We had so many excellent questions during our recent webinar on the Employment Rights Act and tribunal risks that we simply did not have time to get through them all live. From AI-generated grievances and whistleblowing to pregnancy protections, reasonable adjustments, social media and sexual harassment, the questions reflected just how quickly the employment law risk landscape is changing.
So we have pulled together some of the most useful questions we received, with practical answers focused on what employers need to know now and where the law is heading next.
Listen again to our webinar on The Employment Rights Act and tribunal risks
When will the new gender equality and menopause action plan have to be published?
The government’s own Gender Pay Gap Reporting Service now expressly states that action plans are optional for the 2026/27 reporting year. For that cycle, employers may voluntarily publish a plan by 30 March 2027 for most public authorities or 4 April 2027 for private and voluntary-sector employers.
The Employment Rights Act itself does not set the first publication date. Section 33 inserts a regulation-making power into the Equality Act 2010 allowing regulations to require employers with at least 250 employees to develop and publish a gender equality action plan. Those regulations will determine when and how frequently plans must be published, subject to the Act preventing publication being required more frequently than every 12 months. The government continues to describe the plans as becoming mandatory from spring 2027, subject to secondary legislation.
Therefore we maintain our assumption that mandatory reporting will apply to the 2027/28 reporting year. For a private-sector employer, that would mean using the 5 April 2027 snapshot, with the gender pay gap report and associated action plan expected to be published by 4 April 2028. For most public authorities, the equivalent dates would be 31 March 2027 and 30 March 2028.
There is still an important caveat: the secondary legislation setting the mandatory timetable has not yet been made, so employers should not treat April 2028 as legally fixed yet. Organisations with 250 or more employees should work on the assumption that the April 2027 snapshot starts the first mandatory cycle and begin developing the plan during 2027.
What is legally defined as sexual harassment?
Sexual harassment is defined in section 26 of the Equality Act 2010 as unwanted conduct of a sexual nature which has the purpose or effect of violating someone’s dignity, or creating an intimidating, hostile, degrading, humiliating or offensive environment for them.
That can cover much more than physical contact. The EHRC gives examples including sexual comments or jokes, staring or leering, sexual gestures, intrusive questions about someone’s sex life, sexual images or messages, propositions and unwanted touching. It can happen in person, by email, on messaging platforms or through social media.
Importantly, the conduct does not have to be intended to harass. Where the issue is its effect, a tribunal considers the person’s perception, the circumstances of the case and whether it was reasonable for the conduct to have that effect. An employee also does not necessarily have to have expressly told someone to stop before behaviour can be “unwanted”.
The Employment Rights Act does not change this definition. What changes from 30 October 2026 is the employer’s responsibility for preventing it: employers will have to take all reasonable steps to prevent sexual harassment, rather than reasonable steps. The Act also strengthens the position on harassment by third parties such as customers and clients.
From a compliance perspective, we would avoid policies that define sexual harassment only through obvious examples such as touching or explicit propositions. The statutory definition is deliberately wider, so policies, training and risk assessments should cover verbal, non-verbal, digital and environmental conduct as well.
What evidence is there that the 2 year qualification was unfair, and people were disadvantaged?
There was no legal requirement for the government to prove that the two-year qualifying period was “unfair” before changing it. The length of the qualifying period is ultimately a matter of employment policy, and reducing it was a clear political commitment. Labour’s 2024 manifesto promised day-one protection from unfair dismissal as part of its Plan to Make Work Pay. This was implemented through the Employment Rights Act 2025, with the new six-month qualifying period coming into force on 1 January 2027.
Wasn’t sexual harassment already capable of being protected under whistleblowing law?
Yes, potentially. Before the Employment Rights Act change, a disclosure about sexual harassment could already qualify for whistleblowing protection where it tended to show an existing “relevant failure”, most obviously a breach of a legal obligation under section 43B of the Employment Rights Act 1996. Sexual harassment is unlawful under the Equality Act 2010, so there were circumstances where a sexual harassment disclosure could already fall within the whistleblowing regime.
What changed on 6 April 2026 is that the Employment Rights Act 2025 made this explicit. Sexual harassment is now a relevant failure in its own right, alongside matters such as criminal offences and breaches of legal obligations. This removes some of the uncertainty and means a worker does not have to characterise the harassment through another category of wrongdoing.
There is still an important qualification. A report of sexual harassment is not automatically a protected disclosure. The worker must still reasonably believe that the disclosure is in the public interest and that it tends to show sexual harassment has occurred, is occurring or is likely to occur. So we would describe the ERA change as clarifying and strengthening the route to whistleblowing protection, rather than creating protection where none could previously exist.
How do these changes impact on cases of constructive dismissal
The Employment Rights Act does not change the legal test for constructive dismissal. An employee still has to show that the employer committed a serious or fundamental breach of contract and that they resigned in response to that breach. Constructive dismissal claims remain difficult to establish.
What the Act does is increase the potential exposure. From 1 January 2027, the qualifying period for ordinary unfair dismissal, including constructive unfair dismissal, reduces from two years to six months and the compensatory award cap is removed. The tribunal time limit also increases to six months from October 2026. This means more employees will be capable of bringing constructive unfair dismissal claims and successful claims involving higher earners may be considerably more expensive.
There is also particular risk around contractual changes. Imposing a significant change without agreement can already amount to the fundamental breach needed for constructive dismissal, while the ERA separately introduces much stronger restrictions on dismissing employees who refuse certain contractual variations.
Employers should be cautious where an employee raises a serious grievance about changes to pay, duties, working arrangements or treatment. Investigating and trying to resolve the issue before the employee resigns is considerably safer than assuming that a resignation ends the problem.
What is an employer’s responsibility where its disability training does not cover a particular aspect of an employee’s condition?
There is no requirement for disability training to cover every condition or every possible manifestation of disability. However, training is not a defence if the employer knew, or could reasonably have been expected to know, that an employee was disabled and was being placed at a substantial disadvantage.
Under the Equality Act 2010, the duty is to consider and make reasonable adjustments in the individual circumstances. That may mean speaking to the employee, obtaining occupational health or other specialist advice, and considering adjustments even where the issue was never mentioned in the organisation’s standard training. The EHRC describes this as a positive and proactive duty once the employer has the relevant knowledge.
Training can itself become relevant. In Khorram v Capgemini, for example, ADHD awareness training had been recommended as part of a workplace needs assessment, and the tribunal found that failing to provide it contributed to a failure to make reasonable adjustments.
The Employment Rights Act does not change this test. From a compliance perspective, managers should be trained to recognise when they have reached the limits of their knowledge and escalate the issue rather than assuming that something not covered in training does not require action.
Are there any changes in the Act for the Menopause issues/dismissals?
There is no new standalone protection against dismissal because of menopause in the Employment Rights Act 2025, and menopause has not become a separate protected characteristic under the Equality Act 2010.
The main menopause-specific change is the new gender equality action plan regime. Employers with 250 or more employees will be required, once the relevant secondary legislation takes effect, to publish plans showing the steps they are taking to address gender equality, expressly including supporting employees going through the menopause. The government currently says these plans will become mandatory in spring 2027, subject to secondary legislation.
Dismissal risk itself continues to arise primarily under existing law. Menopause symptoms can potentially engage sex, age and disability discrimination protections. In particular, if the effects are substantial and long-term enough to meet the Equality Act definition of disability, the employer may also have a duty to make reasonable adjustments. Those discrimination protections do not require six months’ service.
From a compliance perspective, we would treat menopause-related absence or performance issues cautiously. Before moving towards dismissal, employers should establish whether symptoms may amount to a disability, consider adjustments and occupational health input, and make sure the rationale and process are properly documented.
Does the Employment Rights Act change the position on paternity leave and dismissal?
Yes. Although paternity leave is different from maternity leave because there are already specific statutory protections attached to taking paternity leave.
From 6 April 2026, the Employment Rights Act made statutory paternity leave a day-one right, removing the previous 26-week service requirement.
Employees were already protected from detriment or dismissal because they took, or proposed to take, paternity leave. A dismissal for that reason can be automatically unfair, so the employee does not need to satisfy the normal qualifying period for ordinary unfair dismissal.
The Act also creates a separate power to introduce stronger protection against dismissal after certain periods of statutory family leave. That new post-leave regime covers maternity, adoption, shared parental and neonatal care leave, together with Bereaved Partner’s Paternity Leave. It does not currently extend to ordinary paternity leave.
Employers should already treat any dismissal, disciplinary action or other detrimental treatment connected with paternity leave as high risk. The main ERA change is that employees can now qualify for the leave from day one, so the protection becomes relevant much earlier in the employment relationship.
We have an employee whose funded role ended two years ago. She has remained at home after refusing a number of alternative roles, and we believe she may now be pregnant again. With the current delays in the Employment Tribunal system and further pregnancy protections coming under the Employment Rights Act, could this affect our position?
The first point is that a delay in the tribunal hearing does not normally change which law applies. A dismissal or other employment decision is generally judged according to the law in force when that decision was taken. If a claim is heard after new Employment Rights Act provisions come into force, that does not ordinarily make those provisions retrospective.
However, if the employee is still employed, any new decision you take now is a new legal event. If she is pregnant again and informs the employer, the existing special redundancy protection applies from that point. Where a genuine redundancy exists, a pregnant employee must be given priority for any suitable alternative vacancy that exists. If an employee unreasonably refuses genuinely suitable alternative employment, that can affect entitlement to statutory redundancy pay, although suitability and the reasonableness of the refusal have to be assessed on the particular facts. ACAS specifically cautions that matters such as pay, location, health and disruption to family life can make refusal reasonable.
We would also be careful about acting simply because you believe she is pregnant. The enhanced redundancy protected period formally starts when the employer is informed of the pregnancy. Separately, any decision influenced by an assumption about pregnancy creates obvious discrimination risk.
The Employment Rights Act will strengthen dismissal protection for pregnant women and new mothers further in 2027, although the detailed regulations are not yet in force and the government has not yet settled all of the circumstances in which dismissal will remain permitted. The Act currently provides the regulation-making framework rather than an immediate prohibition. The government’s latest timetable still lists the enhanced protections for 2027.
So the tribunal backlog itself should not be the concern. The greater risk is allowing a difficult employment situation to remain unresolved while the employee remains employed and potentially enters another protected period. Given the two-year history here, we would want the employer to review the redundancy rationale, every alternative role offered and the reasons each was refused, confirm the employee’s current employment status, and take specific employment-law advice before taking any final step. You can contact our partners at WorkNest for more dedicated employment law support.
Do sexual harassment prevention measures also need to cover harassment of employees by third parties, and how should employers approach this?
Yes. In fact, employers should already be considering third-party sexual harassment as part of their existing preventative duty. EHRC guidance is clear that the current duty to take reasonable steps to prevent sexual harassment includes harassment by customers, clients, patients, suppliers, contractors and other third parties.
From 30 October 2026, the Employment Rights Act strengthens this further. Employers must take all reasonable steps to prevent sexual harassment, and a separate provision will make it unlawful for an employer to “permit” a third party to harass an employee. For these purposes, an employer permits the harassment where it happens in the course of employment and the employer failed to take all reasonable steps to prevent it. Importantly, this new third-party protection extends beyond sexual harassment to the other forms of harassment covered by section 26 of the Equality Act.
What amounts to “all reasonable steps” will depend on the workplace and the risks. We would expect employers to assess where employees interact with third parties, expressly cover third-party harassment in policies and training, provide clear reporting routes, and have practical procedures for responding. In a hospitality business, for example, that might include empowering managers to warn or remove customers, banning repeat offenders, addressing lone working and alcohol-related risks, and making clear to staff that commercial relationships do not take priority over their protection. The EHRC specifically recommends assessing third-party risks and having procedures for warning, removing or banning customers where appropriate.
A generic anti-harassment policy will increasingly be difficult to rely on by itself. Employers should be able to show that they identified their actual third-party risks, introduced controls relevant to those risks, acted on previous incidents and reviewed whether those controls were working.
If a candidate refuses a role because of an employee’s social media post expressing protected beliefs, does that prove reputational risk?
It would certainly be evidence of a real impact, and it is stronger than an employer simply asserting that a post might damage its reputation. If the candidate expressly says that the post was the reason they declined the role, the employer can legitimately take that into account when assessing reputational or operational harm.
However, it does not automatically prove sufficient reputational harm to justify disciplinary action or dismissal. In Higgs v Farmor’s School, the Court of Appeal made clear that even where reputational harm, or a risk of it, is established, any restriction on the employee’s expression must still be proportionate. The Court specifically distinguished genuine evidence of people being deterred from engaging with an organisation from more generalised concerns about reputation.
So the employer would still need to consider what was actually posted, whether it was an expression of a protected belief, how it was expressed, whether the employee was identifiable with the organisation, the employee’s role, the scale of the impact and whether a response short of dismissal would adequately address the issue. Our existing guidance makes the same point: a third-party complaint should trigger a structured assessment rather than automatic discipline.
We would suggest to document the candidate’s stated reason carefully, because that may be valuable evidence of actual impact. We would not treat a single candidate declining a job as automatically establishing that dismissal of the employee is proportionate. The evidence strengthens the employer’s case on reputational risk; it does not decide the legal question by itself.
If a workplace has only one unisex toilet, how does that comply with the Workplace (Health, Safety and Welfare) Regulations 1992?
A single unisex toilet can comply with the 1992 Regulations. Regulation 20 normally requires separate toilet facilities for men and women, but expressly provides an exception where each toilet is in a separate room with a door that can be secured from the inside. HSE confirms that men and women can therefore use the same toilet where it is a self-contained, lockable room.
The employer must still provide “suitable and sufficient” facilities. This means one unisex toilet may be sufficient for a very small workplace, but not necessarily for a larger workforce. HSE guidance indicates that for mixed use, workplaces with 1–5 people require at least one toilet, 6–25 require two, and the number increases thereafter.
The facilities must also be readily accessible, adequately ventilated and lit, kept clean, and suitable for the workforce, including any necessary provision or reasonable adjustments for disabled employees.
So the important distinction is between a single self-contained unisex toilet, which can comply, and simply replacing separate male and female facilities with a shared multi-user toilet area. The statutory exception is specifically based on each WC being contained in its own lockable room.
We are seeing an increase in very long, apparently AI-generated grievances, sometimes running to 50 pages. How can an employer reasonably manage these without having to respond line by line?
The fact that a grievance appears to have been drafted using AI does not reduce the employer’s obligation to take the underlying concerns seriously. The current Acas Code requires the employee to set out the nature of their grievance in writing, and the employer should meet with them, investigate where necessary and provide an outcome. There is no requirement, however, for an employer to produce a point-by-point answer to every sentence of a 50-page document.
A sensible approach is to case-manage the grievance. The employer can acknowledge the full document but ask the employee to identify the specific allegations they want investigated, ideally as a numbered list setting out what happened, when, who was involved and what outcome they are seeking. The employer can then confirm the issues it understands it is being asked to determine and invite the employee to correct anything material that has been missed.
This is actually an issue Acas is looking at right now. Its July 2026 draft replacement Code says that concerns have been raised by employers and workers about “unnecessarily long and complex written grievances”. The draft proposes that a short, clear written explanation will often be sufficient initially, with further information supplied later where necessary. That is currently only a consultation proposal, so it should not be treated as law yet, but it strongly supports proportionate management of these cases.
We would therefore avoid writing back simply to say that a 50-page grievance is “unreasonable”. A safer formulation is that the employer has reviewed the material, has identified the substantive allegations it will investigate, and will not provide individual responses to repetitive, irrelevant or purely argumentative material where that is not necessary to determine the grievance.
Employers should still be careful where the document contains allegations of discrimination, harassment, whistleblowing or other statutory breaches, and should consider whether a disability means an adjustment to the normal grievance process is required. But AI should not be allowed to turn grievance handling into an unlimited drafting exercise. Fairness requires the substance to be addressed, not every paragraph generated around it.
If an employee has raised a whistleblowing disclosure about sexual harassment, the internal process has been exhausted and they remain dissatisfied, where can they escalate it externally?
There is no requirement to exhaust the employer’s internal process before making an external protected disclosure. If the concern is genuinely a whistleblowing matter about sexual harassment or wider equality-law breaches, the most directly relevant prescribed body is the Equality and Human Rights Commission (EHRC). The government’s current prescribed-persons list expressly identifies the EHRC for disclosures concerning compliance with equality and human rights legislation.
However, the EHRC is not an appeal body for an individual grievance. Its role is regulatory. It can receive whistleblowing information and decide whether it raises wider equality-law compliance concerns, but it will not ordinarily determine the employee’s individual claim or award compensation. The EHRC itself stresses that whistleblowing protection generally requires a public-interest element wider than one employee’s purely personal circumstances.
If the individual believes they personally suffered sexual harassment, victimisation, whistleblowing detriment or dismissal, the route for obtaining an individual legal remedy is normally Acas Early Conciliation followed, if unresolved, by an Employment Tribunal claim.
There can also be other external routes depending on the facts. If the conduct may amount to a criminal offence, it can be reported to the police. In a regulated sector, such as financial services or healthcare, the relevant sector regulator may also be a prescribed person for aspects falling within its remit. The key point is that “external escalation” is not one single process: EHRC is the obvious prescribed whistleblowing body for systemic equality-law concerns, while the Employment Tribunal is the route for resolving the employee’s individual legal claim.
Are volunteers covered by whistleblowing protection?
Generally, no. A genuine volunteer who has no enforceable employment or worker contract is not usually protected by the statutory whistleblowing regime under the Employment Rights Act 1996. Acas expressly lists volunteers without an enforceable employment contract among those who are normally outside the protection.
The important qualification is that employment status depends on the reality of the arrangement, not simply the label “volunteer”. If someone described as a volunteer is in fact working under an enforceable arrangement that gives them worker status, whistleblowing protection may apply.
Organisations can still choose to allow volunteers to raise concerns through their whistleblowing policy, and that is often sensible governance, particularly for charities. But including volunteers in the policy does not by itself give them the statutory protection against detriment or dismissal that applies to workers.
For the sexual-harassment change specifically, the same status issue applies: making sexual harassment an express category of whistleblowing wrongdoing from 6 April 2026 did not extend statutory whistleblowing protection to genuine volunteers.
What steps should employees in small organisations without an HR department take to raise a grievance?
The absence of an HR department does not remove the employee’s right to raise a grievance or the employer’s responsibility to deal with it fairly. Employers are required to set out in writing who employees should contact if they have a grievance and how they should do so.
Usually, the employee should first try to resolve the issue informally with their line manager where appropriate. If that does not work, or the matter is too serious for an informal approach, they should put the grievance in writing to their manager, a more senior manager, a director or ultimately the employer themselves. If the grievance concerns the person who would normally receive it, it should go to someone else with sufficient authority to deal with it. Acas specifically says that where there is no organisational grievance procedure, the employee should approach their line manager or another appropriate person they feel comfortable speaking to.
The employer should then investigate where necessary, hold a grievance meeting, provide a written outcome and allow an appeal. The employee normally has the right to be accompanied at the grievance and appeal meetings by a colleague or trade union representative.
For a very small organisation, independence can sometimes be difficult because the same few people are involved. That does not make the process impossible. The employer should separate investigation, decision and appeal as far as reasonably practicable, and in particularly sensitive or complex cases it may be appropriate to use an external HR consultant or investigator. The key is not whether there is an HR department, but whether the concern is taken seriously and handled through a fair, documented process consistent with the Acas Code.
Is this protection restricted to people with a formally diagnosed neurodivergent condition?
No. A formal diagnosis is not required for someone to be disabled under the Equality Act 2010. What matters is whether they meet the statutory definition of disability, broadly whether they have a physical or mental impairment that has a substantial and long-term adverse effect on their ability to carry out normal day-to-day activities.
Acas specifically says that a worker does not need a diagnosis for neurodivergence to amount to a disability, and recommends that employers consider support and reasonable adjustments whether or not a diagnosis has been obtained.
That does not mean every employee who describes themselves as neurodivergent automatically has a disability in law. The individual circumstances and functional impact still matter. Equally, an employer should not simply say, “there is no diagnosis, so no adjustment is required”. If the employer knows, or could reasonably be expected to know, that the employee may have a disability and is experiencing a substantial disadvantage, the reasonable-adjustment duty can arise.
The safer approach is to focus on what difficulty the employee is experiencing and what support may be reasonable, rather than making a diagnosis the gateway to assistance. Acas also advises employers not to try to diagnose employees themselves.
How should an employer manage disability discrimination and reasonable adjustments where the employee has not disclosed a disability?
An employee does not have to disclose a disability. However, the reasonable-adjustment duty generally arises only where the employer knows, or could reasonably be expected to know, both that the employee is disabled and that they are likely to be placed at a substantial disadvantage.
The important point is that an employer cannot always rely on “they never told us”. If there are clear indicators, such as repeated sickness absence, difficulty carrying out particular tasks, changes in behaviour or information given to a manager, the employer may be expected to make reasonable enquiries. Acas recommends speaking to the employee privately, asking whether they need support and, where appropriate, seeking occupational health or medical input with their consent. The employer should not try to diagnose the employee itself.
If there was genuinely nothing from which the employer could reasonably have known about the disability, that can be a defence to a failure-to-make-reasonable-adjustments claim. But once warning signs arise, the safer compliance approach is to ask, explore and document, rather than wait for the employee to use the word “disability”.
If an employee only discloses a disability when an alleged misconduct incident is being investigated, should the employer pause the investigation or deal with the disability issue in parallel?
You would normally deal with the two issues in parallel rather than automatically pausing the investigation. A late disclosure does not prevent an employer investigating what happened, and Acas specifically says employers should offer support even where an employee only discloses neurodivergence once a formal procedure has started.
However, from the point of disclosure, the employer should take the disability issue seriously. That may mean asking for further information, obtaining occupational health or medical advice with the employee’s consent, considering reasonable adjustments to the investigation itself, and investigating whether the disability may have contributed to the alleged conduct. A formal diagnosis is not necessarily required.
The important distinction is between investigating the facts and deciding the consequences. You can usually continue gathering witness evidence and establishing what happened while the disability evidence is being considered. But I would be reluctant to reach a disciplinary finding or impose a sanction until any potentially relevant disability evidence has been properly considered.
That is particularly important because of City of York Council v Grosset. For discrimination arising from disability, an employer does not necessarily need to have known that the misconduct was caused by the disability. If the conduct objectively arose in consequence of the disability, and the employer knew or should have known that the employee was disabled when it subjected them to the unfavourable treatment, the employer may have to justify that treatment as proportionate.
So a disclosure made only after the incident does not retrospectively mean the employer should have made adjustments before an incident it could not reasonably have known was disability-related. But once the information is on the table, it should not be ignored simply because it arrived late. The employer should investigate the misconduct and the disability explanation together, make any procedural adjustments required, and ensure the final decision reflects both.
Can employers insure against Employment Tribunal liability and compensation awards?
Yes, to a significant extent. Employers can purchase Employment Practices Liability Insurance (EPLI), sometimes as a standalone policy and sometimes as part of management liability or legal expenses cover. Depending on the policy, it can cover legal defence costs, settlements and tribunal awards arising from claims such as unfair or wrongful dismissal, discrimination, sexual harassment and whistleblowing. Some policies expressly cover damages for injury to feelings as well as judgments and settlements.
However, it is not a complete transfer of employment-law risk. Cover is subject to the policy limit, excess and exclusions. Policies may exclude known disputes that existed before the insurance was taken out, deliberate or criminal conduct, certain wage or contractual liabilities, and fines or penalties. Employers also often have to notify the insurer very early and may need to follow the insurer’s legal advice before dismissing or settling if they want the eventual award to be covered.
It is also important not to confuse EPLI with compulsory Employers’ Liability insurance. Employers’ Liability insurance principally covers compensation where an employee is injured or becomes ill because of their work; it does not ordinarily cover unfair dismissal or discrimination claims.
There is one further distinction around penalties. A tribunal can, in cases involving aggravating features, impose a financial penalty on an employer of up to £20,000 in addition to compensation payable to the claimant. Employers should not assume that such regulatory or statutory penalties will be insured simply because the underlying compensation award is covered.
What are the risks if an employer tends to select remote workers for redundancy?
That could create a significant unfair dismissal and discrimination risk. Remote working is not itself a protected characteristic, but redundancy selection criteria must be fair, objective and non-discriminatory. Acas specifically warns that using flexible working arrangements as a redundancy criterion can amount to indirect discrimination.
The particular risk is that remote workers may disproportionately include groups protected under the Equality Act, for example women with caring responsibilities or disabled employees working remotely as a reasonable adjustment. A practice of scoring remote workers less favourably could therefore amount to indirect sex discrimination or disability discrimination. In the disability context, selecting someone because they work remotely for reasons connected with their disability may also amount to discrimination arising from disability.
That does not mean remote working can never be relevant. If a restructure genuinely requires particular roles to be performed on-site, the employer may have a legitimate business reason for that requirement. But it should be able to evidence why physical presence is necessary, consider whether remote or hybrid working remains workable, apply the criterion consistently and consider reasonable adjustments. Indirect discrimination can potentially be justified, but only where the employer can show that the measure is a proportionate means of achieving a legitimate aim.
If the recipient perceives conduct as harassment related to a protected characteristic, can it still be harassment even if no explicit reference to that characteristic was made?
Yes, potentially. The Equality Act does not require someone to use an explicit slur or directly mention the protected characteristic. The test is whether there was unwanted conduct related to a relevant protected characteristic which had the purpose or effect of violating the person’s dignity or creating an intimidating, hostile, degrading, humiliating or offensive environment.
However, the recipient’s perception is not enough on its own. Where the allegation is based on the effect of the conduct, a tribunal must consider three things: the recipient’s perception, all the other circumstances, and whether it was reasonable for the conduct to have that effect.
So, for example, repeated comments about somebody being “too old-fashioned”, assumptions about what they can physically manage, or mocking a religious practice could potentially be harassment even without expressly saying “because of your age”, “because of your disability” or naming their religion. The connection with the protected characteristic can come from the context and substance of the conduct.
The compliance point is therefore not to ask only, “Did anyone explicitly mention the protected characteristic?” Employers should ask what happened, why it may have been related to the characteristic, how the employee experienced it, the surrounding context and whether that effect was objectively reasonable. Perception matters, but it does not automatically determine the outcome.
Would a workplace WhatsApp group count as a workplace forum even if the employer does not run or participate in it, although some managers are members?
Potentially, yes. There is no special legal category of “workplace forum” that determines liability. The real question is whether the WhatsApp activity is sufficiently connected with work to be regarded as taking place “in the course of employment”. Under section 109 of the Equality Act 2010, an employer can be liable for harassment or discrimination committed by employees in the course of employment even if the employer did not know about or approve it. Acas expressly says that work-linked social media activity can fall within this principle.
So an unofficial WhatsApp group consisting largely of colleagues, used to discuss work, shifts, colleagues or workplace issues may still be work-related even though it uses personal phones and was not created by the employer. EHRC guidance specifically recognises that harassment can occur through social media, messaging tools and personal devices, and says employers should address online harassment in their policies.
The presence of managers would make the position more significant, particularly if they participate in inappropriate conversations, encourage them, or become aware of harassment and fail to act. It may strengthen both the connection with employment and the argument that the organisation knew there was a problem. But a manager simply being a silent member of a private chat would not automatically make every message an act of the employer.
The contrast with a case such as Wilkins is useful. There, the forum was formally workplace-sanctioned, which made the employer connection particularly clear. A private WhatsApp chat requires a more fact-specific assessment. I would look at who created it, why it exists, who belongs to it, what is discussed, whether managers use it for work, and how closely the disputed messages relate to the employment relationship.
From a compliance perspective, employers should therefore avoid policies that say, in effect, “personal WhatsApp is outside our responsibility”. A better position is that harassment, bullying and discrimination involving colleagues may fall within workplace rules even on private messaging platforms and personal devices where there is a sufficient connection with work.
In a diverse workplace, where a term intended as friendly or affectionate in one culture may be offensive in another, does the employee have to raise the issue first and give the employer an opportunity to deal with it before it can amount to harassment?
No. There is no legal requirement for the employee to object first, or for the employer to mishandle an earlier complaint, before conduct can amount to harassment. EHRC guidance expressly says that conduct can be “unwanted” even where the worker did not tell the person at the time that they objected to it.
Cultural context is, however, highly relevant. Under section 26 of the Equality Act, a tribunal considers the employee’s perception, all the circumstances of the case, and whether it was reasonable for the conduct to have the alleged effect. EHRC guidance specifically identifies the race or cultural background of those involved as one of those circumstances. A term may carry an offensive meaning in one cultural context that somebody from another background genuinely did not appreciate.
That means an innocent intention or cultural misunderstanding does not automatically prevent harassment, because conduct can amount to harassment because of its effect even where no offence was intended. But equally, an employee cannot establish harassment merely by saying that they personally found something offensive; the tribunal also applies an objective reasonableness test.
Where prior notification becomes particularly important is in how the employer responds and what happens next. If somebody explains that a term is offensive, the colleague apologises and stops using it, that is a very different risk profile from an employer allowing the same conduct to continue after it has been raised. EHRC guidance even gives an example of an unintentionally offensive racial term where the colleague apologises and the employee accepts the apology.
So from a compliance perspective, employers should encourage people to raise cultural misunderstandings early and resolve them informally where appropriate, but they should not adopt a rule that harassment only becomes actionable after someone has first complained.
Can an employer control the content of an employee’s LinkedIn profile, or is LinkedIn still personal social media even though it is primarily used for work and professional networking?
LinkedIn sits somewhere between the two. A personal LinkedIn profile remains the individual employee’s account simply because it is professionally focused. In Hays v Ions, the High Court itself described the employee’s LinkedIn as his personal account even though he had used it extensively for business and with his employer’s encouragement. The employer’s potential rights arose from confidential client information transferred into the account, not from ownership of the profile itself.
An employer can nevertheless impose reasonable rules on the work-related aspects of an employee’s LinkedIn use. For example, it could require an accurate job title, prohibit disclosure of confidential information, set rules about use of corporate branding, require regulatory disclaimers where relevant, and discipline inappropriate posts where there is a sufficient connection with work. If posting on LinkedIn is genuinely part of the employee’s job, such as in sales, recruitment or marketing, the employer can also set reasonable expectations around that work output.
What is much harder is asserting a general right to determine everything the employee says on their personal profile. Any restriction needs a legitimate reason and should be proportionate, particularly where posts concern political, religious or philosophical beliefs. The Higgs principles relevant here: an employer cannot simply dictate acceptable personal beliefs or expressions because it fears controversy. It needs to distinguish the employee’s personal expression from conduct that creates an identifiable workplace, regulatory or reputational problem.
There is also an important distinction between a personal profile and LinkedIn assets operated on the employer’s behalf. In Whitmar Publications v Gamage, LinkedIn groups were managed by an employee as part of her job, using company systems, and operated for the company’s benefit. The court treated those very differently from a purely personal profile.
Finally, monitoring an employee’s LinkedIn activity is still processing personal data. The ICO says employers must have a clear purpose, act lawfully and fairly, and ensure monitoring is necessary and proportionate. The fact that material is publicly visible does not give an employer an unrestricted right to monitor or use it for any purpose.
Is it correct that the new single-sex guidance applies to services rather than workplace facilities?
Yes. The new EHRC 2026 Code of Practice applies to services, public functions and associations. It is not a new code governing workplace toilets or changing facilities. There is an existing Employment Code, but there is currently no equivalent new post-For Women Scotland workplace code. However employers do not need to wait for further guidance before applying the law.
It is also important to distinguish guidance from law. Even the new statutory Code expressly states that it does not itself impose legal obligations and is not an authoritative statement of the law. Courts and tribunals must take relevant parts into account, but ultimately it is the legislation and the courts that determine the legal position.
There has been no change to the wording of the Equality Act 2010 as a result of the Supreme Court judgment. In For Women Scotland, the Supreme Court held that “sex”, “woman” and “man” in the Equality Act mean biological sex and, importantly, that this is what those terms “were always intended to mean”. The judgment therefore clarified the correct interpretation of the existing law; it did not introduce a new definition of sex.
The same is true of workplace facilities legislation. The Workplace (Health, Safety and Welfare) Regulations 1992 have not been rewritten. Employers still need to comply with those existing requirements on separate male and female sanitary conveniences, subject to the exception for facilities contained in separate lockable rooms, while also complying with the Equality Act protections against discrimination.
What should an employer do if two female employees report feeling uncomfortable because of a male colleague’s looks and comments such as calling them “pretty”, but there are no witnesses?
The absence of witnesses does not mean the complaint cannot be upheld. Acas expressly says employers should not doubt a sexual harassment complaint simply because nobody else witnessed it. The employer should investigate both accounts fairly, hear from the employee accused and assess all of the available evidence.
The alleged behaviour is also capable of falling within the legal definition of sexual harassment. EHRC guidance specifically identifies suggestive looks, staring or leering as conduct that may be sexual in nature. Comments about appearance such as “pretty” are more context-dependent: the word itself is not automatically harassment, but repeated or unwanted comments about appearance, particularly alongside staring or other behaviour, could contribute to an intimidating, degrading, humiliating or offensive environment.
The fact that two employees have independently reported similar behaviour can be relevant evidence, although it does not automatically prove the allegations. The investigator should look at whether the accounts are consistent, whether they describe a similar pattern, when each employee first raised concerns, whether there are contemporaneous messages or notes, and the response of the employee accused. The process should remain neutral rather than starting from an assumption that either side is telling the truth.
Ultimately, an internal investigation does not require criminal-law proof. The employer can reach a finding on the balance of probabilities after a reasonable investigation. If the evidence is genuinely insufficient, it can record that the allegation is not substantiated rather than concluding that it was false. The important point is that “there were no witnesses” is not, by itself, a reason to stop the investigation or reject the complaints.
