Here’s a situation that will be familiar to a lot of neurodivergent employees.
You disclose a disability. You go through the occupational health process. A report lands on your employer’s desk recommending reasonable adjustments: maybe support with workload planning, structured check-ins with your manager, adjustments to how your performance is reviewed. Months pass. Nothing changes. No one mentions the report again. Your manager doesn’t bring it up. You don’t get the support. And then, at the end of the year, you get your performance rating. It’s low. Tied to that rating is a salary increase of zero, while your colleagues receive 5%.
You were set up to fail, and then penalised for failing.
This isn’t an edge case. This is the system working exactly as designed, or rather, exactly as not designed. Because most employers have never stopped to think about what happens when the duty to make reasonable adjustments meets the performance review process. And the people who pay the price for that blind spot are disabled and neurodivergent employees.
The law is clear. The practice isn’t.
The Equality Act 2010, s 20 places a duty on employers to make reasonable adjustments where a disabled employee is placed at a substantial disadvantage by the employer’s provisions, criteria, or practices. Failure to do so is disability discrimination. That’s not a grey area. It applies to every aspect of employment, including performance management.
Section 15 goes further. It says an employer discriminates against a disabled person if they treat them unfavourably because of something arising in consequence of their disability, unless the employer can show the treatment was a proportionate means of achieving a legitimate aim. The causal link can be loose. In Pnaiser v NHS England, the Employment Appeal Tribunal established that the “something arising” need not be the main or sole cause; it must merely have a significant, more than trivial, influence. The Court of Appeal in City of York Council v Grosset confirmed that the employer does not even need to know of the causal connection. This is an entirely objective question for the tribunal.
So when an employer knows an employee is disabled, has a report recommending adjustments, doesn’t implement those adjustments, and then rates the employee poorly on performance, the chain is short and the law is not sympathetic to the employer’s position.
Acas is explicit on this point. Their guidance on neurodivergent employees and performance management states that before using a formal performance, conduct, or capability procedure, an employer must make sure they’ve done all they reasonably can to support an employee. Using a formal procedure without exploring support first, they warn, can cause unnecessary harm and may constitute discrimination.
And yet it happens constantly.
The adjustment gap
The Neurodiversity Index Report 2024, published by the City & Guilds Foundation in partnership with Do-IT Solutions, surveyed 573 UK employees and 93 organisations. It found that 20% of neurodivergent workers were still waiting for promised adjustments to be made. A peer-reviewed study by researchers at UCL’s Centre for Research in Autism and Education found that approximately one-third of autistic employees felt unable to discuss their adjustment needs at all. Of those who did request adjustments, over a quarter were refused and more than one in ten found the adjustment was poorly implemented.
The 2025 Neurodiversity Index, surveying a significantly larger sample of 1,385 respondents, found that a third of neurodivergent employees were dissatisfied with the support they received. 35% had no onboarding support when starting their jobs. 37% of managers had received no training on neurodiversity whatsoever.
This is the gap. Adjustments are recommended but not implemented. Managers are untrained. Support falls through the cracks. And none of this shows up in the performance review. Only the consequences do. The employee’s difficulty with unstructured work, their struggle with communication norms they were supposed to receive help navigating, their missed deadlines that structured check-ins were supposed to prevent. All of this gets recorded as underperformance. The employer’s failure to act simply doesn’t appear in the form.
The financial penalty
When performance ratings determine pay, a biased or unsupported review process doesn’t just affect morale. It creates a measurable financial penalty for being disabled.
The scale of this is already documented. The Office for National Statistics reported a UK disability pay gap of 12.7% in 2023, meaning disabled workers earned, on average, £2 less per hour than non-disabled workers. For autistic workers specifically, the gap was 27.9%: the widest of any disability type. The TUC’s 2025 analysis put the gap at 15.5%, equivalent to over £4,000 per year for a full-time worker. The national picture is clear. But unlike the gender pay gap, which large employers are legally required to report, there is no obligation on individual employers to measure or disclose their own disability pay gap. The data exists at scale. The accountability doesn’t.
Consider a straightforward scenario. Two employees doing the same role. One is neurotypical and receives no adjustments because none are needed. They get a “competent” rating and a 5% salary increase. The other is neurodivergent, was recommended adjustments that were never implemented, performs in a context of zero structured support, and receives a “partial” rating and 0%. After one year, the gap is 5%. After five years of compounding, it’s significant. After a career, it’s tens of thousands of pounds.
The cruelty is in the structure. The employee didn’t fail. The system around them failed. But the system writes the review, and the review determines the pay.
This is not hypothetical. The ONS data confirms that a disability pay gap already exists and has remained broadly stable over the past decade. The question is whether employer failure to implement reasonable adjustments is one of the mechanisms that sustains it. The employer’s failure to implement adjustments becomes invisible, while the employee’s resulting difficulty becomes a permanent record.
The invisible cost
There’s something that never appears on a performance review form, and it’s this: the sheer amount of time, energy, and health that neurodivergent employees spend just trying to get the support they were promised.
Fighting for reasonable adjustments is work. It’s not your job description, but it becomes your job. Writing emails to HR. Chasing occupational health referrals. Explaining your condition, again, to a new manager who wasn’t briefed. Following up on recommendations that were made months ago and never actioned. Researching your own rights because nobody in the organisation seems to know them. Keeping records because you’ve learned that if you don’t, your requests disappear. All of this takes hours. Weeks. Sometimes months or years. And every hour spent advocating for the support you need is an hour not spent on the work you’re being measured on.
Then there’s the masking. While you wait for adjustments that may never arrive, you still have to function in an environment that wasn’t designed for you. So you mask. You perform neurotypicality: forcing eye contact, suppressing stims, manufacturing small talk, pushing through sensory overwhelm, pretending that the open-plan office isn’t slowly dismantling your ability to think. Research on masking is unambiguous about the cost: it’s linked to anxiety, depression, identity erosion, and burnout. Over half of neurodivergent workers surveyed in the 2025 Neurodiversity Index had taken time off work due to neurodivergent-related challenges. Masking is how many of them stayed as long as they did.
And here’s where it becomes a trap with no exit.
By the time the performance review arrives, you are not the person you were at the start of the year. You are depleted, physically, mentally, emotionally, from months of unsupported work, unpaid advocacy for your own rights, and the daily performance of being someone you’re not. Your confidence is eroded. Your health may be genuinely compromised. You may have needed time off that you wouldn’t have needed if the adjustments had been in place. The version of you being reviewed is the version the employer created through their own inaction.
And then they rate that version. And that rating determines your pay.
The employer creates the conditions. The employee absorbs the damage. The employer measures the damage and calls it underperformance. At no point does the form ask: “Did we do what we said we would?” At no point does the process account for the energy spent fighting for basic support, or the toll of masking in an unadjusted environment, or the cumulative impact of both on the person now sitting in the review meeting being told they’re “partial.”
The performance review doesn’t measure the employee. It measures what’s left of the employee after the employer’s system has finished with them.
The descriptor problem
There’s a second layer to this. Even when adjustments are in place, many performance frameworks define competence through behavioural descriptors that are inherently neurotypical.
“Handles change with flexibility and positivity.” “Demonstrates excellent interpersonal communication.” “Proactively builds relationships across teams.” “Adapts quickly to shifting priorities.”
Read those as an autistic employee. Or an employee with ADHD. Or someone with anxiety. These aren’t measuring whether you do your job well. They’re measuring whether you do your job in a neurotypical way. An autistic engineer who delivers technically excellent work but communicates directly rather than diplomatically, or who needs predictability rather than thriving on change, literally cannot achieve a top rating under these descriptors, regardless of their actual output.
Employment Tribunals are beginning to catch up. In Jandu v Marks and Spencer plc, an employee with 22 years’ service was selected for redundancy using three behavioural criteria scored on a four-point scale. The tribunal found she had been marked down for “rushing,” “inaccuracies,” communication tone in emails, and attention to detail, all caused by her dyslexia. The selection criteria, the tribunal held, contained no purely objective elements and left a great deal of scope for subjective opinion, making it difficult for managers to apply them fairly. When Jandu disclosed the link between her scores and her dyslexia, the employer assumed without investigation that she was either lying or mistaken. The tribunal found unfair dismissal, failure to make reasonable adjustments, and discrimination arising from disability. Compensation exceeded £53,000, including aggravated damages.
The direction of travel is clear: performance frameworks that inadvertently penalise disability are discriminatory. But most employers haven’t stress-tested their descriptors. Most HR teams have never asked: “Could an autistic employee achieve the top rating under this framework without masking?” If the answer is no, the framework needs to change.
What good looks like
This isn’t about lowering standards. Nobody is asking for a free pass. It’s about measuring the right things.
Measure outputs, not style. Did the work get done? Was it done well? Was it delivered on time? These are the things that matter. How someone communicates, whether they make small talk, whether they “handle change positively”; these are preferences, not performance.
Adjust the review process itself. If an occupational health report says an employee needs support with performance planning and review, that means the review process needs adjusting, not just the employee’s desk or hours. Written questions in advance. Structured formats. Longer timeframes. A named person to support the process.
Train managers. 37% of managers have had no neurodiversity training. That means more than a third of the people writing performance reviews don’t understand the conditions of the people they’re reviewing. This isn’t a knowledge gap; it’s a liability.
Track the data. Employers should know whether disabled and neurodivergent employees receive lower performance ratings as a pattern. If they do, that’s not a coincidence. It’s systemic, and it needs investigating. The Equality Act 2010, s 149(1)(b) already requires public bodies to have due regard for the need to advance equality of opportunity. That should include scrutiny of performance outcomes by disability status.
Separate the adjustment failure from the performance outcome. If adjustments were recommended and not implemented, the employer cannot in good conscience rate the employee as underperforming. The question is not “did this person meet the standard?” It’s “did we give this person what they needed to meet the standard?” If the answer is no, the rating is not about performance. It’s about neglect.
How the trap works
Here’s how it plays out. Three mechanisms, one outcome.
The first is the framework problem: when competence is defined through neurotypical behavioural descriptors, and neurodivergent employees literally cannot be rated as competent no matter what they deliver.
The second is the support failure: when the adjustments that would have enabled competent performance were never put in place, and the resulting struggle is recorded as the employee’s failure rather than the employer’s.
The third is the most insidious. It’s the depletion cycle. By the time a neurodivergent employee reaches their performance review, they may have spent months fighting for adjustments that never came, masking to survive an environment that was never adapted, and absorbing the physical and mental health costs of both. They arrive at the review already diminished, not by their disability, but by the system’s response to it. And then that diminished version is measured, rated, and financially penalised.
The result is the same every time: financial penalty, stalled progression, eroded confidence, declining health, and eventually, departure. Only 30% of autistic people are in employment in the UK, compared with 82% of non-disabled people. This cycle is one of the reasons why. It doesn’t push people out in a single dramatic moment. It grinds them down, one unsupported review cycle at a time, until leaving feels like the only rational choice. And then the employer records it as attrition.
The question employers should be asking
Not: “Why isn’t this employee performing?”
But: “What did we promise to put in place, and did we actually do it? How much of this person’s energy went into fighting for support instead of doing their job? What did masking cost them? And is the version of this person we’re reviewing the product of their ability, or our neglect?”
If the answer to that last question is neglect (and for 20% of neurodivergent employees still waiting for promised adjustments, it clearly is) then the performance review isn’t measuring the employee. It’s measuring the employer.
And the employer is the one who should be rated partial.
References
- Equality Act 2010, s 15, s 20, s 149
- Pnaiser v NHS England [2016] IRLR 170 (EAT). BAILII
- City of York Council v Grosset [2018] EWCA Civ 1105. BAILII
- Jandu v Marks and Spencer plc (Case No 2200275/2021, London Central ET, 21 April 2022). Judgment
- Acas, ‘Performance, Conduct and Capability — Neurodiversity at Work’ (updated 16 December 2025). acas.org.uk
- City & Guilds Foundation & Do-IT Solutions, Neurodiversity Index Report 2024. cityandguildsfoundation.org
- City & Guilds Foundation & Do-IT Solutions, Neurodiversity Index Report 2025. PDF
- Davies, J., Cooper, K., Mallows, A., & Simons, J. S. (2022). Autistic adults’ views and experiences of requesting and receiving workplace adjustments in the UK. PLOS ONE, 17(8), e0272420. doi.org
- Hull, L., et al. (2017). “Putting on my best normal”: Social camouflaging in adults with autism spectrum conditions. Journal of Autism and Developmental Disorders, 47(8), 2519–2534. doi.org
- Cage, E., & Troxell-Whitman, Z. (2019). Understanding the reasons, contexts and costs of camouflaging for autistic adults. Journal of Autism and Developmental Disorders, 49(5), 1899–1911. doi.org
- Raymaker, D. M., et al. (2020). Defining autistic burnout. Autism in Adulthood, 2(2), 132–143. doi.org
- Office for National Statistics, Disability pay gaps in the UK: 2014 to 2023 (2024). ons.gov.uk
- Trades Union Congress, Disability pay and employment gaps (2025). tuc.org.uk
- Department for Work and Pensions, The Buckland Review of Autism Employment (2024). gov.uk
All URLs accessed 9 March 2026. Statistics from the Neurodiversity Index reports are drawn from convenience samples and should be treated as indicative rather than nationally representative. The ONS autism-specific pay gap figure (27.9%) carries wider confidence intervals due to sub-sample size.