Daniel sent us this one — he's asking about flat hierarchy, not just as a business buzzword but as something that shapes families, social groups, everything. He points to Israel's culture of flat hierarchy in the military and tech, and he's asking the deeper question: how do we actually embrace this fully, not just put it on a poster? And what are the hidden hierarchies we don't even see? The ones that survive every reorg because they were never on the org chart to begin with.
This is one of those topics where everyone nods along at the concept and then goes right back to doing the opposite. The paradox of the startup that calls itself flat while the CEO makes every decision solo, or the family that preaches openness but Dad's opinion is the tiebreaker every time. We all say we want flat, but most of what we call flat is just hierarchy wearing a hoodie.
Hierarchy in casual Friday attire. The tyranny of the approachable boss who's still a boss.
And the timing on this question is sharp. With remote work dissolving traditional office visibility — who's at their desk, who's in the meeting room — the old signals of rank are fading, but the underlying dynamics haven't gone anywhere. If anything, they've gone underground. So the question isn't just how to flatten, it's how to see what's already there.
If flat hierarchy isn't about removing all structure — and it's not — what is it actually? Let's start with the place that made it famous. Israel gets cited constantly as the exemplar, and the reputation is mostly earned, but the mechanics are more interesting than the myth.
So operationally, flat hierarchy isn't the absence of structure. It's the minimization of approval layers and the maximization of decision-making proximity to the work. The person closest to the problem has the authority to solve it without asking three levels up. That's the core. It's not "no managers" — it's fewer gates between insight and action.
The contrast is authoritarian top-down culture, where micromanagement is the symptom, not the disease. The disease is decision-making distance. The person who knows the most has the least authority, and the person with authority knows the least. That gap is where turnover lives.
There's a 2023 Gallup study that found micromanagement is the number one predictor of turnover, accounting for thirty-four percent of voluntary departures in tech. Not low pay, not long hours — being treated like a pawn. And that's exactly what the prompt is getting at. Even if you're overworked, having a voice makes the difference between grinding through and walking out.
Why does it keep coming up? The shorthand is "balagan" — which loosely translates to chaotic, messy, everyone talking at once. But the reality is more specific. Israeli military culture, particularly in units like Unit 8200, trains junior people to challenge superiors as a matter of doctrine. A twenty-two-year-old lieutenant can override a general's tactical plan in real time if they have better information. And that's not insubordination — it's expected.
Unit 8200 is the signal intelligence unit, and it's produced more startup founders than any other military unit globally. Over a thousand companies — Wix, Check Point, Palo Alto Networks — all trace back to alumni of that unit. And the cultural DNA they carry isn't "be chaotic." It's "the best argument wins, regardless of rank." That's the mechanism.
Which is a very different thing from "nobody's in charge." Someone is in charge. The general is still the general. But the general doesn't get to be wrong just because they're the general.
That maps directly to Israeli business culture. There's a concept called "rosh gadol" — literally "big head." It means taking initiative without waiting for permission, thinking like an owner rather than an employee. In a hierarchical culture, rosh gadol gets you in trouble. In a flat one, it's the baseline expectation.
The musical equivalent of middle management — the glockenspiel of corporate approachability. Everyone sees it, nobody needs it, but somehow it's still in the arrangement.
That's a perfect image. And here's the thing — the term "flat hierarchy" was popularized by Ricardo Semler's 1993 book Maverick, describing Semco's radical democratic management model in Brazil. Workers set their own salaries, voted on major decisions, chose their own managers. It was the extreme end of the spectrum. But most companies that try to implement flat hierarchy aren't doing Semco. They're doing something closer to what Amazon calls "two-pizza teams" — small autonomous groups that can be fed with two pizzas, with clear ownership and minimal external approval.
Which is really just the Israeli model with better pizza.
But the principle is the same. Decision-making proximity. The team owns the problem and the solution, and the hierarchy exists to support, not to direct.
We've established the ideal. But before we get to the business applications, we need to understand the mechanism that makes flat work. And that starts with a military unit that treats a twenty-two-year-old's opinion as seriously as a general's.
Unit 8200's training is built around what they call "challenge culture." New analysts are explicitly trained to question orders and propose alternatives. It's not optional and it's not tolerated — it's required. If you see a flaw in your commander's plan and you don't speak up, that's a failure of duty. The feedback loop flattens decision-making without removing rank. Rank still exists. The difference is that rank doesn't confer immunity from being wrong.
This creates something interesting psychologically. In a traditional hierarchy, the junior person is carrying two burdens: the work itself and the emotional labor of deference. They have to phrase things carefully, read the room, wait for the right moment. Challenge culture removes the deference tax. You're not being rude — you're doing your job.
The deference tax. I'm stealing that. And it shows up in business constantly. Meetings where the most junior person has the best idea but spends twenty minutes finding the right way to say it, by which point the conversation has moved on. The idea dies not because it was bad but because the social cost of delivering it was too high.
How does this translate to actual companies? The Israeli startup scene is the obvious case. Wix, Fiverr, Monday dot com — these companies grew out of a culture where arguing with your boss isn't a career-limiting move. But here's where it gets nuanced. Flat hierarchy requires two things simultaneously: high psychological safety and high competence. If either one is missing, it collapses.
This is the tradeoff nobody talks about. Psychological safety means people feel safe taking risks and being vulnerable in front of peers — Amy Edmondson's work at Harvard defined this. Competence means people actually have the domain expertise to make good calls. If you have safety without competence, you get friendly chaos where nobody knows what they're doing. If you have competence without safety, you get silent experts who know the answer but won't speak up. Either way, the flat structure is just theater.
Like a family dinner where everyone's encouraged to speak freely but Uncle Steve's opinion on everything is still treated as fact because he's loud and he's been around forever.
And that brings us to the first hidden hierarchy, the one that survives every reorg. The expertise hierarchy. The person who knows the legacy codebase, or the one who understands the regulatory framework, or the one who's been there since the founding — they have de facto veto power even in a supposedly flat org. Nobody put them in charge, but nothing moves without their nod.
This is the "ask Bob" phenomenon. Bob isn't your manager. Bob has no reports. But Bob wrote the original database schema in 2011 and if Bob says a migration will take six months, it takes six months. Bob is a hidden bottleneck, and nobody can fire Bob because Bob is the only one who knows how anything works.
There was a 2025 study of two hundred Israeli startups that looked at exactly this. The firms with formal flat structures but high expertise concentration — meaning one or two people held most of the critical knowledge — had the same turnover rates as traditional hierarchical firms. The flat structure didn't matter because the real power was concentrated in the experts. People still felt like pawns. They just couldn't point to the org chart to explain why.
Flattening the org chart without distributing expertise is just rearranging deck chairs. The hidden hierarchy is still there, it's just invisible.
Valve Software is the cautionary tale here. Valve is famous for its "no managers" model — the employee handbook is practically a manifesto. No titles, no bosses, desks on wheels so you can physically move to whichever project you want to work on. But what actually emerged, over time, were informal power cliques forming around senior engineers. The people who'd been there longest, who had social capital with the founders, who controlled access to the most interesting projects. The hierarchy didn't disappear. It went underground.
The musical equivalent of beige wallpaper. It's there, it's shaping the room, you just stop noticing it.
That's the expertise hierarchy. But there's a second one that's even harder to see. The social capital hierarchy. Who gets heard in meetings depends on tenure, friendship with the CEO, sheer extroversion, or even just who speaks first and loudest. In a flat org, this can actually get worse because there's no formal process to counter it. At least in a hierarchy, you know the manager speaks last. In a flat meeting, the person with the most social capital dominates and everyone else thinks, well, that's just how the conversation went.
The tyranny of the person who's comfortable with silence. They don't rush to fill gaps, so when they do speak, it lands with weight. Meanwhile the junior person who's been formulating a point for ten minutes gets cut off because they paused for breath.
Extroversion bias is a real, documented phenomenon. Quiet competence loses to confident mediocrity in flat structures all the time. The loudest voice isn't the most informed voice, but without a formal process to surface input, loud wins.
Which brings us to hidden hierarchy number three. The temporal hierarchy. The person who's been there longest doesn't just know the codebase — they know the unwritten rules. Who to cc, which meetings actually matter, which projects are career poison. They can gatekeep decisions without any title because they control the context. New people can't even see what they're being excluded from.
This is the onboarding tax. You join a flat org thinking everyone's equal, and six months later you realize you've been making decisions without information that three people had but never thought to share because "everyone knows that." The temporal hierarchy is a knowledge cartel, and it's the hardest one to break because it doesn't feel like power. It feels like experience.
Like adopting a feral cat. You think you're in charge, but the cat knows where the food is and you don't.
The cat has been there since the Series A. So how do you actually surface and dismantle these hidden hierarchies? This isn't just diagnosis — there are practical techniques. One is round-robin decision-making, where everyone speaks before the leader. The most junior person goes first, and the leader speaks last. This directly counters the social capital hierarchy by structuring whose voice lands when.
The Israeli military's after-action review culture does exactly this. After a mission, the most junior person speaks first. What went wrong, what went right, what would you change. Then you go up the chain. The commander speaks last, because if the commander speaks first, everyone else's answer becomes "what the commander said." Google's Project Aristotle teams adopted this same technique after studying what made high-performing teams work.
There's a fascinating 2024 experiment at a mid-sized tech firm in Tel Aviv that implemented what they called "decision transparency." Every decision was logged with who proposed it, who challenged it, and the final outcome. Not a public shaming thing — an internal log. Turnover dropped eighteen percent in six months. But meeting times increased forty percent. The transparency worked, but it wasn't free.
Forty percent more meeting time is a real cost. Flat hierarchy isn't efficient in the short term. Hierarchy is fast — one person decides, everyone executes. Flat is slow because it requires deliberation. The payoff is in retention, innovation, and decision quality over time. But you have to be willing to pay the upfront tax.
Another technique is anonymous proposal systems. Not for everything, but for decisions where social capital bias is likely to be high. If you know that the CEO's best friend is going to dominate the discussion on a new product direction, you run an anonymous pre-vote. Everyone submits their reasoning in writing, without names attached, and the group discusses the ideas without knowing whose they are. It's not perfect, but it strips out a layer of social hierarchy.
Then there's the decision journal. This is simpler than it sounds. After any significant decision, you write down who actually influenced the outcome. Not who was in the room — who shaped the result. After three months, you look at the journal and ask: is there a pattern? Are the same three people driving every decision? If so, you have a hidden hierarchy, whether your org chart admits it or not.
This is the audit that almost nobody does because the results are uncomfortable. You find out that your "flat" organization is actually run by two extroverts and a founder's college roommate. The org chart was a lie, but it was a comfortable lie.
Covering the covers.
Now, the prompt asks something that most discussions of flat hierarchy ignore entirely — how does this extend beyond work? The family dinner table as a hierarchy laboratory. And this is where it gets genuinely uncomfortable for a lot of people.
Because at work, you can talk about flat hierarchy as a management philosophy. At home, you're talking about power. Who gets the final word. Whether children get to argue with evidence or just obey. And a lot of parents who preach openness at the office run their households like miniature monarchies.
Flat parenting — and I'm using that term loosely, not as an established framework — would mean letting children challenge decisions with evidence, not just emotion. "Because I said so" is the parental equivalent of "because I'm the boss." It ends the discussion but it doesn't build the skill of reasoning. The question is whether you want compliance or capability.
The pushback is obvious. Children aren't employees. They don't have domain expertise. They can't make informed decisions about bedtime or screen time or whether vegetables are negotiable. So flat parenting isn't about giving children veto power. It's about giving them a voice and taking their reasoning seriously, even when you override it.
The practice would look something like this: when a child argues against a rule, you don't dismiss the argument because of who made it. You engage with the substance. "You think bedtime should be later because you're not tired at nine. Here's why we disagree — your body needs more sleep than you feel like it does, and here's the evidence." You're still making the decision, but you're not pulling rank to avoid the conversation.
That's terrifying for a lot of adults because it means you can't coast on authority. You have to actually be right, or at least be able to explain your reasoning. Flat hierarchy, in any context, raises the bar on the people in charge. You can't just say "because I said so." You have to say "because here's why.
The dark side of all this is that flat hierarchy can become a weapon for passive-aggressive control. The leader who claims openness but uses social pressure to enforce their will. The "open door policy" that nobody actually uses because walking through that door is career suicide and everyone knows it. The manager who says "I want pushback" and then punishes the person who pushes back by freezing them out of interesting projects.
This is the performative flat hierarchy. The one that exists in the all-hands presentation but not in the Slack DMs. And it's worse than honest hierarchy because it adds gaslighting to the power dynamic. At least in a top-down culture, you know where you stand.
There's a term for this in the literature — "pseudo-participation." The leader creates the appearance of consultation while having already made the decision. It's democracy as theater. And it's corrosive because people figure it out fast, and then the flat structure has less trust than a hierarchical one. You've burned the bridge and the destination.
How do you tell the difference? Genuine flat hierarchy versus the performative version?
One test: does anyone ever change the leader's mind? If the answer is no, or if it happened once in 2019 and people still tell the story, you don't have flat hierarchy. You have a monarch who's polite about it. Another test: who speaks last in meetings? If it's always the same person, and that person's opinion is treated as the conclusion rather than an input, the flatness is cosmetic.
The "first to speak" rule we mentioned earlier is a forcing function for this. If the leader speaks last, they can't set the frame. The frame emerges from the group. And if the group's conclusion differs from the leader's initial instinct, that's visible to everyone. You can't pretend the discussion was open when the leader spoke first and everyone else just elaborated on the theme.
This connects to something the prompt hinted at — the hidden hierarchies we're not even cognizant of. The ones that feel natural. The person who always summarizes the meeting at the end, subtly steering the takeaways. The person who controls the agenda by being the one who schedules the meetings. The person who "just happens" to have a sidebar with the decision-maker before the formal discussion. None of these are on the org chart.
The pre-meeting meeting. The most powerful room is the one that happens before the room.
In remote work environments, this gets amplified. The pre-meeting meeting is now a Slack DM. The sidebar is a quick huddle that wasn't on anyone's calendar. The hidden hierarchy moves to channels that leave no paper trail. You can't audit what you can't see.
Let's get practical. All of this sounds great in theory, but what do you actually do on Monday morning?
First actionable step: audit your hidden hierarchies. This week, track who spoke last in your last three meetings. Track who made the final call on each decision. Then ask the uncomfortable question: did that person have the most relevant expertise, or just the most seniority or social capital? Don't publish the results. Just look at them yourself. The first step is seeing what's actually there.
If you're brave, extend that audit to your family. Who gets the final word at dinner? Whose preferences shape the weekend plans? If the answer is always the same person, you have a hierarchy. That's not automatically bad, but it's worth knowing.
Second step: implement one flat practice this week. The "first to speak" rule — junior voices first, leader last. Or the decision log — write down who influenced a decision and why, even if it's just a paragraph in a private doc. The point isn't to transform your culture overnight. It's to build the muscle.
For families or social groups, try a "no final word" dinner. No conclusion, no tiebreaker, no summary. Just discussion that ends without resolution. See how uncomfortable it feels. That discomfort is the hierarchy revealing itself. We're so trained to look for the authority figure to wrap things up that the absence feels like failure.
The meta-insight here is that flat hierarchy is a muscle, not a switch. You don't flip it on Monday and have a flat culture by Friday. It requires constant exercise and constant awareness of the hidden hierarchies that creep back in. The moment you stop paying attention, the expertise hierarchy reasserts itself, the social capital hierarchy regrows, and the temporal hierarchy entrenches. Flatness is maintenance, not installation.
That's the thing most companies miss. They announce a flat reorg, remove a layer of middle management, declare victory, and then wonder why nothing changed. The org chart flattened, but the power dynamics didn't. You have to change the practices, not just the boxes.
There's a 2025 Harvard Business Review analysis that looked at a hundred and fifty flat-org companies. Sixty percent reverted to some form of hierarchy within three years, usually due to decision paralysis. The flat structure created so many voices that nothing got decided. And when nothing gets decided, someone eventually steps in and starts deciding, and congratulations, you've reinvented hierarchy, just without the accountability structure.
Decision paralysis is the failure mode of flat hierarchy. It's what happens when you have voice without mechanism. Everyone gets heard, nobody gets resolved. The Israeli model avoids this because the challenge culture has a built-in endpoint — someone still makes the call. The general still commands. The difference is that the general has been forced to defend the decision against the best arguments in the room.
That's the distinction between flat hierarchy and no hierarchy. Flat hierarchy has clear accountability. Decisions get made. The difference is the quality of the inputs before the decision. No hierarchy means nobody's accountable and nothing gets decided. That's not flat — that's formless.
Like a band with no drummer. Everyone's playing, nobody's keeping time.
So the Israeli model isn't "everyone decides." It's "everyone argues, and then the responsible person decides, and then everyone executes regardless of whether they won the argument." The flatness is in the input, not the output. Once the decision is made, the hierarchy reasserts itself for execution. That's the part that gets left out of the TED talks.
That's why it works in the military context. You can't have a debate about whether to take the hill while bullets are flying. The debate happens before. Then you move. The discipline is in the execution, not in the silence during planning.
Let's bring this home with something concrete. If someone listening wants to capture the good parts of flat hierarchy without the chaos, what's the minimum viable practice?
One, separate input from decision. Make it clear when you're in input mode — everyone speaks, rank is suspended, the goal is to surface the best arguments. Then make it clear when you're in decision mode — one person decides, based on the input, and explains their reasoning. The transparency of the reasoning is what maintains trust.
Two, rotate who speaks first. Not just in meetings, but in decision processes. If the same person always frames the problem, they're controlling the outcome regardless of who decides. Framing is power.
Three, write down who actually influences decisions. Not as a punishment, as an awareness tool. Most hidden hierarchies survive because nobody looks at them. The moment you measure them, they start to shift. Sunlight is a surprisingly effective disinfectant.
For families, the adaptation is simpler but harder. When your child argues with you, engage with the argument before you engage with the authority. "Because I said so" should be the last resort, not the first response. You're teaching them that reasoning matters. And you're teaching yourself that authority isn't a substitute for being right.
Flat hierarchy is a practice, not a policy. But it leaves us with an uncomfortable question. If flat hierarchy requires high trust and high competence simultaneously, what happens in environments where those are lacking? Is flat hierarchy a privilege of high-performing teams, or a tool to build them? Can you flatten a dysfunctional team, or does flattening just remove the guardrails?
That's the tension. I think the evidence suggests flat hierarchy is a multiplier. It makes good teams better and bad teams worse. If you have trust and competence, flattening amplifies both. If you have neither, flattening removes the structure that was holding things together, and you get chaos or hidden power grabs. So the question isn't just "should we flatten?" It's "are we ready to flatten?
The future implication that's worth sitting with: as AI decision-support tools become more embedded in workplaces over the next few years, they could cut either direction. They could flatten hierarchies by democratizing information — giving everyone access to the same analysis, the same data, the same recommendations. Or they could reinforce hierarchies by giving leaders better data to control from the top, while everyone else just executes the algorithm's output.
The tool is neutral. The question is who controls the inputs and who sees the outputs. If the AI's analysis is available to everyone, it's a flattening force. If it's a dashboard that only the VP sees, it's a hierarchy amplifier. Same technology, opposite outcomes.
The flat hierarchy conversation isn't going away. If anything, it's about to get more complicated. Which means the practices we talked about — auditing hidden hierarchies, rotating voice, logging influence — those aren't just nice ideas. They're going to be survival skills.
Now: Hilbert's daily fun fact.
Hilbert: Suriname is home to an estimated thirty-seven distinct regional variants of the abacus, many of which use seeds from local palm species as counting beads and are still taught in rural schools as part of the national mathematics curriculum.
Thirty-seven abacus variants.
I have so many follow-up questions and I know I'll never get answers to any of them.
If you want to dig deeper into the Israeli military's decision-making culture, we've linked the Unit 8200 declassified training manual in the show notes. It's a fascinating read on how challenge culture is institutionalized from day one.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop. Find us at myweirdprompts dot com or wherever you get your podcasts. If this episode gave you something to think about, leave us a review — it helps.
Until next time.
