Key takeaways
- Behaviour change is not an event — it is a process, and most organisations design their wellbeing programmes as though it were the former.
- The COM-B model (Capability, Opportunity, Motivation → Behaviour) provides a practical framework for diagnosing why wellbeing behaviours fail to embed.
- Environmental redesign — changing the default conditions of work — is consistently more effective than willpower-based individual interventions.
- The organisations that achieve lasting wellbeing cultures treat change as a multi-year infrastructure project, not an annual initiative.
There is a pattern that appears in organisations with unsettling regularity. A wellbeing strategy is commissioned. A consultant is engaged, or an internal working group convened. There are workshops, surveys, and a launch event. A set of initiatives is rolled out — perhaps a mental health awareness campaign, a resilience training programme, a new employee assistance service. Uptake is reasonable in the first few months. And then, quietly, things drift. The initiatives are still technically available. But they are no longer active, no longer championed, no longer embedded in the real texture of how the organisation operates.
Twelve months later, the organisation begins the process again, wondering why wellbeing never seems to truly take hold.
This is not a resources problem, or a communication problem, or even a leadership problem in the simple sense. It is a behaviour change problem. And until organisations understand how behaviour change actually works — at the individual, team, and organisational level — they will keep cycling through the same disappointing pattern.
Why Wellbeing Behaviours Don’t Stick: The Science
Behavioural science has made enormous strides in understanding why behaviour change succeeds or fails, across everything from health interventions to organisational development. The findings are consistent and, for organisations, instructive.
The dominant model that has emerged from this work is COM-B, developed by Professor Susan Michie and colleagues at University College London. COM-B proposes that any behaviour requires three components to be present simultaneously: Capability (the knowledge and skill to perform the behaviour), Opportunity (the physical and social environment that makes the behaviour possible), and Motivation (the reflective and automatic processes that drive the behaviour). When a behaviour fails to materialise or fails to sustain, it is because one or more of these components is absent or insufficient.
Applied to workplace wellbeing, this model has immediate practical implications. Most organisational wellbeing programmes target motivation — raising awareness, shifting attitudes, inspiring behaviour — while leaving capability and opportunity largely unchanged. The result is that even when motivation temporarily increases (as it typically does immediately after a wellbeing campaign), the behaviour fails to embed because the structural conditions for it have not changed.
Figure 1
Why Wellbeing Behaviours Fail to Embed
The COM-B model applied to workplace wellbeing programmes
C
Capability
Do people have the knowledge and skills to adopt wellbeing behaviours?
Often missing: Line managers lack skills to have wellbeing conversations
O
Opportunity
Does the work environment make wellbeing behaviours possible and normal?
Often missing: Workloads and norms make breaks, flexibility, and recovery feel impossible
M
Motivation
Do people want to engage with wellbeing, and does the culture reinforce it?
Most targeted: Campaigns and awareness events raise motivation — but only temporarily
Most wellbeing programmes target Motivation alone. Lasting change requires all three components to be addressed simultaneously.
A wellbeing campaign that encourages people to take breaks and protect their recovery time will fail if the organisation’s unspoken norm is that being busy signals commitment. It will fail if workloads are genuinely unmanageable. It will fail if line managers never model recovery behaviours themselves. The motivation may be present; the capability and the opportunity are not.
The Role of Defaults and Environment
One of the most robust findings in behavioural science is the power of defaults — the options and conditions that are in place when no active choice is made. In health behaviour research, default-setting interventions consistently outperform willpower-dependent ones. This finding has been replicated in diet, exercise, financial behaviour, and, increasingly, workplace contexts.
For organisations, this translates into a concrete design principle: make the wellbeing-supportive option the easiest, most natural choice. This means structuring meeting culture so that breaks are the default, not the exception. It means designing the physical and virtual workspace to reduce friction for recovery, movement, and social connection. It means building psychological safety into performance conversations so that admitting difficulty is culturally acceptable rather than career-threatening. None of this requires extraordinary effort from individuals. It requires intentional design of the organisational environment.
The research is particularly striking on the role of social norms. People calibrate their behaviour against what they perceive others around them to be doing. When senior leaders are visibly modelling wellbeing behaviours — taking holiday, leaving meetings on time, talking openly about managing energy — employees recalibrate their own sense of what is acceptable. When they don’t, no amount of wellbeing communications will convince people that the organisation’s stated values reflect its real ones.
”You cannot communicate your way to a wellbeing culture. Culture is what people observe others doing, not what they are told is important. If the behaviours at the top contradict the messaging, the messaging loses every time.“
— Carl Buik, Buik Health
From Launch to Embedded: The Change Architecture
The evidence on behaviour change timelines is sobering for organisations accustomed to annual programme cycles. Research on habit formation — most rigorously studied by Phillippa Lally and colleagues at University College London — suggests that new behaviours take an average of 66 days to become automatic, with a range of 18 to 254 days depending on the complexity of the behaviour and the individual. For organisations attempting to shift cultural norms and collective behaviour patterns, the timescale is considerably longer.
This means that any wellbeing strategy designed to deliver meaningful, sustainable change must operate on a multi-year horizon, with a detailed implementation architecture that accounts for the following stages:
Unfreezing — the phase in which existing patterns and assumptions are surfaced and examined. This is where diagnostic work is essential. Without a clear shared understanding of what is currently happening and why, there is no foundation for change. This phase often surfaces uncomfortable truths about workload, management quality, or psychological safety that the organisation has previously avoided examining.
Moving — the phase of active change, during which new behaviours, structures, and norms are introduced. This is where most wellbeing programmes concentrate all of their effort — but it is only one phase of a longer process. The most common failure mode is treating this phase as the entirety of the strategy rather than one part of it.
Refreezing — the phase in which new behaviours become embedded in the fabric of how the organisation operates: in its policies, its performance conversations, its management practices, its meeting culture, and its leadership behaviours. Without deliberate refreezing, the organisation will revert to its previous state when the active change phase ends. This is the phase that most organisations skip entirely.
What Embedded Wellbeing Actually Looks Like
Organisations that have genuinely embedded wellbeing — rather than launched it — share a set of recognisable characteristics. Wellbeing is not an HR programme running alongside the business; it is a set of design principles built into how the business operates. Workload sustainability is part of how teams plan, prioritise, and resource projects. Psychological safety is part of how performance conversations are conducted. Recovery and boundary-setting are modelled by leaders and reinforced by management norms, not just endorsed in a company values statement.
These organisations have also accepted that this work does not have an end point. Wellbeing cultures are not installed; they are continually maintained and adapted. As the business changes — new pressures, new leadership, new ways of working — the conditions for wellbeing must be actively managed. The organisations that do this best treat it not as a project with a completion date, but as an ongoing operational discipline.
The implication for organisations at the beginning of this journey is significant. The question is not ”what wellbeing programme should we run this year?“ It is ”how do we build the organisational architecture that makes wellbeing a consistent, self-reinforcing feature of how we work?“ Those are fundamentally different questions, and they lead to fundamentally different strategies.
The science of behaviour change does not make wellbeing easy. But it makes the path considerably clearer. And it makes it possible, with the right design, to build something that actually lasts.