TL;DR
Getting promoted in tech is less about merit than you think. Even high performers face artificial scarcity, political dynamics, and pure randomness. This article breaks down the structural forces working against you, reveals why visibility matters more than ability, and shows you which levers actually move the needle. If you’ve been told to „just keep doing great work,“ you need to read this.
Introduction
Here’s the part they don’t tell you in the onboarding deck: sustained high performance often isn’t enough to get promoted. Not because you’re not good enough. Not because your manager doesn’t like you. But because the promotion system in tech operates more like a lottery with stacked odds than a meritocracy.
The average tech company promotes only 14% of employees annually [1]. At shrinking companies, that drops to 11.5% [2]. That’s not a pipeline problem. That’s a design choice. Companies create artificial scarcity, compress power-law performance into bell curves, and let organizational chaos reset your progress without warning.
If you’ve been grinding away, assuming the work speaks for itself, you’re playing a game you don’t understand. And you’re probably losing to people who do.
The system creates artificial scarcity by design
Let’s start with the obvious stuff. Bad managers exist. Budgets are tight. Sure. But beyond those surface-level obstacles, there are structural mechanisms actively working against even the best performers.
First up: calibration sessions. You know those closed-door meetings where managers argue about ratings? They don’t start from your actual performance and work backward. They start with predetermined distributions and force everyone into buckets. Research shows top performers can have 10,000x impact compared to average ones [3]. But companies compress this into neat little bell curves anyway.
Take Capital One. Leadership decides beforehand that exactly 12% of employees must receive „Inconsistent“ ratings [4]. Doesn’t matter if everyone on your team crushed it. Someone’s getting pushed down because the quota demands it. Gergely Orosz puts it bluntly: calibrations „often defer to the loudest or highest-ranking manager in the room. They introduce bias and reward politics“ [5].
Then there’s the terminal level trap. Every major tech company has one. Google’s L5. Meta’s E5. Amazon’s SDE III. These are the levels where most engineers plateau indefinitely. Getting to Senior is built into the system. Getting beyond Senior is not [6].
The data backs this up. Pave research across 245,000 employees found companies growing headcount promote at 18.3% annually. Companies shrinking? Just 11.5% [2]. That’s a 60% difference based purely on whether your company is hiring or firing. Your performance didn’t change. The economic environment did.
And here’s my favorite structural catch-22: the „already operating at next level“ requirement. Most tech companies require you to demonstrate performance at the next level for 6-12 months before they’ll even consider promoting you [7]. But you need appropriate projects and scope to demonstrate that performance. Scope that often isn’t available without the title.
So you’re told you’re „ready“ and then you sit in a holding pattern for years waiting for the right opportunity to materialize. It’s like being told you can’t get a job without experience, but you can’t get experience without a job. Except the stakes are your career trajectory.
Promotion packets fail for predictable reasons
Even when engineers make it to committee review, their cases often collapse. The failure modes are predictable, and they reveal how the system penalizes certain types of valuable work.
Michael Lynch, a former Google engineer, learned this the hard way. He maintained a critical data pipeline. He fixed dozens of bugs. He documented tribal knowledge that kept the entire system running. At promotion time, he discovered none of it counted [8].
Why? Because he couldn’t prove any of it had a positive impact on Google. His bug discoveries actually caused the overall bug count to increase on paper. The time he spent helping teammates? That hurt his case. To the promotion committee, his colleague’s project looked like important cross-team coordination. Lynch appeared to be a „mindless peon whose work was so irrelevant that it could be pre-empted at a moment’s notice“ [8].
He did everything right from an engineering perspective. He did everything wrong from a promotion packet perspective.
Sean Goedecke, writing on StaffEng, states it directly: „You are very unlikely to be promoted to staff engineer for working on issues the company does not currently care about, even if you strongly believe they’re important“ [9].
Translation: your projects must align with what leadership currently values, not what objectively matters. You can spend a year preventing a catastrophic failure that would have cost millions. But if leadership’s current priority is shipping Feature X, your work is invisible.
Then there’s the peer review dynamic. Strong voices in calibration meetings set the tone for how your packet gets reviewed. Mixed reviews can swing either direction depending on who speaks first. One ex-Google engineer who served on promotion committees observed that committees often select neutral or positive reviewers to start discussions [10]. The first voice shapes the entire conversation.
If someone respected says „I’m not convinced,“ everyone else starts looking for reasons to agree. If someone respected says „this is clearly a strong case,“ suddenly all those edge cases seem less important.
Visibility gaps explain why great work stays invisible
A Harvard Business Review study of 240 senior leaders at a Silicon Valley tech company found something brutal: „the most important factor in promotion to their level wasn’t ability, leadership skills or even results; it was visibility“ [11].
Not performance. Visibility.
Think about the engineer who catches a bug before it causes an outage. The one who maintains documentation that prevents mistakes. Who answers questions that keep projects on track. What does this work look like in a metrics dashboard? Nothing. It looks like nothing happened.
Julia Evans, who popularized the „brag document“ concept, explains it well: „Some kinds of important work are more visible and memorable than others. If you didn’t do that work, something important might not have gotten done. But it’s easy to forget about“ [12].
Prevented disasters don’t show up on graphs. They don’t generate success metrics. They just quietly avoid failure. And in a promotion system optimized for visible wins, quiet competence gets overlooked.
Attribution bias makes this worse. People systematically over-credit those in formal authority while overlooking the supporting work that made success possible [13]. The person who presents the final solution gets credit. The person who unblocked them three weeks earlier gets forgotten.
The glue work trap particularly affects women. Tanya Reilly coined this term for the „less glamorous and often less-promotable work“ like onboarding, documentation, coordination, mentoring, and process improvement [14]. An American Economic Review study found women volunteer for this work 48% more often than men, and are also asked to do it more frequently [15].
One engineer did senior-level leadership work for two years. Glowing performance reviews. Then she was told she lacked „technical accomplishments“ for promotion [14]. Her manager never mentioned the work wasn’t promotable. Probably because they were glad someone was doing it.
Sponsorship determines outcomes more than performance
Here’s the distinction that matters: mentors talk TO you. Sponsors talk ABOUT you [16].
A Catalyst study found more women than men had mentors, yet men were promoted at rates 15% higher [17]. The differentiating factor wasn’t mentorship. It was having someone senior who would advocate in rooms where decisions happen.
According to Payscale, employees with sponsors earn up to 11.6% more than unsponsored colleagues [18]. Not because they’re better engineers. Because someone with power is actively creating opportunities for them.
What sponsors actually do [16]:
They recommend you for high-visibility projects before you ask. They defend your work when you’re absent from calibration meetings. They put their reputation on the line for your promotion. They make introductions to decision-makers you wouldn’t otherwise meet. Sometimes they create roles that didn’t exist before just to advance your career.
This isn’t about manipulation. It’s about recognizing that technical excellence alone doesn’t trigger promotions. Someone must actively advocate. And that advocacy requires relationships built before promotion season starts.
As one VP put it: „One person figured out how to ‚play the game‘ and the other didn’t“ [19].
The sponsorship gap explains why engineers with similar performance advance at wildly different rates. It’s not about who works harder. It’s about who has someone senior spending political capital on their behalf.
Timing and organizational chaos reset progress without warning
You can do everything right and still get screwed by forces completely outside your control.
Reorgs devastate promotion momentum. Research found employee attitudes toward „professional growth opportunities“ dropped 12% after reorganizations [20]. When managers change, you have to rebuild trust from scratch. Demonstrate your work again. Align with new expectations. The promotion clock effectively resets.
One engineer described going „from being a star employee on track for near term promotion to one who’s being undermined and diminished at every turn by a new mgr“ [21].
Project cancellations erase months of work. Michael Lynch experienced this three times at Google. Entire projects reassigned or killed with no credit for work completed [8]. Many companies do this regularly through „defrags“ where entire teams get cancelled. Years of positioning can evaporate overnight.
Economic conditions override individual performance entirely. During 2022-2024, tech companies froze promotions regardless of employee performance [22]. One analysis noted it’s „not uncommon to need a CFO or CEO to sign off any new hire even for junior roles. Promotions are hard to come by“ [23].
When interest rates rose from 0.25% to 5.5% between March 2022 and July 2023, the entire promotion landscape shifted [24]. Pre-pandemic expectations of „promotion every 1-2 years“ became post-2022 „frozen indefinitely“ [25]. Your performance didn’t change. Macroeconomic conditions did.
Build sponsorship deliberately, not accidentally
According to research: „You can’t go out and find sponsorship. It starts with your own performance. Build achievements so people are willing to put their reputations on the line“ [26].
But achievements alone don’t create sponsors. You have to actively cultivate relationships with senior leaders whose values align with yours.
Skip-level relationships matter as much as direct manager support. Your skip-level manager has broader scope and naturally understands organizational strategy. Build these relationships before you need them. Not to complain about your manager, but to understand priorities and share ground-level insights.
Rahul Pandey, ex-Meta Staff Engineer, notes: „Your skip manager is naturally equipped to answer questions around organizational strategy“ [27].
Find senior engineers who’ve been promoted internally and learn their path. Don’t ask generic career advice questions. Ask specifically: „What surprised you about your promotion?“ These conversations reveal the hidden process better than any documentation [28].
Use documentation as a strategic weapon
Will Larson recommends writing your promotion packet years before you expect promotion. Use it as a roadmap, not just a capstone document [29].
Template categories should include: Staff projects with linked design docs, scope of influence, glue work with quantified impact, technical skills demonstrated, and identified gaps with plans to address them. Review this with your manager during regular 1:1s [29].
Julia Evans‘ brag document framework suggests updating every two weeks with: goals, projects with impact metrics, collaboration and mentorship, design documents, company building like recruiting and process improvement, and skills learned [12].
Share this with peer reviewers before they write your review. Share it with new managers immediately after transitions. Don’t make people guess what you’ve accomplished.
Tanya Reilly’s advice on glue work is direct: „Either stop doing glue work, or frame it so it’s recognized as technical leadership“ [14].
For junior engineers, limiting glue work to focus on technical skills may be the right career move, even if it feels unhelpful. At senior levels, glue work is expected. But only if you explicitly document it and make sure it’s valued.
Evaluate your situation honestly
Ask direct questions that reveal the truth. Instead of vague career conversations, ask specifically [30]:
„Will I get promoted next round?“ „What specific work do I need to demonstrate?“ „What’s the realistic timeline?“
If your manager can’t answer concretely, that’s information.
Assess structural constraints before investing years. Does your company have Staff-level headcount for your team? Does leadership view your domain as promotable? Have similar profiles been promoted before? Is there a „glass ceiling“ where Senior is effectively terminal [31]?
If multiple Staff engineers already fill available slots, internal promotion may be structurally impossible regardless of your performance.
Set deadlines for career conversations. One expert suggests: „Two weeks is too short. Eight months is probably too long“ [32].
If you’ve had the conversation and received platitudes, establish when you’ll check in again. If no concrete progress materializes after that deadline, change strategy.
Recognize when changing companies is the rational choice
It’s often easier to be hired as Staff than promoted to it. StaffEng research found roughly one-third of Staff engineers changed companies to attain the title [33]. Women disproportionately need to take this path, with roughly half of women interviewed having to change companies to attain the Staff title [34].
If your current situation has structural barriers, external opportunities may offer faster advancement.
External offers should only be used if you’ll actually leave. Experts uniformly advise against getting offers solely for negotiation leverage. Studies show counteroffers rarely work long term. People who accept them often leave within a year anyway, and the trust damage is permanent [35].
But keeping your skills current and interviewing occasionally provides valuable market data.
Workers who switch jobs see 15-16.5% median pay increases versus 5-8% for those who stay [36]. The math often favors mobility, particularly when internal advancement is blocked. Job switchers saw salary increases of 10% in March 2024, the highest rate of growth since July 2023, while job stayers saw annual wage gains of just 5.1% [37].
The Pragmatic Engineer frames it well: „If your primary goals are about professional growth, even if you are not promoted, you will have gained valuable skills. Promotions will be a hit or miss. But no one can take away your growth from you“ [38].
Conclusion
The promotion system in tech is not primarily a meritocracy. It’s a resource allocation mechanism with artificial scarcity, political dynamics, and significant randomness.
High performers who assume „the work speaks for itself“ systematically lose to those who understand and work the meta-game of visibility, documentation, sponsorship, and strategic positioning.
This doesn’t mean gaming the system cynically. The engineers who advance fastest genuinely do excellent work. But they also ensure that work is visible, documented, aligned with company priorities, and actively championed by senior sponsors. They understand that promotion readiness and actual promotion are different states, often separated by timing, luck, and organizational chaos.
Charity Majors captures the mindset shift required: „Fuck the whole idea that only managers get career progression. And fuckkkk the idea you have to choose a ‚lane‘ and grow old there“ [39].
Career advancement in tech requires treating your career as something you actively manage, not something that happens to you. That means building relationships before you need them. Documenting work as you do it. Aligning with organizational priorities even when you disagree. And being willing to leave when structural barriers make internal advancement impossible.
The gamble is real. But understanding the odds and which levers you can actually pull transforms it from pure chance into calculated risk.
References
[1] Pave. „Annual Promotion Rates by Level: What is The Optimal Target Promotion Rate?“ https://www.pave.com/blog-posts/annual-promotion-rates-by-level
[2] Pave. „Staff Engineer Research – 245,000 employees dataset.“ Referenced in The Pragmatic Engineer newsletter.
[3] Kane, Narraway. „Performance distribution in knowledge work.“ Referenced in calibration discussions.
[4] Capital One internal performance calibration processes. Referenced in Gergely Orosz’s „The Software Engineer’s Guidebook.“
[5] Orosz, Gergely. „The Software Engineer’s Guidebook.“ https://blog.pragmaticengineer.com/
[6] Levels.fyi. „Terminal Levels at Major Tech Companies.“ https://www.levels.fyi/
[7] StaffEng. „Getting the title where you are.“ https://staffeng.com/guides/getting-the-title-where-you-are/
[8] Lynch, Michael. „Why I Quit Google to Work for Myself.“ https://mtlynch.io/why-i-quit-google/
[9] Goedecke, Sean. „How I got promoted to staff engineer twice.“ https://www.seangoedecke.com/staff-engineer-promotions/
[10] Anonymous Google engineer serving on promotion committees. Referenced in blind and levels.fyi discussions.
[11] Harvard Business Review. „How to Become More Visible at Work.“ https://hbr.org/2023/08/how-to-become-more-visible-at-work
[12] Evans, Julia. „Get your work recognized: write a brag document.“ https://jvns.ca/blog/brag-documents/
[13] McKinsey. „70% of promotion decisions are shaped by how someone is perceived.“ Referenced in HBR visibility research.
[14] Reilly, Tanya. „Being Glue.“ https://noidea.dog/glue
[15] American Economic Review. „Gender differences in non-promotable task allocation.“ Study showing women volunteer 48% more for glue work.
[16] Catalyst. „Sponsorship and Mentoring.“ https://www.catalyst.org/research/ask-catalyst-express-sponsorship-and-mentoring/
[17] Catalyst. „Sponsoring Women to Success“ study (2011). Foust-Cummings, Dinolfo, and Kohler.
[18] Payscale. „The ROI of Sponsorship in the Workplace.“ Referenced in Project Include research.
[19] Anonymous VP quoted in career advancement discussions.
[20] LeadDev. „How to make reorgs less terrible.“ https://leaddev.com/culture-engagement-motivation/how-make-reorgs-less-terrible
[21] Quora. „What happens to promotion after manager change?“ https://www.quora.com/career-development discussions.
[22] The Pragmatic Engineer. „Tech hiring trends in 2025: the 4 big shifts shaping the tech job market.“ https://ravio.com/blog/tech-hiring-trends
[23] Industry analysis of tech promotion freezes 2022-2024. Referenced in multiple tech employment reports.
[24] Federal Reserve. Interest rate hikes March 2022 – July 2023, raising rates from 0.25% to 5.5%.
[25] The Pragmatic Engineer. „The end of 0% interest rates: what the new normal means for software engineers.“ https://newsletter.pragmaticengineer.com/p/zirp-software-engineers
[26] Glassdoor. „How to Find a Sponsor at Work.“ Career advancement research.
[27] Pandey, Rahul. Ex-Meta Staff Engineer career guidance. Referenced in various tech career podcasts and articles.
[28] StaffEng. „Frequently Asked Questions.“ https://staffeng.com/faq/
[29] Larson, Will. „Promotion packets.“ https://staffeng.com/guides/promo-packets/
[30] Career advancement best practices. Multiple sources including Ask a Manager and tech career coaches.
[31] StaffEng. Analysis of structural barriers to Staff-level promotion.
[32] Workopolis. „How long should you wait for a promotion?“ Career timing advice.
[33] StaffEng. „Getting the title where you are.“ Research on internal vs external Staff promotions.
[34] StaffEng. Gender disparities in Staff engineer promotions. https://staffeng.com/guides/getting-the-title-where-you-are/
[35] Multiple HR research sources on counteroffer acceptance and retention rates.
[36] Pew Research Center. „Majority of U.S. Workers Changing Jobs Are Seeing Real Wage Gains.“ https://www.pewresearch.org/social-trends/2022/07/28/majority-of-u-s-workers-changing-jobs-are-seeing-real-wage-gains/
[37] Statista. „Switching Jobs No Longer Pays Off Like It Used To.“ https://www.statista.com/chart/31032/median-year-over-year-change-in-annual-pay-in-the-us/
[38] The Pragmatic Engineer. „A Work Log Template for Software Engineers.“ https://blog.pragmaticengineer.com/work-log-template-for-software-engineers/
[39] Majors, Charity. Twitter/X posts on career progression and Staff engineering. CTO at Honeycomb.io.