Polkey Revisited: Pal v Accenture (UK) Ltd [2026] EAT 12

The decision of the Employment Appeal Tribunal (“EAT”) in Pal v Accenture (UK) Ltd [2026] EAT 12 provides a significant restatement of the correct approach to Polkey reductions, particularly in the context of progression-based performance models and overlapping disability claims. The judgment is a reminder that Polkey is a predictive, evidence-based exercise focused on what the employer would have done, not what the tribunal considers fair in hindsight.

Factual background

The claimant commenced employment with the respondent, a global professional services firm, in August 2009 as an Analyst. She was promoted to consultant level in 2011 and to a manager position in September 2013. The respondent operated a progression-based performance model, commonly described as an “up or elsewhere” system, under which employees were expected to demonstrate continuous development towards the next level of seniority. Promotion from Manager to Senior Manager was ordinarily expected within three to four years.

In September 2018, the claimant informed her employer that she required urgent surgery for ovarian cysts and was subsequently diagnosed with endometriosis. She had two periods of sickness absence related to this condition between September 2018 and January 2019, followed by a phased return to work.

The claimant was assessed as “Not Progressing” in performance reviews in August 2018 and March 2019. She was dismissed on 3 July 2019 following a meeting concerning her performance. Her appeal was unsuccessful.

The Employment Tribunal found that the dismissal was unfair because the respondent had failed to comply with its own Disciplinary and Appeals Policy. However, it applied a 100 per cent Polkey reduction, concluding that the claimant would have been dismissed in any event. It also rejected the claimant’s claims of disability discrimination, finding that she was not disabled within the meaning of the Equality Act 2010.

The EAT decision

The EAT allowed the claimant’s appeal on all grounds.

The Polkey reduction and the counterfactual error [71-84]

A compensatory award may be reduced as per section 123(1) Employment Rights Act 1996, to an amount the tribunal considers just and equitable in all the circumstances. That said, a Polkey reduction can only be made where the employer established that it would, or might have, fairly dismissed the claimant if it had the opportunity to correct the error that rendered the dismissal unfair [51]. The central error in this case was the Tribunal’s approach to the Polkey counterfactual. Rather than assessing what the respondent would have done had it complied with its own procedures, the Tribunal constructed a hypothetical process, effectively assuming that the employer would have introduced a new policy closely resembling the flawed process it had actually followed.

The EAT held that this was an error of law. A Polkey reduction must be grounded in evidence of the employer’s likely conduct if allowed to remedy the procedural defects that rendered the dismissal unfair. It is not permissible for a tribunal to speculate about alternative policies or processes in the absence of evidence that the employer would have adopted them.

The key question, therefore, is what the employer would or might have done, not what the tribunal considers would have been fair.

Capability and progression-based models

The EAT also addressed the interaction between Polkey and the respondent’s ‘up or elsewhere’ model. It emphasised that capability should be assessed by reference to the work the employee was employed to do under the contract, not by reference to readiness for promotion.

Dismissal based on a failure to progress to the next level may not amount to dismissal for capability if the employee is adequately performing their current role [63]. The Tribunal had failed to analyse whether, had the respondent complied with its own procedures, the application of the progression model would have provided a potentially fair reason for dismissal at all [90].

Disability and discrimination arising from disability

The EAT described the Tribunal’s analysis of disability as ‘wholly inadequate’ [103]. It had failed to engage properly with the statutory definition of disability, including the likelihood of recurrence of symptoms and whether the claimant’s condition would have continued to have a substantial adverse effect absent medical treatment.

The Tribunal had also failed to appropriately consider whether the respondent had actual or constructive knowledge of the claimant’s disability. The EAT noted that, had the respondent followed its own policies, further investigation would have taken place, during which the claimant’s medical condition should have been explored.

Similarly, the Tribunal failed to analyse whether the claimant’s dismissal was because of something arising in consequence of her disability, such as sickness absence or a phased return to work, or whether dismissal was a proportionate means of achieving a legitimate aim. These failures undermined both the discrimination findings and the Polkey assessment.

Practical implications

The decision reinforces several important points for practitioners.

First, Polkey reductions are evidentially demanding. Employers seeking substantial reductions must adduce clear evidence explaining how a fair process would have unfolded and why dismissal would probably have followed. Tribunals are not entitled to fill evidential gaps with their own assumptions.

Secondly, where progression-based performance models are relied upon, careful analysis is required to distinguish between performance in role and readiness for promotion. That distinction may be determinative both of fairness and of the Polkey counterfactual.

Thirdly, Polkey cannot be assessed in isolation from other claims. Errors in analysing disability, knowledge, or causation may fatally undermine findings on remedy.

What can parties do?

For respondents: Evidence is paramount

Respondents seeking to rely on Polkey must be able to demonstrate, with specificity, what would have occurred had a fair procedure been followed. This requires:

• witness evidence from relevant decision-makers;
• contemporaneous documentation supporting the employer’s rationale and intended process; and
• a clear explanation of how concerns would have been addressed through warnings, reviews, or alternative measures.

Without such evidence, tribunals may be unable to conclude that dismissal was likely, let alone inevitable.

For claimants: Test the counterfactual

Claimant representatives should rigorously challenge assertions that dismissal would have occurred in any event, including by:

• scrutinising the coherence and credibility of the employer’s evidence;
• exploring whether alternative outcomes, such as redeployment or further support, were realistically available; and
• considering whether, alongside procedural failings, there is an absence of genuine consideration.

Conclusion

Pal v Accenture does not change the law on Polkey, but it sharpens its application. By re-emphasising that the counterfactual must be supported by evidence from the employer, this case is a timely reminder that employers who fail to effectively prove inevitability cannot expect total reductions.

References

Pal v Accenture (UK) Ltd [2026] EAT 12