What Managers Look for When Promoting Employees

Every company has a rubric. It lists competencies, impact areas, and behavioral expectations by level. It's the official answer to "what do I need to do to get promoted?"

It's also not the full picture.

The rubric tells you what the criteria are. It doesn't tell you how the decision actually gets made. It doesn't explain why one engineer who meets the criteria gets promoted while another who also meets them waits another cycle. And it doesn't prepare you for the fact that the most important part of the process happens in a room you're not in, during a conversation you'll never hear.

Managers know things about how promotions work that they rarely say out loud. Not because they're hiding something, but because the informal signals that determine outcomes are harder to articulate than a rubric item. Here's what they actually look for.

The person is already operating at the next level

This is the single most consistent thing managers and engineering leaders say about who gets promoted. Not "shows potential to operate at the next level." Already doing it.

Gergely Orosz, author of The Pragmatic Engineer, describes the standard at most tech companies: you're expected to act and take responsibility at the next level for six to twelve months before anyone considers you for promotion. The promotion is a formal acknowledgment of a change that's already happened, not a bet on what might happen next.

What "next level" actually looks like varies by company, but one framing from Rahul Pandey (ex-Meta, ex-Pinterest) cuts through the noise. A former manager told him: "Working extremely hard to achieve next-level impact is actually a negative signal for promotion. Burning the midnight oil isn't sustainable and implies that the candidate cannot tackle problems of larger scope."

The real signal isn't effort. It's how you approach problems. A mid-level engineer picks up a defined task and executes it well. A senior engineer sees an ambiguous situation, figures out what needs to happen, and drives it forward without waiting for someone to break it down. The difference isn't skill level. It's operating mode.

The manager can tell a compelling story in 90 seconds

Your manager does not get a 30-minute presentation to argue for your promotion. In most calibration meetings, they get one to two minutes per person. In that window, they need to describe what you did, why it mattered, and why you specifically should be promoted over other candidates.

If the story is thin, the outcome defaults to "not this cycle."

This is why the calibration room determines your outcome more than the rubric does. When budget constraints force a fixed number of promotions per team or organization, the people who get through are the ones whose managers can argue most compellingly. A manager who says "she did great work this year" loses to a manager who says "she led the API migration that cut timeout incidents by 40% and unblocked two dependent teams."

Your job is not just to do the work. It's to make sure your manager has the raw material to build that 90-second case. Specific outcomes, quantified impact, names of teams and stakeholders who benefited. If your manager has to reconstruct your contributions from memory under pressure, you're gambling on how much they remember.

The person is visible to people beyond their manager

A study by the Center for Creative Leadership examined 64 promotions at three Fortune 500 companies and found two numbers worth sitting with:

• In 73% of cases, the person promoted had already worked directly with the hiring manager or that manager's boss
• In 81% of cases, the promoted person had a career-facilitating relationship with someone higher in the organization who helped their visibility

The researchers noted that bosses did not feel compelled to look beyond people they already knew. Familiarity, not just merit, drove the decision.

This pattern shows up everywhere once you look for it. On Team Blind, engineers stuck at the same level for years describe a consistent pattern: they were doing high-impact work that nobody outside their immediate team knew about. The infrastructure engineer whose service enabled three product teams. The backend engineer who removed a bottleneck that saved senior engineers two days a week. Excellent work, zero visibility.

Visibility isn't optional. If the people in the calibration room don't know your name, your manager is arguing for a stranger. That's a harder case to win than arguing for someone the room already recognizes.

How visibility actually works

Visibility doesn't mean self-promotion in a cringey way. It means:

• Writing internal docs about what you built and why it matters, then sharing them with teams affected by the work
• Presenting your work at team-level syncs one level up from where you normally operate
• Showing up in cross-team projects where directors and staff engineers can see how you work firsthand
• Asking teams you unblocked to mention the impact in their own updates

Sean Goedecke, who was promoted to staff engineer twice (at Zendesk and GitHub), describes the mechanism bluntly: "You can't get promoted to staff without your manager doing a lot of work: writing the promo packet with you, figuring out what senior leaders are looking for, schmoozing senior leaders ahead of time so they know your name, and making sure you're assigned lead to key projects."

Your manager's advocacy matters enormously. But they can only advocate for someone the room is prepared to hear about.

The person makes the manager look good

This one makes people uncomfortable. But the incentive structure is real.

When your manager promotes you, it signals to their leadership chain that they're effective at developing talent. A team that produces promotions is a well-run team. A manager whose reports stagnate for years has a harder time arguing for their own advancement.

This alignment works in your favor. Your manager wants to promote you, because your success reflects well on them. But they can only do it if you give them the ammunition and the visible results.

The flip side is less comfortable: some managers avoid promoting their best people because it's disruptive to replace them. On Team Blind, this dynamic gets discussed regularly. If you're a high performer managing large projects, your departure from the current role creates an operational problem your manager doesn't want. Promotion happens when it's in both your interest and your manager's interest, which is why getting your manager to actively fight for your promotion requires more than just doing good work.

"Potential" matters more than performance

This is the finding that surprises most engineers.

A study from MIT Sloan and the National Bureau of Economic Research analyzed 30,000 employees across multiple review cycles and found a striking asymmetry:

• Moving from a medium to high potential rating increased promotion likelihood by 75%
• Moving from a medium to high performance rating increased promotion likelihood by 27%

Potential was roughly 2.8 times more influential than performance in determining who got promoted.

What does "potential" mean in practice? It's the manager's subjective assessment of how much more you can grow. And it's where bias enters the process most aggressively. The same study found that women received 8.3% lower potential ratings than men despite receiving higher performance scores. This gap in potential ratings explained up to 50% of the overall gender promotion gap.

For the engineer trying to get promoted, the implication is concrete: demonstrating that you can handle bigger scope, more ambiguity, and more organizational complexity matters more than executing your current role perfectly. You need to show trajectory, not just output. That's the informal difference between "great at their level" and "ready for the next one."

The person self-advocates (and it affects the rating)

Most engineers believe their work should speak for itself. The data says otherwise.

A Harvard Kennedy School study tracked four review cycles at a multinational firm. During one cycle, a technical glitch prevented managers from seeing self-evaluations before assigning ratings. That cycle, average ratings dropped and became less correlated with self-scores. When the glitch was fixed and managers could see self-evaluations again, ratings rose back to normal.

Self-ratings exert social influence on manager evaluations. When you rate yourself highly and present evidence to support it, your manager's rating is pulled upward. When you undersell yourself, the manager has less reason to push the number higher.

Separately, research from Harvard Business School and Wharton found that men rated their own performance 33% higher than equally performing women. Workers who self-rated higher were more likely to be hired and offered higher pay. The gap persisted even when workers were told their actual scores and how they compared to peers.

Self-advocacy isn't arrogance. It's an input the system is designed to receive. When you undersell yourself in a self-review, you're not being modest. You're providing your manager with weaker ammunition for a calibration conversation where ammunition is everything.

The calibration room is political (and that's normal)

Calibration meetings are where final ratings and promotion decisions happen. They're also zero-sum. Organizational budgets for compensation increases and promotion slots are fixed. Every promotion approved for one team is one fewer available for another.

The dynamics of that room matter:

• Anchoring bias. The first manager to advocate strongly for a rating tends to sway others. If your manager speaks early and confidently, the outcome skews in your favor.
• Manager credibility. Not all managers advocate with equal effectiveness. One calibration participant described it: "Promotion is largely a function of how good a manager is at verbally promoting their staff. Folks whose manager has a tongue of gold get promoted at the expense of others."
• Budget constraints. Promotion budgets create hard caps. "When it's gone, it's gone. Being generous with ratings or promotions is just spreading the same amount of butter over more bread."
• Forced distributions. Many companies still use bell curve guidelines that limit how many people can receive top ratings. What started as guidelines became quotas that managers had to hit regardless of actual performance.

None of this means the process is hopelessly corrupt. It means the process is human. The rubric defines what "promotable" looks like. The calibration room decides who, among everyone who's promotable, actually gets through. Understanding the difference is the first step toward navigating it.

Sponsorship doubles your odds

The 2025 McKinsey and LeanIn Women in the Workplace study found that employees with sponsors — senior people who actively advocate for your career, not just offer advice — are promoted at roughly twice the rate of those without.

A sponsor is different from a mentor. A mentor answers your questions. A sponsor mentions your name in rooms you're not in. They steer opportunities toward you. They vouch for you when a promotion is on the line.

The research found that at the entry level, 45% of men had sponsors compared to 31% of women. The gap narrowed at senior levels but never closed. Four in ten entry-level women had not received a promotion, stretch assignment, or leadership training in the past two years.

Building a sponsor relationship isn't transactional. It happens when a senior person sees your work, believes in your potential, and decides to invest their own credibility in your advancement. The most reliable path to attracting sponsorship is to do visible, high-impact work on projects that your potential sponsor cares about, and to be clear about where you want your career to go.

What this means for you

The rubric is the ticket to entry. Meeting the criteria makes you eligible. But eligibility is not the same as getting promoted. The people who get promoted understand that the process has two layers: the formal criteria and the informal signals that determine who, among those who qualify, actually advances.

What you can control:

• Document your impact continuously, not at review time. Give your manager specific, quantified wins they can use in their 90-second calibration pitch.
• Operate at the next level visibly, not quietly. The work doesn't count if the right people don't see it. Cross-team projects, internal documentation, and presentations one level up all build the familiarity the CCL study found in 73% of promotion cases.
• Self-advocate in your self-review. Rate yourself based on evidence, not modesty. Your self-review is an input the system is designed to receive. Underselling yourself gives your manager less to work with.
• Invest in the manager relationship. Your manager's willingness and ability to advocate for you is the single most important variable. Make their job easy. Bring them the evidence, the framing, and the story.
• Build relationships one level up. If your skip-level knows your name and your work, your manager's pitch in calibration lands on familiar ground instead of blank stares.

The system isn't perfectly fair. But it's more navigable than most engineers realize, because most engineers are only paying attention to the rubric.

---

CareerClimb helps you build the evidence, track your wins, and prepare for the conversations that shape how your manager talks about you in calibration. Your AI coach Summit works with your specific situation — your manager, your company, your gaps — so the case is ready when the moment comes. Download CareerClimb