Platform The IWE Benefits Sectors Resources How It Works About Book a call
Home The measurement case for boards
Workplace health strategy

The measurement case for boards

The short answer

A board does not decline workplace-health spending because it disagrees that people matter. It declines because the paper in front of it does not answer the questions a board is obliged to ask: what is the exposure, what are we required to do, what changed last time, and how would we know if this worked.

Those are answerable. Not with a benchmark, and not with an engagement score, but with a record of what you knew, when you knew it, what you did about it and what happened next. This page sets out what goes in that record, what the law actually requires, and the things measurement cannot tell you, which matter just as much when the questions get hard.

No figures quoted Last reviewed 16 August 2026

What a board actually asks

Four questions, in roughly this order. A proposal that answers all four is a decision paper. One that answers the first and skips the rest is an appeal to values, and boards decline those politely and repeatedly.

Question one

Is there an exposure?

  • What do we know about the health of this workforce.
  • How do we know it, and how current is it.
  • Where are we guessing.
Question two

What are we required to do?

  • What the general duty covers.
  • What has to be assessed, and how often.
  • What guidance we are being measured against, formally or otherwise.
Question three

What would change?

  • What specifically we would do differently.
  • What it costs, against what is already being spent.
  • Who owns it and by when.
Question four

How will we know it worked?

  • What we will measure before and after.
  • What result would tell us to stop.
  • What we will not be able to attribute, and why.

Question four is the one that fails proposals, and it fails them at the second meeting rather than the first. A programme that cannot say in advance what would count as it not working is difficult to fund twice.

What the law actually requires

Less specific than most people expect, and harder to satisfy with a purchase than most people hope. Two provisions carry the weight.

The general duty under the Health and Safety at Work etc. Act 1974 covers health and welfare at work, so far as is reasonably practicable, and is not confined to physical safety. The Management of Health and Safety at Work Regulations 1999 require a suitable and sufficient assessment of the risks to employees, and psychosocial risk is not carved out of that.

What the requirement does not do is tell you how, and that cuts both ways.

  • No instrument is prescribed. The regulations require a risk assessment, not a particular questionnaire, and not a particular vendor.
  • No frequency is prescribed. "Suitable and sufficient" is the test. A risk assessment that describes an organisation as it was two years ago is a weaker answer to that test than one that does not.
  • Buying a service is not the same as discharging a duty. The Court of Appeal in Daw was explicit that a counselling service is not by itself an answer. That is the single most useful sentence to put in front of a board that thinks the EAP settles it.
  • What you knew matters. The Hatton line of cases turns on what the employer knew or ought to have known. A dated record of what you asked, what you were told and what you did has a value that is not really about wellbeing at all.

This is a description of where the duties come from, not advice about your own position. Whether a duty applies, and what satisfying it looks like for your organisation, is a question for your legal and health-and-safety advisers. ISO 45003 is guidance and is not certifiable; following it is neither a requirement nor a defence.

What the evidence base supports

The useful research question is not whether wellbeing matters. It is which features of work predict harm, because those are the ones an employer can change, and they are the ones a board can be asked to fund.

The job demands-resources tradition is the workhorse here: strain rises when the demands of a job outrun the resources available to meet them, and resources include control, support, clarity and fair treatment as well as headcount. It is useful to a board because every term in it is something an organisation decides rather than something a workforce is.

The HSE Management Standards put that into six areas an assessment is expected to cover, and they are a reasonable spine for a board paper.

  • Demands: workload, work patterns, the environment.
  • Control: how much say people have in how they work.
  • Support: from the organisation and from colleagues.
  • Relationships: conflict, and how unacceptable behaviour is dealt with.
  • Role: whether people understand their role and whether it conflicts with itself.
  • Change: how organisational change is managed and communicated.

Two things follow that are worth saying out loud in the paper. The first is that these are properties of jobs, so the interventions are changes to work rather than services bought for people. The second is that they are measurable with published instruments rather than invented ones, which is what makes a before-and-after comparison mean anything.

What to actually put in front of them

One paper, four parts, in the order the questions were asked. The failure mode is a deck that is mostly context; the useful version is mostly record.

1
Part one
What we know, and how

The current picture of the workforce, the date it was taken, the method, and the parts where you are relying on judgement rather than evidence. Say which is which. A board reads an unhedged claim as either naive or sales, and neither helps you.

2
Part two
What we are required to do

The duties, the assessment requirement, and the guidance you are being measured against. Short. This section exists so that nobody in the room is deciding whether workplace health is optional.

3
Part three
What we propose, and what it displaces

The specific changes, the owner of each, the date, and the cost set against what is already being spent. Most organisations are already funding a good deal of this through insured benefits nobody uses, and a line comparing the two is more persuasive than a new ask.

4
Part four
How we will know

The measure, the cadence, the comparison you will make, and the result that would tell you to stop. Include what you will not be able to attribute. Volunteering the limits of your own evidence is the thing that makes the rest of it credible.

5
Standing
Who owns it between meetings

A named owner with the authority to change something, a group that reviews the data on a stated cycle, and a decision log. Governance is the part that survives a reorganisation, and it is usually the part missing when a programme quietly stops.

This is the record the Alltoogether platform is built to produce: measurement with its limits stated, dated evidence, sourced comparisons only, and nothing below five people.

What measurement cannot do

Put this section in the paper. It is the one that gets the programme funded a second time, because a board that has been oversold once will discount everything that follows.

Five limits, all of them real.

  • It cannot establish cause. Two series that move together are a pattern worth a conversation, not proof that one caused the other. Anything that promises attribution is promising more than the method supports.
  • It cannot tell you about a named person. Nor should it. Organisation-wide reporting behind a minimum group size is what makes people answer honestly, and honest answers are the entire value of the exercise.
  • It cannot benchmark you against your sector. National statistics are household surveys with no employer in them and self-defined industry classifications. They are context. Your own baseline is the comparison that means something.
  • It cannot predict. A measure of how work feels now is not a forecast of absence or turnover later, and treating it as one is how a programme loses its credibility in year two.
  • It is not clinical. A workplace measure is not a diagnosis, not screening, and not a substitute for occupational health or clinical care.

If a supplier offers you individual-level insight, prediction, or a sector benchmark drawn from their own client base, ask what the sample is and what the confidence interval is. Ask us the same question. The answers are the fastest way to tell a measurement programme from something that only looks like one.

Questions people ask

Is measuring workplace health a legal requirement?

Assessing risk is. The Management of Health and Safety at Work Regulations 1999 require a suitable and sufficient assessment of the risks to employees, and psychosocial risk is not excluded from that. What is not prescribed is the method, the instrument or the frequency, so how you assess it is a judgement, and whether your judgement is sufficient is a question for your own advisers.

Does having an employee assistance programme cover us?

Not on its own. The Court of Appeal in Intel v Daw was explicit that a counselling service is not a panacea by which an employer discharges its duty in all cases. An EAP is a support route. It is not an assessment, and it is not a record of what you knew and what you did.

Can we compare ourselves to our sector?

Only loosely, and it is usually the weakest part of a board paper. The published national figures come from a household survey with no employer in it, and the industry classification is chosen by the respondent. It is context. The comparison that carries weight is your own organisation against its own baseline.

What if the measurement shows something bad?

Then you know, and knowing is the position the law and the case law both treat as better than not knowing, provided you act on it. The risk that is worth worrying about is a programme that surfaces a problem and produces no decision, which is why the governance and the decision log matter more than the instrument.

Sources and limitations

What this page is, and is not

This page is general information about making an evidence-based case for workplace health, for people who have to justify it to a board or an executive group. It is not legal advice, not health and safety advice, and not a statement about what any particular organisation is required to do. Whether a duty applies to you, and what discharging it looks like, depends on your circumstances and is a question for your own legal and health-and-safety advisers. Alltoogether is an appointed representative of Sante Partners Ltd, which is authorised and regulated by the Financial Conduct Authority.

legislation.gov.ukHealth and Safety at Work etc. Act 1974, s.2
Supports: that the general duty on an employer covers health and welfare at work, so far as is reasonably practicable, and is not limited to physical safety
Published or updated: 1974, as amended. Retrieved: 16 August 2026.
Limitation: the statutory duty is qualified by what is reasonably practicable, which is decided on the facts of a case and not by any general standard.
legislation.gov.ukManagement of Health and Safety at Work Regulations 1999, reg. 3
Supports: that a suitable and sufficient assessment of the risks to employees' health is a legal requirement, which is the hook the measurement argument hangs on
Published or updated: 1999, as amended. Retrieved: 16 August 2026.
Limitation: the regulation requires a risk assessment. It does not prescribe an instrument, a frequency, or any particular method of measuring.
Health and Safety ExecutiveManagement Standards for work-related stress
Supports: the six areas of work design that a psychosocial assessment is expected to cover
Published or updated: current guidance. Retrieved: 16 August 2026.
Limitation: guidance, not legislation. HSE states it does not expect an employer to meet all six standards.
International Organization for StandardizationISO 45003:2021, psychological health and safety at work
Supports: that there is an international guidance standard for managing psychosocial risk, which a board can be pointed at
Published or updated: 2021. Retrieved: 16 August 2026.
Limitation: guidance, and not certifiable in the way ISO 45001 is. Following it is not a legal requirement and not a defence.
Court of AppealHatton v Sutherland [2002] EWCA Civ 76
Supports: that what an employer knew, or ought to have known, is central to whether a duty was breached, which is why a record of what you knew and when has value beyond the wellbeing programme
Published or updated: 2002. Retrieved: 16 August 2026.
Limitation: one authority in a line of cases, and later qualified. See Daw below.
Court of AppealIntel Corporation (UK) Ltd v Daw [2007] EWCA Civ 70
Supports: that providing a counselling service is not by itself a discharge of the duty, which is the reason a board should ask what changed rather than what was bought
Published or updated: 2007. Retrieved: 16 August 2026.
Limitation: decided on its own facts. It does not make counselling irrelevant, only insufficient on its own.
Office for National StatisticsSickness absence in the UK labour market, 2025 edition
Supports: that national absence context exists and is published, for a board that asks how the organisation compares
Published or updated: released 1 May 2026. Retrieved: 16 August 2026.
Limitation: a household survey with no employer in it, and industry is self-defined by the respondent. It is national context and not a benchmark for any organisation, which is the point made on this page.
AlltoogetherHow this page was written
Supports: the structure of the argument and the decision to quote no statistic
Published or updated: rebuilt 16 August 2026. Retrieved: 16 August 2026.
Limitation: this page is about how to make a case, not a source of figures. Where a number would be persuasive it is a number about your own organisation, and this page cannot supply it.

Related guides

ZF

Zak Fenton · Founder, Alltoogether

Written by Zak Fenton, MSc Workplace Health and Wellbeing (Distinction), founder of Alltoogether, a UK employee-benefits broker and workplace-health platform. Alltoogether is an appointed representative of Sante Partners Ltd, which is authorised and regulated by the Financial Conduct Authority (914023).

Last reviewed 16 August 2026.

Where your evidence
would fall over

The Strategy Audit works through what you currently know about your workforce, what you are required to assess, and which parts of that a board would push back on. It is a structured review with an output you can take to a meeting, not a sales call with a document attached.

Open the Strategy Audit →