The Data Black Professionals Already Know
Your results were strong. Your output was measurable. And your review did not reflect either. Performance review bias is not a feeling. It is a documented, data-evidenced mechanism of the Excellence Tax. Here is what the research shows.
What Performance Review Bias Actually Is
Performance reviews are presented as objective. They are not.
They are a distillation of subjective impressions, cultural assumptions, and proximity biases accumulated over months of working relationships. For Black professionals, those impressions are shaped by a set of racialised filters that most organisations have never audited and most managers have never been asked to examine.
This matters because the review is not just a conversation. It is a document. It follows you. It informs promotion decisions, pay decisions, and the informal judgements made about your potential in rooms you are not in. When the process that produces it is structurally biased, the consequences are not contained to a single meeting. They travel.
Performance review bias is not about one manager having a bad day. It is not about a personality clash. It is about a system that was designed around a particular kind of professional, a particular set of cultural norms, and a particular standard of what “high performance” looks and sounds like. That system has not been redesigned. It has been given a new name. And the people it was never designed to see accurately are still being assessed by it.
The Research Data
The Excellence Tax research draws on data from over 1,000 Black professionals across the UK. The findings on performance review bias are among the most consistent in the entire dataset.
Black professionals consistently rated their own performance higher than their managers rated them. When we cross-referenced those self-ratings with objective output metrics, the Black professionals’ assessments were the more accurate of the two.
This is not a confidence gap. This is not overestimation. The managers were wrong. The professionals were right. And the ratings that went on record were the managers’.
The promotion data reinforces this. Black professionals at equivalent tenure and equivalent performance ratings were promoted at a rate 34% lower than their non-Black counterparts. Same role. Same years of service. Same output scores. Different outcomes.
34% lower promotion rate. Not explained by performance. Not explained by tenure. Explained by the process.
The language data tells the same story from a different angle. When we analysed the specific words used in performance reviews, a clear pattern emerged. Black professionals received descriptors like “abrasive”, “difficult”, “intense”, “not quite the right fit”, and “hard to manage”. Their non-Black colleagues, assessed on the same behaviours in the same environments, received “assertive”, “driven”, “direct”, and “strong performer”.
The behaviour was not the variable. The race of the person performing it was.
What It Looks Like in Practice
Performance review bias does not always arrive as an obviously unfair rating. It is often quieter than that. Here is what it looks like across three common scenarios.
The review that does not match the year
You have delivered. The project landed on time. The client relationship held. You stepped in when a colleague was absent and absorbed the additional workload without complaint. You came into your review prepared, with a clear account of what you had done and the outcomes it produced.
Your rating is “meets expectations”.
Your manager struggles to give specific examples when you ask why it is not higher. There is some general commentary about “continuing to develop” and “building relationships across the business”. Nothing concrete. Nothing that maps onto the year you just had.
Your non-Black colleague, who delivered a comparable project with comparable outcomes, received “exceeds expectations”. You know this because they told you.
The gap is not explained by the work. It is not explained by the evidence in the room. It is explained by the impressions that accumulated over twelve months in an environment where you were held to a standard that was never written down and never applied equally.
The language trap
Your review contains the following under development areas: “Would benefit from being more considered in how she presents her views to senior stakeholders. Can sometimes come across as quite certain, which closes down discussion.”
You ask for an example. Your manager cannot give one that is specific enough to be useful. You ask what “more considered” means in practice. The answer is vague. You ask how this maps onto the work you delivered. It does not, clearly.
What you have been given is feedback that is impossible to act on and impossible to disprove. It is subjective enough to be deniable and specific enough to be damaging. It will sit in your record. It will be referenced at the next calibration meeting. And it will be used to explain, to people who were not in the room when it was written, why you are not quite ready for the next level.
This is the language trap. The feedback is not designed to help you improve. It is designed to create a paper trail that justifies an outcome that has already been decided.
The sponsorship gap
Your manager tells you that you are doing well. Your clients rate you highly. Your peer feedback is strong. You are, by every visible measure, performing at a level that should translate into progression.
But nothing moves.
You are not being put forward for the stretch assignments that build visibility. You are not being mentioned in the conversations where the next round of promotions is being shaped. You are not being advocated for in the calibration meetings where your manager’s assessment of you meets the assessments of other managers and a collective decision is made about who is ready.
Sponsorship, the active act of going into a room and putting your name forward, is not evenly distributed. Our research finds that Black professionals are significantly less likely to have a senior advocate who does this for them, regardless of their performance level. The result is a career that is performing well on paper but not moving. Not because of what you are delivering. Because of who is not speaking up for you when you are not in the room.
“I kept being told I was doing brilliantly. I kept waiting for that to mean something. Eventually I understood that doing brilliantly and being moved forward were two separate decisions. And only one of them involved my work.”
The Psychological Impact
There is a specific kind of damage that comes from excellence that goes unrecognised. It is different from the damage of failing. It is, in some ways, harder to process.
When you fail and receive a poor review, there is at least a logic you can work with. When you deliver and receive a poor review, the logic breaks. The thing you were told would protect you, strong performance, turns out not to protect you. The effort you put in turns out not to be the variable. And the question that follows, quietly, persistently, is: if this is not about the work, what is it about?
That question is destabilising. Our research finds that Black professionals who have experienced repeated performance review bias show a specific pattern: a gradual decoupling of effort from expectation. They continue to deliver. They often continue to deliver at a very high level. But the belief that delivering will lead to recognition, the motivational contract that sustains most professionals through difficult periods, begins to erode.
This is not disengagement. It is a rational adjustment to an environment that has repeatedly shown that the reward does not follow the effort. And it carries a significant cost. Not just for the professional, but for the organisation that is extracting the performance without providing the recognition that would sustain it.
The longer-term psychological effects of this sustained experience connect directly to racial battle fatigue and the weathering effect. Both are explored in detail in the Workplace Healing hub.
What the Professional Can Do With This Knowledge
Naming performance review bias accurately is the first and most important thing. Not because naming it changes the system, but because naming it correctly means you stop explaining it incorrectly. You stop looking for the flaw in your work. You stop trying to adjust your performance in response to a process that is not actually responding to your performance.
Build your own evidence base throughout the year. Do not wait for the review cycle to construct the account of what you have delivered. Document outputs as they happen. Keep records of client feedback, measurable results, and instances where you went above the scope of your role. When the review arrives, you have a case built on specifics, not impressions. This does not guarantee a fair outcome. But it changes the nature of the conversation and creates a record that exists independently of your manager’s account.
Understand the difference between a manager problem and a system problem. Some performance review bias is localised. It is one manager, in one team, applying one set of preferences. That is addressable, though not easily. Some performance review bias is systemic. It is consistent across managers, across teams, across review cycles. When it is systemic, the solution is not to find a better way to present yourself to the same process. The solution is to understand what the process is and is not capable of producing, and to make decisions accordingly.
Know what a paper trail looks like and why it matters. If you are considering raising a formal concern, or if you think you may need to at some point, the strongest position you can be in is one where you have documented comparisons. Not just your own reviews, but the pattern. What language was used about you. What language was used about colleagues in equivalent roles. What the outcome differential was. This kind of evidence is what moves a grievance from a personal complaint to a structural one.
What Organisations Must Do Differently
The performance review process will not fix itself. It requires deliberate structural intervention. Four changes are necessary.
Audit review ratings by race, tenure, and output metrics. This is the starting point. Pull the data. Cross-reference performance ratings with measurable output data and with the demographic profile of the professional being reviewed. If your organisation has been running annual reviews for a decade, you have ten years of data that will tell you, with clarity, whether your process is producing equitable outcomes. Most organisations have not looked. The ones that have looked have found what the research predicts.
Standardise the evidence required for ratings.
A rating of “exceeds expectations” should require the same quality of specific, evidenced examples as a rating of “does not meet expectations”. Subjective impressions should not be sufficient justification for any rating, positive or negative. When managers are required to provide the same level of evidence regardless of the rating direction, the volume of vague, impressionistic, and racialised feedback decreases significantly.
Address the sponsorship gap directly.
Calibration meetings, the sessions in which managers collectively decide who is ready for progression, need to be restructured. Who is being put forward, and on what basis. Whether the evidence being used is consistent across candidates. Whether the professionals who lack a vocal advocate are being assessed on the same terms as those who have one. The sponsorship gap does not close through awareness. It closes through process accountability.
Separate style preferences from performance assessments. Communication style, cultural difference, and personality are not performance criteria. If they are appearing in performance reviews as barriers to progression, that is a process failure. Review frameworks need to define, explicitly, what constitutes a performance assessment and what constitutes a personal preference. And managers need to be trained to recognise the difference.
One Mechanism of Fifteen
Performance review bias is one of fifteen mechanisms through which the Excellence Tax operates. It sits alongside tone policing, microaggressions, the double performance standard, and the sustained labour of code-switching as part of a pattern of extraction that our research has now documented at scale across UK workplaces.
The compounding effect is significant. When the review process is biased, and the language used to describe you is racialised, and the sponsorship that would move you forward is withheld, and the cognitive load of navigating all of this is unaccounted for in how your capacity is assessed, the result is a professional whose excellence is being extracted and whose record is being written by a process that was never designed to see them accurately.
That is not an individual problem. It is an organisational one. And it has a cost that organisations are currently paying without understanding what they are paying for.
About the Research
Your Organisation Is Already Paying This Cost: The Excellence Tax™ is not a future risk. It is a current expenditure — in talent, in productivity, in trust. If you are a leader who wants to understand and address what your data is not telling you, let’s talk
Book a Briefing with NatashaRead Next
- Back to the Corporate Survival hub
- Previous article: Tone Policing
- Next article: Microaggressions
Natasha Williams
1 comment