Your Meritocracy Isn’t Working — Here is How to Fix It

The research has existed since 1998. The conditions have never been created. And people’s careers are the cost of that gap.

Yet you keep using that word. Every talent review. Every promotion decision. Every all-hands where leadership explains why the best people are rising. Every DEI statement reassuring employees the system is fair because it’s based on merit.

Meritocracy.

Here’s the thing: the word has a definition. A precise, researchable, testable definition. And the systems required to make it real have been documented in peer-reviewed research since 1998.

Frank Schmidt and John Hunter published the validity evidence in 1998. Claudia Goldin and Cecilia Rouse published the blind audition data in 2000. Emilio Castilla and Stephen Benard documented the paradox of meritocracy in 2010. The blueprint has been available for twenty-five years. The conditions it describes have not been built in a single large organization at scale.

So when you say you have a meritocracy, one of two things is true. Either you haven’t read the research, in which case you’re using a word you don’t fully understand. Or you have, and you built something different anyway, in which case you’re using a word that doesn’t apply.

Neither is a comfortable position for an organization that has been telling its employees their outcomes reflect their merit.

The problem isn’t the concept. Merit, advancing people based on what they contribute, what they know, how they perform, is not just defensible. It’s the most equitable organizing principle available. The problem is that organizations have claimed the word while systematically refusing to build the architecture. And the people paying for that gap aren’t the executives debating talent philosophy. They’re the employees whose careers are shaped by evaluation systems that were never as fair as advertised.

This post does two things. First, it explains exactly what the word means and what the research says genuine meritocracy requires. Second, it describes what building it actually looks like, specifically enough that any organization can compare it against what they have and see the gap clearly.

That gap has a name. It isn’t ignorance. It’s a choice. And choices can be changed.

The Origin Story Nobody Talks About

When sociologist Michael Young coined the term ‘meritocracy’ in his 1958 satirical novel The Rise of the Meritocracy, he wasn’t coining a compliment. He was writing a warning. Young imagined a dystopian future where intelligence and effort replaced birth and blood as the basis for social stratification. His argument was that this wasn’t progress. It was a more insidious form of inequality, because it made the hierarchy feel deserved.

The great irony is that the word Young invented as a pejorative was immediately adopted enthusiastically as a positive organizing principle by the very institutions he was critiquing. It does extraordinary rhetorical work. It validates success, explains failure, and makes the entire distribution of power feel like the natural result of effort and talent rather than a social construction shaped by access, networks, race, gender, and class.

Young’s warning has proven prescient. The belief in meritocracy doesn’t reduce inequality. In many measurable cases, it makes it worse.

What the Research Actually Says

Let’s be specific, because vagueness is where this word goes to hide.

In 2010, Emilio Castilla and Stephen Benard published their paradox of meritocracy study in the Administrative Science Quarterly. In organizations that explicitly promoted a meritocratic culture, managers were more likely to award higher bonuses to male employees than to equally performing female employees. Male employees received bonuses averaging $46 higher than identically-rated female counterparts specifically in the condition where meritocracy was emphasized as a value. That paper is fifteen years old. Most organizations still haven’t redesigned their reward systems in response to it.

The mechanism is psychological. When people believe a system is fair, they become less vigilant about their own biases. The belief in fairness functions as moral license. Organizations most committed to the rhetoric of meritocracy are often the least equipped to detect when it’s failing.

Van Dijk et al.’s 2020 CSI-W model adds the structural dimension. Workplaces don’t merely reflect existing inequality. They amplify it through nine interlocking mechanisms. Initial differences in opportunities shape performance, which shapes subsequent opportunities, in a compounding cycle that makes early advantage look like inherent talent. Running for years, that math produces career gaps that look like merit differences. They are not.

This is how organizations launder inequality. Not through overt discrimination, but through the quiet mathematics of accumulated advantage dressed in the language of fairness.

The Power Problem: A Structural Observation, Not a Moral Accusation

For an organization to build a genuine meritocracy, it would need to interrogate the conditions that produced its current leadership. Think about who holds power in most large organizations: disproportionately male, disproportionately white, disproportionately from higher socioeconomic backgrounds, graduates of a small number of elite institutions. This is documented in census surveys, executive compensation studies, and board composition analyses repeated year after year.

Did those people get there purely on merit? Or through a combination of genuine talent and access to networks, education, early-career opportunities, mentors who saw themselves in them, a professional culture built for people like them? The honest answer is both. Their success is real. It was also scaffolded by structural advantages that were never equally available.

A genuine meritocracy would make that visible. Visibility threatens legitimacy. Sociologist Jo Littler describes this as circular validation: success proves the system works, and the system working proves that success is deserved. The logic is self-sealing. It explains why the blueprint has sat largely unused for twenty-five years.

The Tools Meritocracy Uses to Protect Itself

Constructed Criteria

Research by Uhlmann and Cohen (2005) shows that evaluators define merit criteria after they’ve already decided who they prefer, unconsciously reshaping what good looks like to favor their preferred candidate. Unless criteria are locked before any individual is assessed, they aren’t criteria. They’re rationalizations.

Opportunity Hoarding

Organizations repeatedly invest development resources in the same small group of already-recognized employees. Those employees accumulate experience and credibility at a compounding rate that eventually reads as merit. Van Dijk et al. identify this as a structural feature of most talent systems, not a coincidence.

The Language of Objectivity

When a manager rates someone 3 out of 5 on ‘executive presence,’ they aren’t reporting an objective measurement. They’re reporting a social judgment inflected by race, gender, disability status, and cultural background. The numerical format obscures the subjectivity. The apparent objectivity becomes a shield.

Meritocracy as Explanation for Inaction

If the system is already meritocratic, then intervention is not just unnecessary. It’s anti-meritocratic. This framing converts the absence of corrective action into a principled position. It is a choice dressed as a principle.

When ‘Culture Fit’ Became Meritocracy’s Most Useful Disguise

Culture fit is the concept that allows every other bias to survive the formal screening process. It’s the moment after all structured criteria have been applied when a decision-maker says ‘I just don’t think they’d fit here,’ and the candidate is out. Bias with a veneer of reasonableness.

Lauren Rivera’s research at Northwestern, published in the American Sociological Review, documented this across elite professional service firms. Hiring decisions were driven heavily by assessments of ‘polish’ and ‘fit’ that were, in practice, evaluations of class background and participation in expensive extracurricular activities. The firms believed they were selecting for excellence. They were selecting for familiarity.

A genuine meritocracy retires ‘culture fit’ entirely. In its place: specify which attributes are actually required for success in a given role, distinguish those from attributes merely familiar to people already in power, and document the difference before any candidate is evaluated.

The Diversity-Meritocracy False War

The framing that diversity and meritocracy are opposing forces isn’t just wrong. It’s precisely backwards.

Diverse evaluation panels produce more accurate assessments of candidate quality than homogeneous ones by reducing the shared blind spots homogeneous groups systematically carry. The exclusion of women, people of color, working-class candidates, and people with disabilities from senior leadership pipelines isn’t evidence of insufficient merit. It’s evidence of structural barriers that prevent merit from becoming visible, developed, and recognized. Removing those barriers isn’t trading quality for diversity. It’s removing quality-reducing distortions from a system currently misallocating talent at scale.

Any leader who frames equity work as a concession to external pressure rather than a precondition for genuine meritocracy hasn’t grappled with what the research says. Or has grappled with it, found it uncomfortable, and chosen the narrative that better protects the current distribution of power.

The framing that diversity and meritocracy are opposing forces isn’t just wrong. It’s precisely backwards.

Diverse evaluation panels produce more accurate assessments of candidate quality than homogeneous ones by reducing the shared blind spots homogeneous groups systematically carry. The exclusion of women, people of color, working-class candidates, and people with disabilities from senior leadership pipelines isn’t evidence of insufficient merit. It’s evidence of structural barriers that prevent merit from becoming visible, developed, and recognized. Removing those barriers isn’t trading quality for diversity. It’s removing quality-reducing distortions from a system currently misallocating talent at scale.

Any leader who frames equity work as a concession to external pressure rather than a precondition for genuine meritocracy hasn’t grappled with what the research says. Or has grappled with it, found it uncomfortable, and chosen the narrative that better protects the current distribution of power.

The following is not a wish list. It’s the minimum viable architecture for an organization that wants to use the word ‘meritocracy’ without misusing it. Each element is grounded in research. None is prohibitively expensive. All have been systematically avoided by most organizations for decades

1. Audit the Inputs Before You Evaluate the Outputs

You cannot evaluate merit neutrally when the inputs to performance have been unequally distributed. Start with an honest inventory of who received developmental inputs: stretch assignments, high-visibility nominations, leadership program enrollment, senior sponsorship. Break it down by demographic group. If stretch work is going disproportionately to employees already overrepresented in your leadership pipeline, your performance data is a record of who was given the chance. Not a neutral record of merit.

2. Lead With Hard Metrics, Not Behavioral Observations

Behavioral criteria like ‘demonstrates executive presence’ or ‘leads with influence’ are social judgment in numerical costume. The rubric creates the illusion of standardization. The bias operates underneath it.

Goldin and Rouse’s blind orchestra audition study showed a 50% increase in women advancing when visual information was removed. Schmidt and Hunter’s meta-analysis showed work samples dramatically outperform narrative behavioral assessments in predictive validity. The research case for output-first evaluation has existed for twenty-five years.

The detailed framework for how to build this, including criteria by role and how to weight each evaluation tier, is in the free Meritocracy Audit guide linked at the bottom of this post.

3. Lock In Criteria Before Any Individual’s Name Is Attached

Castilla’s MIT Sloan research found that requiring managers to provide specific, criteria-grounded justifications for ratings, reviewed by someone other than the manager, is one of the most effective bias-reduction interventions available. No rating without a written rationale tied to pre-committed criteria. No nomination without documented justification. No promotion decision without a specific performance evidence log.

Criteria defined after evaluation begins are not criteria. They are rationalizations.

4. Separate Potential from Performance, Then Audit Both

Potential ratings are more demographically skewed than performance ratings. They reflect not what evaluators observe, but what they imagine. Imagination is deeply inflected by similarity bias and stereotypes about who looks like a future leader. If potential ratings correlate with race or gender after controlling for performance, your potential assessment is measuring something other than potential.

5. Break the Nomination Monopoly

When access to opportunity is controlled by one manager, access skews toward whoever is most visible and socially compatible. Use transparent nomination processes with pre-defined criteria. Diverse panels for significant opportunities. Blind review of application materials where feasible. Track who has been nominated three or more times in the past year and who has never been nominated once.

6. Calibrate Across Managers With Demographic Transparency

Mandatory calibration sessions where managers defend ratings in a group context, with explicit attention to demographic patterns. The goal isn’t consensus. It’s visibility. If one manager’s ratings show systematic demographic skew against comparable performance evidence, that pattern must become visible before it calcifies into compensation and promotion disparities.

7. Link Manager Accountability to Equity Outcomes

If managers aren’t accountable for equity outcomes, including opportunity distribution, demographic representation in nominations, and rating patterns, those outcomes won’t change. These metrics should carry weight in manager evaluation equal to revenue or headcount targets. A manager whose team shows systematically skewed opportunity distribution produced that imbalance.

8. Publish What You Find

Transparency is the mechanism that makes accountability durable. Internal findings that stay internal create a conversation among people already motivated to minimize what the data shows. Your employees already know the opportunity distribution isn’t equal. Publishing the data doesn’t reveal something new. It demonstrates that leadership is finally willing to see what everyone else has been living.

The Thing Nobody in Power Wants to Admit

The research documenting what genuine meritocracy requires has been available since 1998. Schmidt and Hunter. Goldin and Rouse. Castilla and Benard. Van Dijk. These appear in the American Economic Review, Administrative Science Quarterly, Psychological Bulletin — the most-read journals in their fields. Any organization with an HR function and an executive who attended a business school in the last twenty years has had access to this evidence.

And yet the systems those papers describe have not been built. Not at scale. Not in a single large organization that can point to documented, longitudinal, externally verified outcomes and say: we built the conditions, and here is what they produced.

That is a choice. Made repeatedly. By people with the authority and the resources to make a different one.

Many of the leaders who most enthusiastically endorse meritocracy are beneficiaries of a system that was never meritocratic. Schools more accessible to families with money. Organizations where people who looked like them were already in leadership. Mentors who invested in them partly because of affinity. Room to fail and recover that employees from underrepresented backgrounds are rarely afforded.

None of that makes them fraudulent people. But it does mean their career trajectories aren’t clean evidence that the current system rewards the best people. The organizational incentive to build a genuinely meritocratic system is structurally low for precisely the people who hold the power to build it. It doesn’t require bad intentions. It just requires being human. And it explains why twenty-five years of available evidence hasn’t produced twenty-five years of organizational change.

So. About That Word.

This is where most posts of this kind land softly. They catalogue the problem, note it’s complex, and suggest that leaders ‘reflect’ on their practices.

This is not that post.

You’ve been using the word. For years, in most cases. In talent reviews, in offer letters, in the presentations leaders give about their culture. And now you know what it actually requires: output-first metrics constituting the majority of evaluation weight. Pre-committed criteria defined before any individual’s name is attached to an assessment. Work samples evaluated blind where possible. Cross-manager calibration with demographic transparency. Opportunity auditing that tracks who gets developmental inputs, not just who gets good ratings. Manager accountability tied to equity outcomes. And the organizational courage to publish what you find.

That is what the word means. That is what twenty-five years of research says is required to make the outcomes reflect the claim.

The cost of the gap isn’t abstract. It’s the people inside your organization not doing their best because they don’t believe your system works.

So here’s the test. Not the suggestion.

If you use the word ‘meritocracy’ to describe how your organization makes talent decisions, that claim is auditable.

Here’s the minimum viable evidence that you’re using it accurately:

If you can’t produce that evidence, you’re not a meritocracy. You’re an organization that keeps using the word while not building the thing.

No organization at scale has fully built this yet. The blueprint has existed for twenty-five years. The question was never whether it was possible. The question is whether you’ll be the one who actually builds it.

You keep using that word. Now you know what it means. The only question left is what you’re going to do about it.

Performedi

Performedi builds the data infrastructure and decision architecture to close the gap between meritocracy as a word and meritocracy as a practice. Opportunity audits. Output-first evaluation frameworks. Accountability systems that make the invisible visible. If you’re ready to use the word accurately, we’re ready to help you build what it requires.

References

Castilla, E. J., & Benard, S. (2010). The paradox of meritocracy in organizations. Administrative Science Quarterly, 55(4), 543-576.

van Dijk, H., Kooij, D., Karanika-Murray, M., De Vos, A., & Meyer, B. (2020). Meritocracy a myth? A multilevel perspective of how social inequality accumulates through work. Organizational Psychology Review, 10(4), 240-269.

Uhlmann, E. L., & Cohen, G. L. (2005). Constructed criteria: Redefining merit to justify discrimination. Psychological Science, 16(6), 474-480.

Goldin, C., & Rouse, C. (2000). Orchestrating impartiality: The impact of ‘blind’ auditions on female musicians. American Economic Review, 90(4), 715-741.

Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124(2), 262-274.

Ridgeway, C. L. (1991). The social construction of status value: Gender and other nominal characteristics. Social Forces, 70(2), 367-386.

Rivera, L. A. (2012). Hiring as cultural matching: The case of elite professional service firms. American Sociological Review, 77(6), 999-1022.

Young, M. (1958). The rise of the meritocracy. Thames and Hudson.

Littler, J. (2017). Against meritocracy: Culture, power and myths of mobility. Routledge.