Most people who advocate for their colleagues have the easy arguments covered. Investing in people feels good, and it looks good, internally and to the outside world. The trouble is that the person you most need to convince has heard the feel-good case before and has learned to discount it to zero. It sounds like advocacy, and advocacy is exactly what a budget-holder is trained to resist.

We ran into this ourselves while building a cost model for proactive mental health support. So we went looking for a way to make the argument in language a CFO can't dismiss. What we found wasn't a better statistic. It was a better frame, and a method that follows from it. This is how we built ours, and how you can build your own.

Here's what's below. First, the method: the one reframe that changes the conversation, and the four steps we used to turn it into numbers. Then, at the end, two things to take with you. A free Excel calculator you can download and plug your own numbers into, so you don't have to build it by hand. And the story of how we used this exact method to make our own case for Proactive EAP, so you can see it work on a real decision rather than in the abstract.

The one move that changes the conversation

Here is the whole thing in a sentence. Stop counting what it costs to act, and start counting what it costs to do nothing.

Almost every people-and-culture business case is built the wrong way round. It leads with the price of the program, and the price of a program is a cost, and a CFO's entire job is to minimise costs. You've framed yourself as something to be reduced before you've finished your first slide.

Turn it around. The money is already being lost. People in distress already cost the business, through claims, through absence, through the quiet underperformance of people who show up but can't fully function, through the ones who leave. That loss is being paid right now, every year, whether or not anyone funds a single initiative. It just never appears as a line item, so nobody sees it.

Once you show the loss, the program stops being a cost and becomes what it actually is: a way to recover money already leaving the building. The question in the room changes from "can we afford to do this" to "can we afford to keep paying for the alternative." That is the move. Everything below is how to make it concrete.

Step one: break the loss into pillars, not one scary lump

A single big number is easy to dismiss. "$220 billion a year across Australia" is the kind of figure a sceptic bats away in a sentence, because it's about the whole country, not about them. The trick is to break the loss into distinct, recognisable pieces, each one something the reader can picture happening in their own organisation.

We settled on five. You can adapt these, but they cover the ground:

One: claims and workers' compensation. The direct, visible cost when something goes wrong badly enough to be formally recorded. In Australia, work-related mental health claims keep people off work about five times longer than the average serious claim, roughly 35 weeks. (Safe Work Australia, 2022–23.) These are the costs your finance team can already see, which makes them the easiest place to start.

Two: absenteeism and presenteeism. The bigger, quieter one. Presenteeism, people at their desks but unwell and underperforming, is the single largest cost of poor mental health to employers. Deloitte found it accounts for more of the total than absenteeism and turnover combined. (Deloitte UK, 2024.) This is the pillar most business cases miss, precisely because it never shows up in any system.

Three: turnover and replacement. When people leave, replacing them costs somewhere between 50% and 200% of their annual salary depending on the role. (Widely replicated across HR research.) Poor mental health and low engagement are direct drivers of avoidable turnover.

Four: insurance and premiums. Claims experience feeds premiums. A worse claims history costs more to insure, year after year, long after the original incident.

Five: disputes and culture cost. The claims that turn adversarial, the disputes that consume management time, the corrosive effect on everyone watching how a struggling colleague gets treated.

Breaking it into pillars does two things. It makes the loss specific enough to feel real, and it lets you use the low end of every published range, so you can look the sceptic in the eye and say every figure here is conservative. That sentence wins more arguments than any single number.

Step two: ground each pillar in a number you can actually use

This is where most articles hand you a global statistic and leave you to figure out what it means for you. The macro figures are real, but on their own they invite the easiest dismissal there is: that's the world, that's not us. So the job is to translate each one down to your own headcount.

A few of the most useful findings, and how to localise them.

On the total cost. Australia's Productivity Commission put the cost of mental ill-health at $200–220 billion a year, with the workplace-specific slice at $12.2–39.9 billion. (Productivity Commission, 2020.) Useful for context, useless for your business case on its own. Don't lead with it.

On return. This is the number to reach for when someone asks what they get back. In Australia, PwC and beyondblue found every $1 invested in a mentally healthy workplace returns about $2.30 on average. (PwC, 2014.) Deloitte's UK work puts it at £4.70 per £1 across 26 studies. (Deloitte, 2024.) The WHO estimates $4 per $1 on scaled-up treatment for depression and anxiety. (Chisholm et al., Lancet Psychiatry, 2016.)

The way to use a return multiple:

Proposed spend × conservative multiple = projected return $50,000 program × 2.3 = ~$115,000 in projected benefit

State plainly that the multiple assumes the program is implemented well, and that you've used the most conservative published figure. Overclaiming here is how you lose credibility in one line.

On prevalence, localised. Around one in five working-age people experience a mental health condition in a given year. Turn it into your own team:

Headcount × 20% = people on your team likely affected this year 100 employees × 20% = ~20 people

Suddenly it isn't a national statistic. It's twenty people the reader can almost name.

On engagement, localised. Gallup's global disengagement figure ($8.8–10 trillion, 9% of GDP) is the most dismissible number in the whole field, because it's the most abstract. Don't quote the trillions. Use the per-person finding instead: disengaged employees run roughly 18% less productive with markedly higher absence. (Gallup, State of the Global Workplace.) Applied to a salary:

Salaries of disengaged staff × ~18% = productivity lost One $85,000 role, disengaged ≈ ~$15,000 of lost output a year

That is a number a CFO feels, because it's denominated in one salary they recognise, not the GDP of Japan.

The principle across all of these: never hand over a figure you haven't divided down to their people and their dollars. The macro number is the evidence. The localised number is the argument.

Step three: give it a total and a per-employee cost

Once you've built the pillars for your own organisation, add them up, then do one more thing that matters more than it should. Divide the total by headcount.

A total does the heavy lifting for scale. A per-employee figure does the heavy lifting for relatability. "We're carrying roughly this much in avoidable people-cost each year, which is about X per employee" is a sentence that works in two directions at once. The total makes it serious. The per-head number makes it impossible to file under "someone else's problem," because everyone in the room can multiply it by their own team in their head.

Per-employee framing also travels. A manager who hears "about X a head" can apply it to their own eight people without any further help from you. You've handed them a portable version of your argument that they can carry into rooms you're not in.

Step four: forecast it, because inaction compounds

The last thing we learned, and the one that surprised us most, is that the cost of doing nothing is not a flat annual figure. It grows.

An unaddressed culture problem doesn't hold steady. Claims accumulate and push up premiums. The people with the most choices leave first, so the loss concentrates. Distress spreads through teams. Each year of inaction makes the next year more expensive, while a proactive investment does the opposite: it costs a little up front and then bends the curve down as prevention compounds in your favour.

You don't need false precision to show this. An illustrative index makes the point cleanly. Set this year's cost of inaction at 100 and show the two paths:

Do nothing: Year 1 = 100, Year 2 ≈ 145, Year 3 ≈ 210, Year 4 ≈ 295, Year 5 ≈ 410 Invest now: Year 1 = 100, Year 2 ≈ 102, Year 3 ≈ 98, Year 4 ≈ 92, Year 5 ≈ 85

The exact slope is arguable and you should say so. The shape is not. One line climbs and one line flattens, and by year five they are a long way apart. A business case that shows only this year's number is arguing on the sceptic's terms, where the spend looks large against a single year's loss. A business case that shows the five-year divergence reframes the spend as the cheaper path, which is what it is.

The calculator: don't do this by hand

We built the four steps into a spreadsheet, because doing the arithmetic for a specific organisation is fiddly and we'd rather you spent your energy on the argument than the sums. You put in your team size, your salary, and your current level of support, and it returns the five pillars, a total, a per-employee figure, and the five-year forecast, with a chart and every source attached. It's free to download and yours to keep.

It's deliberately conservative. Every assumption sits on the low end of the published range, and they're all listed on their own tab so you can see exactly what each number rests on and adjust anything you'd defend differently. The point isn't to hand you our figure. It's to save you the build, so you can walk in with your own.

Free Excel calculator: the cost of inaction

Put in your team size, average salary, and current level of support. It returns the five pillars, a total, a per-employee figure, and the five-year forecast — with a chart and every source attached. Deliberately conservative, and yours to keep.

Prefer a quick on-screen estimate before you open the spreadsheet? Our Hidden Cost of Inaction calculator answers four questions and returns a live figure for your organisation in seconds.

How we used it: our Proactive EAP case

We didn't write this in the abstract. This is the method we used to make our own case for shifting from reactive to proactive mental health support.

The old model waited. It waited for someone to reach crisis, notice it, find the number, and call, at the point they were least able to do any of it. Framed as a cost, proactive support was an easy no: more money, for something that looked like what we already had.

So we turned it around. We costed what the reactive model was already losing, across the five pillars, and then we showed the two curves side by side: the reactive cost climbing as claims and turnover compounded, against the proactive cost holding flat and then bending down. The proactive investment didn't show up as new spending. It showed up as the cheaper of two paths we were already choosing between, whether we admitted it or not.

That's the case the numbers made for us. The tool above is how we made it, handed to you to make yours. If you want to see the model applied end to end, our Proactive EAP page walks through the same reframe as a live service.

The method is the thing to keep

The calculator is just the four steps automated. The method is what's worth holding onto, and it's yours to use with or without our tool: turn the cost into a loss, break the loss into pillars, make it relatable with a per-employee number, and show how it compounds. Do that, and you walk into the room with the one thing the feel-good case never had. A number the CFO can't wave away, because it was their kind of number all along.

The Cost of Poor Mental Health at Work: How to Build the Business Case (With a Free Calculator)