Why Your Best Employees Quit (The Corporate Black Hole Effect)

A few years ago, I was consulting for a Fortune 500 company that had just gone through a painful round of layoffs.

The CEO pulled me aside after a workshop and said something I will never forget: “Yu-kai, we kept the wrong people.”

He wasn’t talking about skill sets or credentials. He was talking about something harder to measure. The people who survived the cuts were the ones who had built relationships with decision-makers. The people who were let go were the ones who had quietly delivered the best work.

His company had accidentally optimized for politicians and punished performers. And by the time he realized it, his best talent had already walked out the door.

If you are a leader reading this and that line made you flinch, you already suspect the same thing is happening to your team. The four-type framework below explains the mechanism, and the one decision that determines which way it goes.

This pattern shows up in almost every large organization I have worked with. And after years of observing it across dozens of companies and governments (including my work advising Ukraine’s digital transformation), I developed a framework that maps every employee into one of four types. (I first wrote about the basic Corporate Player Types matrix years ago, and later expanded on how to become irreplaceable using this framework.)

My colleague Joris Beerda later gave these types names borrowed from astrophysics, and those names stuck because they capture something the original labels never could: the gravitational pull each type exerts on the people around them.

Author Credibility: Yu-kai Chou

Yu-kai Chou created the Octalysis Framework after studying gamification since 2003 — years before the term entered mainstream vocabulary. As a Human-Systems Architect & Behavioral Designer, his framework has been applied by LEGO, Microsoft, Porsche, Coca-Cola, Salesforce, and MrBeast, impacting over 1.5 Billion Users.

Chou has taught the Octalysis methodology at Harvard, Stanford, Yale, Tesla, Google, BCG, and IDEO.

His work has been cited by Harvard, Stanford, MIT, Forbes, Wall Street Journal, Wired, US Department of Energy, NIST, NSF, NCBI, US Department of Education, ClinicalTrials.gov, and Google Scholar — with 3,700+ more academic publications. Explore his books here.

TEDxLausanne · Gamification to Improve Our World17 min

If you want the deeper version of this matrix and the full intervention playbook, start with the Octalysis Framework.

The Two Axes That Define Every Employee

Before I walk you through the four types, you need to understand the two dimensions I’m measuring.

Axis 1: Performance. This is straightforward. Can this person deliver results? Not “do they seem busy” or “do they work long hours.” Can they actually produce high-quality output that moves the business forward?

A critical distinction here: performance does NOT mean business impact. An employee might do exceptional work that nobody notices because they lack the second axis. Business impact requires both dimensions working together.

Axis 2: Political Skill. Before you cringe at the word “political,” hear me out. I don’t mean manipulation or backstabbing. Political skill is the ability to build relationships, communicate your value, mobilize people across teams, and make others feel invested in your success.

Some people call it “emotional intelligence” or “networking ability.” I call it political skill because in a corporate environment, that’s exactly what it is. The person who convinces three VP teams to cooperate on a shared goal is exercising political skill. So is the person who takes the VP to lunch every week to make sure their mediocre project looks important.

Same skill. Wildly different intentions. That difference is everything.

The Four Corporate Player Types

Plot those two axes on a grid, and you get four quadrants. Four types of employees that exist in every organization on Earth.

The distribution is not even. In large corporations, Novas are the majority, often 70% or more. Protostars and Black Holes each make up small minorities. Stars are an extreme minority, often just a handful of people in a thousand-person organization. In startups, the ratios invert: mostly Protostars and Stars, because the environment cannot yet support the other two types.

Novas (Low Performance + Low Political Skill)

The majority of any organization falls here. These are the people who show up, do their job at an acceptable level, collect their paycheck, and go home.

Their internal monologue sounds like this: “Why should I go above and beyond? I worked hard last year and didn’t get a bonus. What’s the point?”

Novas aren’t lazy. Many of them started as something more. They were once excited about their work, maybe even passionate. But repeated signals from the organization taught them that extra effort doesn’t get rewarded. So they stopped trying.

In astrophysics, a nova is a star that has already burned through its fuel. That’s exactly what these employees are: burnt-out protostars that will likely never shine again, just surviving.

Here’s what makes this dangerous. Novas aren’t a problem by themselves. They become a crisis when they’re the only type left. If your organization is full of Novas, it means every other type has either left or given up.

Black Holes (Low Performance + High Political Skill)

These are the employees who do mediocre work but have perfected the art of appearing valuable.

Their signature move: “I’m going to invite the VP to lunch this week and make sure he knows how critical my role was in the project.” Never mind that their actual contribution was minimal. The VP walks away thinking this person is indispensable.

Black Holes network relentlessly. They claim credit. They position themselves at the center of every visible initiative. They make something harsh sound pleasant, even if it means exaggerating their involvement or downplaying others’ contributions.

The name “Black Hole” is not an insult. It’s a physics metaphor that is terrifyingly accurate. In astrophysics, a black hole warps the space around it so that everything nearby gets pulled toward it. Corporate Black Holes do the same thing to organizational culture. They warp incentives so that political behavior becomes the rational strategy for everyone.

And here’s the poetic part that my colleague Joris captured perfectly: “If Novas are under the influence of Black Holes, they run the risk of turning into Black Holes themselves.”

That one sentence describes cultural death in most organizations.

Protostars (High Performance + Low Political Skill)

These are your best workers who hate politics.

Their belief system: “If I keep doing great work, eventually someone will notice and promote me.” They put their heads down, deliver exceptional results, and trust that meritocracy will handle the rest.

It almost never does.

The tragedy of Protostars is that their work speaks for itself, but nobody hears it because they refuse to amplify it. Schmoozing with executives? Not their style. Taking credit in meetings? Feels dishonest. Playing the lunch game? Beneath them.

In astrophysics, a protostar is a young star that hasn’t yet ignited nuclear fusion. It has all the mass and potential energy of a star, but something is missing. For corporate Protostars, that missing element is visibility.

MIT Sloan’s research on the Great Resignation found that toxic culture was 10.4 times more powerful than compensation in predicting whether someone would quit. When I read that statistic, I immediately thought of Protostars. They are the ones who leave first, because they have options. They can perform anywhere. And they will go somewhere that recognizes it.

This is connected to Core Drive 2: Development & Accomplishment in my Octalysis Framework. Protostars are driven by genuine mastery and progress. When the environment tells them that mastery doesn’t matter, that political maneuvering matters more, they experience what I call an Anti Core Drive response. The thing that motivated them becomes the thing that repels them.

Stars (High Performance + High Political Skill)

Stars deliver exceptional results AND know how to mobilize people. They perform AND they build coalitions.

Their thinking: “I need to figure out how to get all the VP teams to cooperate so we can hit projections this year.” Beyond doing great work themselves, they create the conditions for great work to happen across the organization.

In most companies I have seen, true Stars are a small minority. They usually do well regardless of culture because they have both the skill to produce and the savvy to be recognized.

The problem is that Stars are also the most poachable. Every competitor wants them. Every recruiter targets them. If your organization doesn’t protect and empower its Stars, someone else will happily take them.

The One Decision That Determines Everything

Here’s the insight that makes this framework actually useful, not just a categorization exercise.

Your company’s culture is determined by a single thing: what happens when a Black Hole gets promoted.

Not your mission statement. Not the company values poster in the lobby. Definitely not your “culture deck” on your website. What actually happens when a politically skilled but mediocre performer gets rewarded over a high performer who doesn’t play the game.

Let me trace the cascade.

Scenario A: The Black Hole Gets the Promotion.

Every Protostar in the organization sees it happen. They think: “I delivered twice the results, but this person got ahead because they had lunch with the VP every week.”

Protostars don’t suddenly become political. It’s not in their nature. Instead, they lose motivation. Their performance drops. They slide from Protostar to Nova. Some of the best ones leave entirely.

Meanwhile, the Novas see the promotion too. Their takeaway is different: “Oh, THAT’S how you get ahead here.” The ones with any social skill start networking, credit-claiming, and visibility-seeking. They become Black Holes.

Give that pattern enough promotion cycles, and you start losing Protostars while gaining more Black Holes. The entire gravitational field of the organization shifts toward politics. Performance becomes secondary. The people who remain are either surviving (Novas) or gaming (Black Holes).

This is how organizations die from the inside. Not through one catastrophic failure, but through a slow, invisible cultural shift that makes high performers feel stupid for trying.

Scenario B: You Actively Discourage Black Hole Behavior.

The approach isn’t punishing Black Holes. It’s simpler than that: stop rewarding political behavior when it isn’t backed by performance. Make the criteria for advancement transparent and tied to measurable output.

At the same time, you invest in your Protostars. Not by telling them to “be more political” (they’ll resent that), but by giving them visibility. Put them in front of leadership. Assign them to cross-functional projects where their work becomes impossible to ignore. Teach them the people skills they’re missing, but frame it as leadership development, not politics.

For Novas, you add what I call Core Drive 1: Epic Meaning & Calling. You connect their daily work to something larger. You add recognition systems that make effort visible. You gamify the environment so that progress feels tangible.

For Stars, you empower them with autonomy (Core Drive 3: Empowerment of Creativity & Feedback) and protect them from organizational drag.

Result: More Protostars develop into Stars. More Novas re-engage as Protostars. Black Holes either adapt or self-select out. The gravitational field shifts toward performance.

The Google Exception (And Why You Probably Can’t Copy It)

Early Google managed something remarkable: a company composed almost entirely of Stars and Protostars.

How? They made the filter so aggressive that Black Holes and Novas rarely got through the door.

Google’s hiring process was famously brutal. Brain teasers. Multiple interview rounds. Technical challenges that even experienced engineers found intimidating. But the real filter wasn’t the difficulty. It was the signaling.

Part of Google’s early advantage was that many candidates were willing to trade some compensation for mission, status, and the chance to work with exceptional peers. In Octalysis terms, that means stronger pull from Core Drive 1: Epic Meaning & Calling (“I want to change the world”) and Core Drive 5: Social Influence & Relatedness (“I want to work with the smartest people”).

Getting hired at Google became a credential in itself. Having “Google” on your resume carried enormous status. That Core Drive 2: Development & Accomplishment signal created a self-reinforcing loop that attracted more Stars and Protostars.

Here’s the reality check: this model breaks down at scale.

Once companies like Google and Facebook entered true hypergrowth, maintaining all-Star quality became dramatically harder. The interview pipeline could not stay equally rigorous forever. Some Black Holes got through. Some Novas slipped in. And once they were inside, the cultural dynamics I described above started playing out.

If you’re running a 50-person startup, you might be able to hire exclusively Stars and Protostars. If you’re running a 5,000-person organization, you need a system that works with all four types. Which means you need to get intentional about incentives.

The Death Spiral: When Only Novas and Black Holes Remain

Some of the most spectacular corporate failures in history follow a predictable pattern.

BlackBerry. Blockbuster. Yahoo in its later years. Sears. And more recently, Boeing, where decades of promoting financial and political operators over engineers produced exactly the culture every safety whistleblower has been describing since 2018.

The founding team is usually Stars and Protostars. They build something extraordinary. The company gains market share, revenue momentum, and brand recognition.

Then something subtle happens. New hires ride the momentum without creating it. The company’s success becomes self-sustaining for a while. Revenue keeps coming in because of past decisions, not current innovation. Everyone looks productive because the numbers are good.

But underneath the surface, the composition has shifted. The Stars and Protostars who built the company have either left for new challenges or been promoted into roles where they’re too far from the work. The remaining workforce is increasingly Novas (just keeping their jobs) and Black Holes (playing internal politics for shrinking resources).

BlackBerry’s culture in its final years was described by employees as “uninspired and sterile” with a “resistance to change and an insular mindset.” That’s what a Nova/Black Hole organization sounds like. Nobody is innovating because the Novas don’t see the point and the Black Holes are too busy protecting their territory.

The illusion: “Everyone’s not really doing much, but the company still makes money.” This works until it doesn’t. And when it stops working, it stops fast.

By the time leadership realizes the Stars are gone, it’s often too late. You can’t hire your way back to a Star culture overnight. The Black Holes will resist any new high performer who threatens their position, and the Novas will watch passively from the sidelines.

Part of why this is so hard to catch in time: Stars do not downshift when they’re upset. A demoralized Nova checks their phone in meetings. A demoralized Protostar misses deadlines. A demoralized Black Hole gets louder. But a demoralized Star keeps delivering at 100% right up to the day they email their two-week notice. The only visible signal is subtle. They stop saying “we” and start saying “the company.” They stop volunteering for projects that stretch beyond the current quarter. By the time you act on those signs, three offers from competitors are already in their inbox.

What To Actually Do About This

If you’re a leader reading this, here’s the diagnostic framework.

Step 1: Audit your last ten promotions. How many went to people with visible political skill versus measurable performance? If more than half went to people known for relationships rather than results, your organization is drifting toward Black Hole dominance.

Step 2: Identify your Protostars by name. These are your most talented people who never show up on leadership’s radar. They’re often the ones other employees go to for help, the ones who quietly fix things that are broken, the ones whose work is referenced in meetings they aren’t invited to. If you can’t name at least three Protostars, you’re not looking hard enough.

Step 3: Check your Novas’ backstory. Talk to your disengaged employees. Not about their performance. Ask them what they were excited about when they first joined. Ask them when that changed. You’ll often discover that a specific event turned them from Protostar to Nova: a promotion they deserved that went to someone else, a project that was canceled for political reasons, a manager who took credit for their work.

Each of those stories is a data point about your culture.

Step 4: Change the reward signal. Make one visible, undeniable promotion decision based purely on performance. Not quietly. Publicly. Let everyone see that the person who delivered the best results got recognized, even though they never played the political game. That single action will do more for your culture than a hundred town halls about “our values.”

Step 5: Invest in Protostar development. Create a program specifically for high performers who lack political skill. Don’t call it “political training.” Call it “leadership development” or “executive communication.” Teach them how to present their work, build cross-functional relationships, and communicate their impact without feeling like they’re selling out. You’re not asking them to become politicians. You’re helping them become Stars.

This maps directly to my Octalysis Framework. Core Drive 3 (Empowerment of Creativity & Feedback) comes alive when Protostars get the tools and autonomy to grow. Core Drive 5 (Social Influence & Relatedness) kicks in when they connect with mentors and peers. And Core Drive 2 (Development & Accomplishment) gets reinforced through visible milestones that track their leadership growth.

The Real Test of Culture

I’ve worked with organizations on six continents. I’ve advised governments trying to motivate entire populations and startups trying to keep ten people engaged. The pattern never changes.

The organizations that thrive are the ones where Protostars feel safe, Black Holes feel exposed, Novas feel a reason to re-engage, and Stars feel empowered.

The organizations that die are the ones where the opposite is true.

And the difference between those two outcomes almost always traces back to a small number of promotion and reward decisions that signal what the organization actually values.

Not what it says it values. What it actually rewards.

If you look around your organization and see a lot of Novas, the question isn’t “How do I motivate these people?” The question is “What did we do to turn them into Novas?”

Answer that honestly, and you’ll know exactly where to start.

Want to hear me walk through this framework live? Here is the original 5-minute lecture this post is built on.

Frequently Asked Questions

What are the four corporate player types?

The four types are Stars (high performance + high political skill), Protostars (high performance + low political skill), Black Holes (low performance + high political skill), and Novas (low performance + low political skill). These are mapped on a 2×2 matrix of Performance versus Political Skill, and every employee in every organization falls into one of these quadrants.

Why do the best employees quit first?

High performers with low political skill (Protostars) quit first because they have options. MIT Sloan research found that toxic culture is 10.4 times more powerful than compensation in predicting attrition. When Protostars see politically skilled but mediocre performers (Black Holes) getting promoted ahead of them, they leave for organizations that will recognize their contributions based on merit.

How does rewarding politicians destroy company culture?

When a politically skilled but mediocre performer gets promoted, it sends a signal throughout the organization. High performers lose motivation and either leave or disengage (becoming Novas). Meanwhile, disengaged employees see that political behavior gets rewarded, so they start playing politics (becoming Black Holes). Give that pattern enough promotion cycles, and the organization shifts away from performance toward political maneuvering.

Can you build an all-star organization?

Early Google came unusually close through aggressive hiring filters, strong mission signaling, and the status of working with exceptional peers. But that model gets harder to maintain at hypergrowth scale. Most organizations need systems that work with all four types, focusing on preventing Black Hole behavior from being rewarded and investing in developing Protostars into Stars.

What should leaders do when their best employees start leaving?

Audit the last ten promotions. If more than half went to politically skilled people without measurable performance, the cascade has already started. Stop rewarding Black Holes, name your Protostars publicly, and protect your Stars from the bureaucratic friction that signals you do not distinguish them from a median employee. The single most important diagnostic is honest: what does the company actually reward, not what its values poster claims?

WOULD YOU LIKE YU-KAI CHOU TO WORK WITH YOUR ORGANIZATION?