first two years they have lower performance ratings (Knowledge at Wharton). The position is filled by a strong specialist with zero lag for learning the business.
The uncomfortable part
In the order of items in this checklist I'm about 70% confident. Each separate item I'm ready to defend before anyone, and the sequence is a matter of judgment, and a person with a different team would sort differently.
Without any reservations I'm ready to defend something else, and it's what most managers resist: your team's quality is mostly a description of you. Gallup's 70% figure says this in the language of variance. Five cases say this in the language of Tuesday afternoon.
Reading this is unpleasant if the team isn't working. And this is the only version of the situation where you have any lever at all - which makes it useful.
What a good manager really is
All serious leadership research converges on four things, and charisma isn't among them. Integrity: your actions match your words so much you can be predicted. Development: you grow your people, not feel threatened by their growth. Empathy: you understand how a message will land before you send it. Accountability: you openly admit your own mistakes, because only that makes admitting others' mistakes safe.
A team doesn't become a dream team because it was told to. It becomes one because someone did forty boring minutes a week for two years and never once punished anyone demonstratively.
Frequently Asked Questions
How do I know if I'm a good manager?
Hire us. Team Audit & Capability Mapping conducts confidential engagement survey and bottom-up feedback for you and delivers a six-month plan; Team Mentoring Retainer keeps you on that plan.
Do it yourself. Three numbers that don't depend on your opinion: anonymous bottom-up feedback survey (13 Google items), team engagement score at team level (manager accounts for 70% of variance) and unwanted turnover versus company average.
What really makes a team high-performing?
Hire us. Process Design & SOP and Role Mapping & Job Design give structure; Team Development builds safety to use it.
Why do good employees leave companies where they're treated well?
Hire us. Team Audit & Capability Mapping finds who's about to leave and why, before resignation; Employee Onboarding Programme closes the leak at start.
Do it yourself. Talk: 52% of those who left believe departure could have been prevented, and with 51% no one talked in the last three months (Gallup, 2019); toxic culture predicts departure ten times stronger than pay (MIT SMR, 2022).
What to do in the first 90 days in a manager position?
Hire us. Team Audit & Capability Mapping gives starting point in first four weeks; Role Mapping & Job Design fixes structure before hiring.
Do it yourself. Twelve questions once a year, short pulse every quarter, both by manager, both with written response within 30 days. Survey without visible response teaches people to stop answering honestly.
Recognized three bad markers? This is a process problem with a known solution. Write to me; first conversation is free.
Comments
No comments yet. Be the first.