How To Quantify The Productivity Impact Of Poor Mental Health At Work

Most CHROs cannot answer a simple question when Finance asks it: what is poor mental health actually costing us in lost productivity, and how do we know? Not because the cost isn't real, but because the data needed to size it usually sits in five different systems, owned by five different teams, none of which were built to talk to each other.

This guide sets out a repeatable method for quantifying that cost and building a defensible business case around reducing it, using organisational-level measurement rather than anecdote or industry averages that don't reflect your workforce.

The honest starting point is that most organisations aren't measuring this at all. Engagement surveys ask how people feel. Absence data sits with occupational health. Performance data sits with line managers. None of it is stitched together into a single view of how stress, burnout and poor mental health are showing up in output, error rates, absence and attrition.

That fragmentation is the actual problem, and it predates any specific number. An organisation running a legacy Employee Assistance Programme, a separate wellbeing app and a manual engagement survey has three data sets that were never designed to be joined. Each vendor can tell you about their own utilisation. None of them can tell you what changed for the business.

This is also why so many CHROs default to citing external, industry-wide statistics in a board deck rather than their own numbers. It's an understandable shortcut, but it's the wrong one. A generic figure from outside your organisation says nothing about your workforce, your industry, or your risk factors, and it evaporates the moment a CFO asks where it came from and whether it applies here. The credible business case is built from your own data, measured over your own population, over a defined period.

Quantifying the productivity impact of poor mental health follows the same basic structure regardless of company size or sector. It has four steps.

Step 1. Establish a baseline. Before you can show movement, you need a starting point on the actual drivers of workforce wellbeing: stress levels, workload pressure, psychological safety, and managerial and organisational support. A diagnostic assessment that measures these drivers directly, and compares the result to external benchmarks, gives you a defensible starting line rather than a guess. This is the function Insights & Assessments performs at Unmind: real-time dashboards and a diagnostic assessment that surface stress signals, resilience scores and risk factors by team and region, benchmarked against comparable organisations.

Step 2. Segment by risk, rather than just by average. An organisation-wide average tells you almost nothing useful. The instructive number is what's happening among your highest-stress population, because that's where the productivity drag concentrates and where an intervention has the most to recover. Segmenting by team, region or risk factor is what turns a wellbeing score into a targeting tool.

Step 3. Connect the wellbeing data to business metrics. This is the step most organisations skip. Wellbeing scores need to be set against absence data, engagement and, where available, work-impairment measures that combine presenteeism and absenteeism into a single productivity figure. Without this connection, you have two data sets sitting next to each other rather than one story.

Step 4. Calculate the projected financial impact and re-measure. Once the wellbeing and business metrics are connected, the productivity gap can be converted into an estimated financial impact, and the whole cycle repeated at intervals so leadership can see whether the trend is moving in the right direction. This is what makes the business case durable rather than a one-off deck for a single budget cycle.

The clearest evidence for this model isn't theoretical. A global FMCG employer, operating across four countries (Canada, Mexico, the Netherlands and India), ran a six-month research study to test exactly this question: could a digital mental health platform move the needle on wellbeing, and would that translate into measurable business outcomes.

The results were specific and traceable. Among high-stress employees, stress reduced by 16%, and overall work impairment (the combined measure of absenteeism and presenteeism) fell by 10%. Modelled against those improvements, the study projected a 3.49x return on investment and an estimated $1M impact from improved productivity.

What makes this study useful for a business case isn't the headline ROI multiple on its own. It's the structure behind it: a defined population, a baseline, a measurement period, and a business-metric conversion at the end. That's the same four-step model above, run at enterprise scale.

For most enterprise employers, the productivity drag isn't invisible because nobody is spending on mental health. It's invisible because the spend already in place isn't built to prove anything back.

The category norm for a traditional EAP is under 2% engagement, and a legacy EAP running at 2 to 4% utilisation is providing legal cover rather than care. That gap matters because it's not a wellbeing statistic, it's a productivity one. A benefit line that almost nobody uses cannot be reducing presenteeism, absence or attrition at any meaningful scale, no matter what the contract says on paper. The cost of poor mental health keeps accruing underneath a benefit that technically exists but functionally isn't reaching the people who need it.

This is the gap that a measurement layer closes. Rather than reporting how many people logged in, organisational-level insights connect wellbeing trends to the metrics a CFO actually cares about: absenteeism, turnover and productivity. That's the difference between a utilisation report and a business case.

Once the baseline is established and the wellbeing-to-business-metric link is in place, the ROI conversation becomes concrete rather than aspirational. Data from the Unmind platform shows a 4.6x return for every $1 invested. That figure holds because it's built on the same underlying logic as the FMCG study above, wellbeing improvement converted into a measurable productivity return, rather than an industry-wide estimate applied to a workforce it was never measured against.

For a Finance partner reviewing this business case, the questions to ask of any vendor claim are the same ones this guide has been building toward. What population was measured. Over what period. Against what baseline. Converted into which business metric, using what method. A number that can answer all four is usable in a board deck. A number that can't should be treated as marketing rather than evidence.

Measurement only pays off if it changes what happens next. Two things matter most once the diagnostic data is in hand.

The first is making support easy enough to reach that people actually use it before a problem escalates. Nova, the AI coach built by Unmind, gives employees confidential, in-the-moment support and a pathway into human care, therapy or coaching when it's needed. The retention data suggests this matters more than it might sound: employees who engage with Nova show 49% monthly retention on the platform, compared with 39% for those who don't, which is a meaningful signal that easy-access support keeps people engaged with the tools that are meant to help them.

The second is equipping the people closest to the problem before it shows up in a productivity number. Managers are usually the first to notice a team member struggling, long before it appears in an absence report or an engagement score, which is why manager capability is often as consequential to the productivity outcome as the individual-level support. For more on how manager training specifically connects to business impact, see this guide to the business impact of mental health training for leaders, published by Unmind.

If your organisation doesn't yet have a baseline for how stress, workload pressure and psychological safety are affecting productivity, that's the first gap to close, rather than the ROI calculation. Insights & Assessments from Unmind is built specifically to establish that baseline: a diagnostic assessment and real-time dashboard that segments risk by team and region and benchmarks it externally, so the business case starts from your own data rather than an industry average.

From there, the ROI page from Unmind sets out how wellbeing improvement converts into a financial return, using the same evidence base referenced throughout this guide, so Finance and HR are working from one shared model rather than negotiating between two.

The organisations that get funding for this work aren't the ones with the biggest external statistic. They're the ones who can show their own baseline, their own movement, and their own number.

• The productivity impact of poor mental health is hard to quantify because the relevant data (wellbeing, absence, performance) usually sits in disconnected systems, rather than because the cost isn't real.
• A repeatable model has four steps: establish a wellbeing baseline, segment by risk, connect wellbeing data to business metrics, and calculate and re-measure the financial impact.
• A four-country FMCG research study using this model found a 3.49x projected ROI and an estimated $1M productivity impact, driven by a 16% stress reduction and a 10% reduction in work impairment among high-stress employees.
• Legacy EAP utilisation sitting at 2 to 4%, against a category average under 2% engagement, means a large share of existing mental health spend is not reaching the people driving the productivity cost.
• Data from the Unmind platform shows a 4.6x return per $1 invested, built on the same wellbeing-to-business-metric conversion as the model above.
• Reducing the impact, rather than just measuring it, depends on accessible support (Nova shows 49% monthly retention versus 39% for non-users) and equipped managers, who are usually the first to notice a problem before it reaches a productivity report.