The Milking Stool

Several weeks ago, I came across the idea of the “happy–productive worker” and began reading more. Since then, I have disappeared down a variety of theoretical and empirical rabbit holes: job satisfaction, happiness and performance, work design, human capital, hedonic adaptation, sustainable careers, and the evidence behind different workplace interventions. Along the way, a more fundamental question kept emerging: what are workplace wellbeing programmes actually for?

This blog is the culmination of what I found down those rabbit holes and of some of the reflexivity that accompanied them. It explores whether workplace wellbeing is intended to restore lost functioning, protect people from harm, enhance performance, or sustain healthy and productive working lives, and why clarity about that purpose matters for the interventions we choose, the outcomes we measure, and the responsibilities placed on organisations and managers.

---

Workplace wellbeing has become an extraordinarily broad category. Under the same heading, an organisation might offer counselling to an employee with depression, redesign an overloaded job, provide mindfulness training, develop managers, introduce flexible working, subsidise exercise, improve career development, or organise a wellbeing day. These interventions may all have value, but they clearly do not do the same thing. That raises a deceptively simple question that should perhaps precede any discussion about return on investment: What is the wellbeing intervention actually for?

Is it intended to restore functioning that has already been lost? To prevent deterioration in the first place? To enhance functioning above its existing level? Or to enable healthy and effective functioning to remain sustainable over years or decades?

Without answering that question, organisations risk comparing fundamentally different interventions, measuring the wrong outcomes, and, perhaps most importantly, claiming evidence for purposes the intervention was never demonstrated to achieve.

Wellbeing is not a single outcome

Part of the difficulty is conceptual. Terms such as wellbeing, happiness, job satisfaction, engagement, mental health and psychological functioning are often used as though they describe essentially the same thing. They do not. Cropanzano and Wright (2001), for example, observed that research on the “happy–productive worker” had operationalised happiness in multiple ways, including job satisfaction, positive affect, the absence of negative affect, the absence of emotional exhaustion, and psychological well-being. Different operationalisations produced different relationships with performance. Job satisfaction is primarily an evaluation of one’s job, not simply an emotional state. This matters because the apparently straightforward claim that “wellbeing improves productivity” isn’t fully explained or supported by scientific evidence.

Which aspect of wellbeing? Which aspect of productivity? Over what period? And through what mechanism?

The evidence illustrates the problem. Judge et al. (2001), in a meta-analysis involving 312 samples and more than 54,000 workers, estimated a corrected correlation of around .30 between overall job satisfaction and job performance. However, Bowling (2007) later found that this connection became much weaker once factors such as personality and work-related self-evaluations were taken into account. It nearly disappeared altogether after considering organisation-based self-esteem (OBSE).  Neither finding means that employee experience is irrelevant to performance. Rather, they raise a more interesting possibility that good work produces both wellbeing and performance.  Perhaps wellbeing and performance sometimes improve together because the same working conditions make both possible.  That is quite different from saying: [make workers happy] → [workers produce more].

Four different jobs for workplace wellbeing

A useful starting point is to distinguish four objectives:

The distinctions are not absolute. A good work-redesign intervention might simultaneously PROTECT, ENHANCE and SUSTAIN. Identifying the primary objective forces organisations to specify the causal claim they are making, which makes evidence easier to interpret.

RESTORE: return to work is not necessarily recovery

There is strong evidence that well-designed interventions can improve occupational outcomes for people experiencing mental ill-health. For example, Reme et al. (2015) evaluated work-focused CBT combined with individual job support in a large multicentre randomised controlled trial involving people with common mental disorders. The intervention increased or maintained work participation compared with usual care. Lerner et al. (2020) similarly found that adding a work-focused intervention to integrated depression care improved depressive symptoms and reduced self-reported at-work productivity loss compared with integrated care alone. These are important findings, but they do not justify the much broader proposition that “Providing psychological therapy restores productivity”. Clinical recovery, work participation and work functioning are distinct outcomes. Arends et al. (2019) tracked workers with common mental health problems after they returned to work and found that recovery patterns were quite varied. Forty-two per cent belonged to a slow-recovery group characterised by persistent high symptoms and moderate-to-low work functioning despite a relatively rapid return to work. This suggests that [return to work] ≠ [symptom recovery] ≠ [restored functioning] ≠ [restored productivity]. This also raises questions about the provision of generic Employee Assistance Programmes. Conventional CBT may be an evidence-based treatment for particular mental health conditions. But CBT in which work is discussed is not necessarily equivalent to an intervention deliberately designed to restore occupational functioning. We should remember that an intervention can be evidence-based for one objective without being evidence-based for another.

PROTECT: why wait for functioning to deteriorate?

RESTORE becomes ethically and economically problematic when it substitutes for prevention. The Job Demands–Resources model offers a useful explanation. Demands can contribute to strain and health impairment, whereas resources have motivational functions and can buffer the impact of demands (Demerouti et al., 2001; Bakker and Demerouti, 2007). Conservation of Resources theory similarly proposes that stress arises particularly from the actual or threatened loss of valued resources and provides a basis for understanding potentially compounding processes of resource loss (Hobfoll, 1989). An organisation might therefore operate an impressive pathway, but when considered more broadly, it is actually a disappointing circle:

Becoming highly efficient at the 6 o’clock position in the circle is not the same as having a good wellbeing strategy. Sometimes the more important intervention is upstream or at the 12 o’clock position in this circle, i.e., removing or reducing excessive demand. This is where evidence of organisational-level intervention matters. Fox et al. (2022), in a review of 83 organisational- and group-level experimental and quasi-experimental studies, examined interventions involving scheduling, job/task changes, relational dynamics, and participatory processes. The evidence is mixed, and context matters. Importantly, it demonstrates that the workplace itself can be an intervention target. 

ENHANCE: happiness is not a productivity technology

There is nevertheless credible evidence that positive psychological states can influence performance. Oswald, Proto, and Sgroi (2015) experimentally induced positive affect and found approximately a 12% increase in productivity on a piece-rate task among treated participants. This provides unusually strong causal evidence, albeit in a deliberately constrained experimental setting. Bellet, De Neve and Ward (2024) provide complementary field evidence from BT call centres. Using repeated measures of happiness, detailed administrative performance data, and quasi-experimental variation associated with weather and visual exposure to it, they found positive effects of happiness on sales productivity. The effects were driven primarily by workers converting more calls into sales, with smaller effects through calls per hour and schedule adherence; they did not find effects on attendance or break-taking. These studies matter. But neither demonstrates that any intervention labelled “wellbeing” will increase productivity. The evidence chain would need to establish [specific intervention] → [specific psychological construct changes] → [relevant performance outcome changes] → [effect persists sufficiently long to matter]. That is a much higher evidential bar.

Work design may offer a more direct route. Hackman and Oldham (1976) demonstrated that characteristics such as autonomy, feedback, skill variety, task identity, and task significance can influence psychological states associated with intrinsic motivation. Deci and Ryan’s (2000) Self-Determination Theory similarly identifies autonomy, competence, and relatedness as important conditions for psychological growth, wellbeing, and effective functioning. More recently, Knight and Parker’s (2021) systematic review identified 55 heterogeneous work-redesign intervention studies: 39 reported positive performance effects, two reported negative effects, and 14 reported mixed effects. Motivation, quick action, and learning emerged as key factors, while implementation and the overall organisation’s situation served as important constraints. This suggests a different question: How can we design work that enables both people and organisations to function well?

SUSTAIN: wellbeing needs a time dimension

Most wellbeing conversations implicitly concern the present. Yet a worker can be healthy and productive today while working in a way that is unlikely to remain viable for another five years. De Vos, Van der Heijden, and Akkermans (2020) offer an important counterpoint with their sustainable careers model. They conceptualise career sustainability dynamically across person, context and time, with health, happiness and productivity as three core indicators. That shifts the question again. SUSTAIN does not necessarily mean “keep somebody performing at today’s level in today’s job indefinitely”. A sustainable trajectory may involve development, adaptation, rebalancing, recovery, and/or changes in responsibilities to enable continued contribution. Sometimes sustaining a career requires changing it. This is particularly important for experienced workers whose knowledge, judgement and tacit expertise have accumulated over decades.

Human capital: what exactly are we preserving?

Human Capital Theory adds an economic dimension. Schultz (1961) and Becker (1964) conceptualised investments in areas such as education, knowledge, skills and health as investments in productive human capacity. From this perspective, the four objectives can also be understood as follows:

• RESTORE = recover the capacity to deploy existing human capital.
• PROTECT = prevent avoidable erosion or underutilisation of accumulated human capital.
• ENHANCE = develop additional capability.
• SUSTAIN = keep accumulated capability viable and deployable over time.

This exposes a curious contradiction in some organisations. They may spend decades investing in someone’s education, professional development, experience, tacit knowledge and relationships, while simultaneously operating working arrangements that progressively deplete the health and psychological resources needed to deploy that accumulated capability. A RESTORE-heavy wellbeing strategy may therefore represent poor human-capital management as well as poor prevention. There is, however, an essential ethical qualification. Workers are not assets owned by organisations. Indeed, Schultz himself recognised the moral difficulty inherent in treating people as capital. The economic argument can strengthen the business case for prevention. It cannot provide its complete ethical justification. An entry-level worker does not have a weaker entitlement to protection from avoidable harm merely because they are cheaper to replace.

Hedonic adaptation: what remains when the novelty wears off?

There is another reason to be cautious about interventions designed primarily to “make people happy”. People adapt. Hedonic adaptation complicates rather than invalidates workplace wellbeing interventions. Diener, Lucas and Scollon (2006) demonstrate that adaptation is neither complete nor uniform: different dimensions of wellbeing and different individuals follow different trajectories, and lasting changes in wellbeing can occur. Consequently, workplace interventions should not be judged solely by their initial effect on subjective wellbeing—or by whether that initial uplift persists unchanged.

A structural intervention may cease to feel novel while continuing to reduce exposure, preserve health or support effective functioning. This principle raises an important workplace question: What remains when the novelty has worn off? A wellbeing day may initially feel excellent. A new benefit may initially increase satisfaction. A recognition initiative may initially generate positive affect. But novelty can diminish. By contrast, workers may also adapt psychologically to adequate staffing, reasonable workloads, autonomy, good supervision and flexible working. They may stop noticing these things. Yet these conditions can continue to produce substantial functional benefits. That gives us a crucial distinction: experienced benefit is not the same as functional benefit. Nobody needs to remain perpetually delighted that their workplace is adequately staffed for adequate staffing to remain valuable. For PROTECT and SUSTAIN interventions, becoming normal may actually be evidence of successful institutionalisation rather than intervention failure.

Managers should not be amateur happiness engineers

This has an important implication for managerial development. If organisations accept the simplified proposition that [happy workers] = [productive workers], they may conclude that managers should be responsible for making their teams happy. That is a mistake. Managers have neither the competence nor the legitimate authority to determine another adult’s emotional state. Happiness reflects not only work but also health, personality, relationships, expectations, values and events outside the workplace. Even worse, making happiness a managerial KPI can create perverse incentives. A manager may sometimes need to challenge unsafe behaviour, manage poor performance, say no, resolve conflict, allocate unpopular work or deliver difficult feedback. Competent management can therefore produce legitimate short-term dissatisfaction. If [good manager] = [happy employees], then [employee displeasure] → [perceived managerial failure] → [avoidance of difficult conversations] → [appeasement]. This can turn wellbeing management into people-pleasing management. Managers should instead be architects of the conditions for good work. Negative emotion should be treated as information rather than as something that automatically requires correction. Sometimes dissatisfaction is an entirely appropriate response to badly designed work.

Managers are not therapists either

The distinction also protects professional boundaries. A manager noticing distress might appropriately notice, ask, listen, consider work factors, and then act where appropriate, signposting or referring. They should not normally proceed through a notice, diagnose, formulate, counsel and treat pathway. A manager can be psychologically informed without becoming a psychologist. Similarly, an organisation can promote wellbeing without attempting to manage employees’ emotional lives.

The occupational health clinician has a different responsibility again

This distinction is particularly important in occupational health. The Faculty of Occupational Medicine’s current Ethics Guidance for Occupational Health was published in 2026. Its framework reinforces the professional independence of occupational health advice and the importance of evidence, competence and ethical consideration. This means occupational health should be wary of becoming merely the organisation’s highly efficient RESTORE department. Repeated referrals indicating excessive demand, impairment, absence, rehabilitation, return and recurrence may reveal the adequacy or inadequacy of organisational prevention. Occupational health clinicians cannot necessarily redesign organisations, but they can and should recognise, assess, communicate and appropriately escalate material work-related risks. Independent clinical opinion is not separate from the occupational health service: it is a core part of what is being commissioned. The organisation commissions the expertise; it does not determine the clinical opinion.

So what should organisations measure?

The framework suggests that using the same dashboard to measure every wellbeing intervention is conceptually flawed. A counselling intervention might appropriately be evaluated through clinical change, functioning and work participation. A workload intervention might require measures of exposure, health, error, absence and retention. A work redesign intervention might examine autonomy, learning, motivation, quality, and performance. A career-sustainability strategy requires something harder: health, happiness and productive capability over time. And an employee satisfaction score should not automatically be relabelled as a “wellbeing score”. The metric should follow the objective, not the other way around.

Four questions before buying another wellbeing intervention

Perhaps every wellbeing proposal should begin with four questions:

• What are we trying to do—RESTORE, PROTECT, ENHANCE, or SUSTAIN?

• What mechanism should enable this intervention to achieve that objective?

• What evidence shows that it affects the outcome we actually care about—not merely an adjacent one?

• What will we measure, and for how long to know whether the effect persists?

There should perhaps be a fifth one:

• Are we intervening at the cause—or simply becoming better at managing its consequences?

What is wellbeing for, actually?

There is no single correct answer. Organisations should support workers’ recovery from illness and protect them from avoidable harm. There is nothing inherently objectionable about helping healthy workers flourish or perform better. Both workers and organisations have legitimate interests in careers remaining healthy, rewarding and productive over time. The problem arises when those purposes become blurred. A tertiary intervention cannot demonstrate primary prevention. An improvement in symptoms cannot automatically demonstrate restored work functioning. An association between satisfaction and performance cannot demonstrate that increasing satisfaction will improve performance. A short-lived increase in positive affect cannot demonstrate sustainable wellbeing. And a positive return on investment cannot, by itself, establish that an intervention represents ethically adequate workforce health management. Perhaps the most useful shift is from asking: “Does workplace wellbeing work?” to asking: “What work are we asking wellbeing to do?” Once that question is answered, the evidence becomes easier to interpret, appropriate outcomes become clearer, and responsibilities can be allocated more sensibly. For managers, that means creating the conditions for good work rather than becoming amateur happiness engineers. For occupational health, it means distinguishing treatment and rehabilitation from prevention and retaining independent professional judgement. For organisations, it means recognising that protecting health, developing capability and sustaining careers are related but not interchangeable objectives. And for workers, perhaps most importantly, it means recognising that their health and wellbeing have value independently of whether making them happier generates another percentage point of productivity. A mature workplace wellbeing strategy should therefore be able to answer four very simple questions:

What are we restoring? What are we protecting? What are we enhancing? And can we sustain it?

---

References

Arends, I., Almansa, J., Stansfeld, S.A., Amick, B.C., van der Klink, J.J.L. and Bültmann, U. (2019) ‘One-year trajectories of mental health and work outcomes post return to work in patients with common mental disorders’, Journal of Affective Disorders, 257, pp. 263–270.

Bakker, A.B. and Demerouti, E. (2007) ‘The Job Demands–Resources model: State of the art’, Journal of Managerial Psychology, 22(3), pp. 309–328.

Becker, G.S. (1964) Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. New York: National Bureau of Economic Research.

Bellet, C.S., De Neve, J-E. and Ward, G. (2024) ‘Does employee happiness have an impact on productivity?’, Management Science, 70(3), pp. 1656–1679.

Bowling, N.A. (2007) ‘Is the job satisfaction–job performance relationship spurious? A meta-analytic examination’, Journal of Vocational Behavior, 71(2), pp. 167–185.

Cropanzano, R. and Wright, T.A. (2001) ‘When a “happy” worker is really a “productive” worker: A review and further refinement of the happy–productive worker thesis’, Consulting Psychology Journal: Practice and Research, 53(3), pp. 182–199.

Deci, E.L. and Ryan, R.M. (2000) ‘The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior’, Psychological Inquiry, 11(4), pp. 227–268.

Demerouti, E., Bakker, A.B., Nachreiner, F. and Schaufeli, W.B. (2001) ‘The Job Demands–Resources model of burnout’, Journal of Applied Psychology, 86(3), pp. 499–512.

De Vos, A., Van der Heijden, B.I.J.M. and Akkermans, J. (2020) ‘Sustainable careers: Towards a conceptual model’, Journal of Vocational Behavior, 117, 103196.

Diener, E., Lucas, R.E. and Scollon, C.N. (2006) ‘Beyond the hedonic treadmill: Revising the adaptation theory of well-being’, American Psychologist, 61(4), pp. 305–314.

Faculty of Occupational Medicine (2026) Ethics Guidance for Occupational Health Practice. 9th edn. London: Faculty of Occupational Medicine.

Fox, K.E., Johnson, S.T., Berkman, L.F., Sianoja, M., Soh, Y., Kubzansky, L.D. and Kelly, E.L. (2022) ‘Organisational- and group-level workplace interventions and their effect on multiple domains of worker well-being: A systematic review’, Work & Stress, 36(1), pp. 30–59.

Hackman, J.R. and Oldham, G.R. (1976) ‘Motivation through the design of work: Test of a theory’, Organizational Behavior and Human Performance, 16(2), pp. 250–279.

Hobfoll, S.E. (1989) ‘Conservation of resources: A new attempt at conceptualizing stress’, American Psychologist, 44(3), pp. 513–524.

Judge, T.A., Thoresen, C.J., Bono, J.E. and Patton, G.K. (2001) ‘The job satisfaction–job performance relationship: A qualitative and quantitative review’, Psychological Bulletin, 127(3), pp. 376–407.

Knight, C. and Parker, S.K. (2021) ‘How work redesign interventions affect performance: An evidence-based model from a systematic review’, Human Relations, 74(1), pp. 69–104.

Lerner, D., Adler, D.A., Rogers, W.H., Ingram, E. and Oslin, D.W. (2020) ‘Effect of adding a work-focused intervention to integrated care for depression in the Veterans Health Administration: A randomized clinical trial’, JAMA Network Open, 3(2), e200075.

Oswald, A.J., Proto, E. and Sgroi, D. (2015) ‘Happiness and productivity’, Journal of Labor Economics, 33(4), pp. 789–822.

Reme, S.E., Grasdal, A.L., Løvvik, C., Lie, S.A. and Øverland, S. (2015) ‘Work-focused cognitive–behavioural therapy and individual job support to increase work participation in common mental disorders: A randomised controlled multicentre trial’, Occupational and Environmental Medicine, 72(10), pp. 745–752.

Schultz, T.W. (1961) ‘Investment in human capital’, American Economic Review, 51(1), pp. 1–17.