What does "reasonable" really mean in reasonable adjustments?

If you work in HR, you have almost certainly been asked some version of this question.

A manager has come to you with a proposed adjustment for a neurodivergent employee. Maybe they have proposed it themselves. Maybe the employee has asked for it. Maybe an occupational health report has suggested it. The question on the table is the same one every time. Is it reasonable?

It is one of the most common questions I get when I am training HR teams. And it is also where I see the most mistakes being made. The good news is that the answer is not vibes. There are five tests an adjustment has to pass to be reasonable and once you have the tests in your head, the question becomes much easier to answer.

What is the biggest mistake people make with reasonable adjustments?

Most HR people I meet do the same thing when this question lands on their desk. They Google it. They find a blog with a list of “adjustments for ADHD” or “adjustments for autism” and they work down the list trying to decide which bits to implement.

This is the wrong starting point.

Reasonable adjustments are not generic packages. A list of standard adjustments for employees with ADHD or any other condition might give you useful prompts but it cannot tell you whether a particular adjustment is reasonable for a particular person in a particular role in a particular workplace. That answer only comes from looking at all three together.

There is a structured way to do this and you do not need to be a lawyer to use it. There are five tests an adjustment has to pass to be reasonable. If it passes all five, it is reasonable. If it fails one of them, it probably is not and that is the place to start the conversation with the manager or the employee about what would work instead.

Test 1: Is it effective?

A reasonable adjustment exists to do something specific. It exists to lower a barrier so that the neurodivergent employee is on a level playing field with their colleagues. If the adjustment does not actually do that, it is not reasonable. It is just expensive.

This is where the blog-list approach falls apart most often. You can implement an adjustment that is associated with ADHD support and find that for this person, in this role, it makes no measurable difference. That does not mean the adjustment is bad in principle. It means it is not the right one here.

The right question is never “what do ADHD people need” or “what do autistic people need”. The right question is “what is the specific barrier for this person and is this specific adjustment lowering it”.

Sometimes the first thing you try does not work. That is normal. There is usually more than one way to lower a given barrier, so if an adjustment is not working you change it for something else and measure that. What is not acceptable is putting something in place, not measuring whether it works and continuing to call it reasonable.

Test 2: Is it practical?

The second test is whether the adjustment can actually be implemented in the environment the person works in.

Practicality is contextual every single time, which is exactly why blog lists are so misleading. The same adjustment can be transformational in one workplace and impossible in another.

Take noise-cancelling headphones, which appear on almost every “support for ADHD at work” list you will read. In an open-plan office, they can be a genuine help. They lower the sensory load, they signal to colleagues that someone is in focused work and they cost very little. On a construction site, in a fire station, in any environment where the person needs to hear what is happening around them, they are not just impractical. They are dangerous, which brings us to the safety test in a moment.

Or take project management software. A piece of structured planning software can be a real change for a neurodivergent employee who needs visible task structure. But if the rest of the team works in a different system and the employee now has to maintain their work in two places, the adjustment has not reduced their friction. It has doubled it.

Communication protocols are another one. In a stable team where the same people work together every day, a clear agreed communication protocol can be enormously helpful. In a role where the person regularly works with different teams, different agencies or members of the public, you cannot dictate the communication style of people outside your organisation and so the adjustment is impractical to implement.

The test is not “could this adjustment exist somewhere in the world”. The test is “does this adjustment work in the place this person actually works”.

Test 3: Is it affordable?

Affordability is the test that gets argued about most and it is also the one where employers most often get the maths wrong.

The affordability test is not just the sticker price of the adjustment. You have to weigh the cost of the adjustment against the cost of not doing it. CIPD figures put the average cost of replacing an employee at around £30,000 when you account for recruitment, lost productivity, onboarding and the gap in delivery while the role is empty. If you are looking at a £1,000 adjustment and calling it unaffordable but the alternative is the person leaving or being managed out, the comparison is not £1,000 against zero. It is £1,000 against £30,000.

Affordability also scales with the size of the employer. The Equality Act 2010 explicitly recognises that what is reasonable for a 5,000-person organisation is not the same as what is reasonable for a five-person business. Larger employers are expected to absorb more cost. Smaller employers are expected to absorb less and Access to Work exists to support the gap.

Access to Work is the bit most HR people underuse. It can fund equipment, software, support workers, coaching and travel adjustments that would otherwise be beyond what a smaller employer could afford. If you are getting close to “we cannot afford this” as your answer, the conversation to have is with Access to Work, before you start asking whether you can employ the person at all.

The headline test on affordability is simple. Is the cost of the adjustment genuinely disproportionate to what your organisation can absorb, taking the cost of not doing it into account? If the answer is yes, you may have an affordability ceiling. If the answer is no, you do not and “we cannot afford it” is not the reason to refuse.

Test 4: Is it safe?

The fourth test is health and safety and unlike the others it is non-negotiable. If an adjustment would compromise the safety of the individual or anyone around them, it is not reasonable. There is no version of the conversation where it becomes reasonable.

This matters more in some roles than others. In construction, in the emergency services, in care, in driving roles, in any environment where safety is a foreseeable risk and where reduced awareness or impaired response could cause harm, this test is key. Noise-cancelling headphones on a building site is the obvious example. Reduced supervision for someone whose role involves safeguarding is another. A flexible start time that means a critical handover is unsupported is a third.

The principle is that the adjustment has to fit the job, not just the person. Where the job carries safety responsibilities, those responsibilities are part of the role the person needs to be able to do and any adjustment that conflicts with them is not reasonable.

This is also a place where Test 1 and Test 4 work together. If the only way to lower the barrier compromises safety, the adjustment fails both tests and the right response is not to push the adjustment through anyway but to look for a different way to lower the barrier. There is almost always one.

Test 5: Can they still do the job?

The fifth test is whether the person, with the adjustment in place, can still do the job.

This is the test that gets dressed up in language about “the essential functions of the role” but in plain English it comes down to a clean question. With this adjustment, can the person still deliver what the job requires?

A reasonable adjustment can include redeployment to an alternative role where the strengths fit better and the barriers are lower. That is not a failure, it is a sensible outcome. But whichever role the person ends up in, with the adjustments in place, they need to be able to do that role.

What a reasonable adjustment is not is “take all the bits of work you find hard and give them to your colleagues”. I see this happen, with the best of intentions and it is a problem on two fronts. It is unfair on the colleagues who are now carrying that work and it is patronising to the neurodivergent employee. It treats them as someone who cannot do the work, rather than as someone who can do the work with the right environment.

There is a meaningful difference between redistributing work fairly to play to the strengths of everyone in the team, which can be a sensible change and quietly reducing one person’s workload to less than what their role requires. The first is workplace design. The second is a slow-motion capability issue dressed up as inclusion.

What do all five tests have in common?

Look back across the five tests and you will notice something. Every single one of them is about the individual and the environment together. Not the individual on their own. Not a list of adjustments off the internet. The person, in their actual role, in their actual workplace, with the actual barrier they are facing.

That is the bit blog lists cannot ever give you. A blog can tell you what has worked for other people in other situations. It cannot tell you whether it will work for the person in front of you. The five tests are how you make that call.

For a deeper read on the legal background to all of this, our guide to reasonable adjustments covers the Equality Act definition, the duty on employers and what happens at tribunal when things go wrong.

Where a Workplace Needs Assessment fits in

There is a tool that is designed to do exactly this work and that is a Workplace Needs Assessment.

A WNA is the structured process for bringing the individual’s specific barriers and the workplace context into the same conversation. A good assessment looks at what the person is finding difficult, what the role actually requires, what the environment makes possible and what does not work - then recommends adjustments that pass the five tests above by design, not by luck. If you would like to know what to expect from the process, we have a full walkthrough on the knowledge hub.

If you are sitting with an adjustment question and you genuinely do not know whether it is reasonable, a WNA is the right next step. It will either confirm the adjustment as the right answer, with the evidence to support it or it will surface a different answer that fits the person and the role better.

If you would like to talk through whether a Workplace Needs Assessment would help in a specific situation, book a free consultation with me and we can work it out together.