Culture is too broad a target when the organisation needs a specific action to change.
"Improve the culture."
"Increase awareness."
"Encourage sustainable behaviour."
"Build an innovation mindset."
These phrases are common in transformation programs because they point toward something real: outcomes often depend on what people repeatedly do. But they are poor intervention specifications.
A culture cannot click an approval button, escalate a safety concern, choose a lower-impact travel option, use a new planning routine or challenge a poor-quality forecast. People do those things in specific contexts, under specific pressures, with specific beliefs about effort, consequences and what others expect.
If leadership wants behaviour to change, the unit of design should be the behaviour.
The Strategic Context
Juvan and Dolnicar's 2017 study of environmentally sustainable tourist behaviour challenged a convenient assumption: that different pro-environmental behaviours share the same underlying drivers.
They examined seven specific behaviours and found that the significant predictors differed across behaviours. Self-identity emerged as a driver across all seven, and involvement in organised environmentalism was associated with all seven. But other factors such as guilt, personal norms, environmental knowledge, education and place of residence mattered differently depending on the action.
The study's conclusion is strategically useful well beyond tourism: interventions should be behaviour-specific rather than built around a generic theory of the person.
The researchers also pointed to an alternative: change the choice architecture so that the desired behaviour becomes easier or more likely without first having to change deep beliefs.
What Leaders Commonly Misread
The first error is assuming knowledge causes action.
Training can be necessary. It is not sufficient.
An employee may know the correct procurement process and still bypass it because the approved route is slow. A manager may understand the value of lessons learned and still skip the review because the next project has already started. A traveller may understand environmental impacts and still choose the faster or cheaper option.
The second error is treating all behaviours in the same program as if they have the same barrier.
"Adopt the new system" may contain dozens of distinct behaviours: entering data correctly, checking a dashboard, escalating an exception, retiring a spreadsheet and using a new approval path. Each can fail for a different reason.
The third error is trying to change values when changing the environment would be easier.
If the desired option is difficult, hidden, slow or risky, a communications campaign may have little effect. Sometimes the rational intervention is to redesign the workflow.
Reframing the Issue
Behaviour change should be treated as an engineering problem with human variables.
Define:
actor → behaviour → context → barrier → driver → intervention → evidence
For example:
Actor: project manager
Behaviour: escalates forecast variance before the monthly review
Context: high schedule pressure and concern about appearing weak
Barrier: perceived social penalty for early escalation
Driver: leadership response and peer norms
Intervention: reward early visibility; require variance commentary at defined thresholds; publicly distinguish bad news from poor management
Evidence: average time between variance emergence and escalation
This is much more actionable than "create a culture of transparency."
Different Behaviours Need Different Levers
The tourism research illustrates this directly.
In its data, reducing the number of vacations and choosing more sustainable transport shared several drivers, including self-identity, social norms, guilt and organised environmentalism. Other behaviours had different patterns. Environmental knowledge was specifically associated with choosing providers that operated sustainably, while formal education was associated with choosing certified providers. Personal norms were not uniformly predictive across all behaviours.
The lesson is not to copy tourism interventions into business.
It is to stop assuming that one leadership message can modify every action in a transformation.
A cybersecurity program may need different interventions for password behaviour, phishing reporting, privileged access and software installation.
A safety program may need different interventions for PPE use, near-miss reporting, pre-start risk assessment and stopping work.
An AI adoption program may need different interventions for trying a tool, verifying outputs, protecting confidential data and escalating high-consequence uncertainty.
The program theme can be broad. The behaviour design cannot.
Identity Matters, but It Is Not a Shortcut
Self-identity was the most consistently associated cognition in the tourism study. In organisational settings, identity can also be powerful.
Engineers may act in ways consistent with being rigorous problem solvers. Nurses may strongly identify with patient safety. Project professionals may identify with accountability for delivery. Craftspeople may identify with quality.
Leaders can use identity by connecting desired behaviour to a legitimate professional standard: "This is what good engineering judgement looks like here."
But identity appeals become manipulative if the system makes the behaviour unreasonable.
Telling employees that "owners take accountability" while denying them decision rights will not create ownership. Asking people to "speak up" while punishing inconvenient information will not create psychological safety.
Identity and system design must agree.
Norms Can Be Designed Through Leadership Behaviour
People infer what is expected not only from policy but from observation.
What does the executive team praise?
What happens when a project leader reveals a problem early?
Which behaviours receive status?
What shortcuts are tolerated from high performers?
If the formal message says one thing and the lived norm says another, the lived norm usually wins.
This is why behavioural change starts with leader behaviour. A governance system is itself a choice architecture: it signals which information must be surfaced, who can decide, and what happens after a decision.
Choice Architecture Can Outperform Persuasion
A choice architecture changes the environment in which decisions are made.
Examples include:
- making the safer or lower-impact option the default;
- reducing the number of steps required for the desired action;
- placing information at the point of decision rather than in annual training;
- requiring active justification for a high-risk exception;
- automatically escalating when defined thresholds are crossed;
- pre-populating high-quality options;
- making consequences visible immediately.
This can be more reliable than hoping people remember a message delivered months earlier.
The ethical test is straightforward: the architecture should support legitimate organisational outcomes without concealing material consequences or eliminating reasonable choice where choice matters.
Decision Framework
Before launching a behavioural intervention, leadership should answer seven questions.
1. What exact behaviour needs to change?
Use an observable verb.
Not "be collaborative", but "invite the affected operations lead before approving the design change."
2. Who performs it?
Avoid "the organisation". Name the role or actor group.
3. In what context does the behaviour occur?
Time pressure, hierarchy, customer interaction and digital interface conditions can all change behaviour.
4. What currently competes with the desired action?
People may be responding rationally to a conflicting KPI, workload, incentive or process delay.
5. Which driver is plausible?
Knowledge, identity, norm, effort, perceived control, reward, fear, habit or convenience?
Use evidence rather than preference.
6. Can the environment be redesigned?
Ask whether the desired action can become easier, more visible or the default.
7. What behavioural measure will prove change?
Measure the action itself before relying on attitude surveys.
From Strategy to Execution
Immediate action: choose one important behavioural outcome in a current transformation and rewrite it in actor-behaviour-context form. Remove vague cultural wording.
Medium-term capability building: build behavioural diagnostics into change management. Combine qualitative observation, process data and small experiments. Do not assume the same intervention will work across sites, roles or behaviours.
Long-term strategic positioning: develop an operating model in which governance, incentives, digital interfaces and leadership signals reinforce the behaviours the strategy requires. Culture then becomes an emergent property of repeated system-consistent behaviour rather than a communications campaign.
Related article: Slow-Onset Risks Fail When Organisations Treat Them as Events
Signals to Monitor
Watch for:
- high awareness scores with low behavioural compliance;
- repeated training on the same issue;
- leaders blaming "resistance" without identifying a concrete barrier;
- desired behaviour requiring significantly more effort than the alternative;
- KPIs that reward the opposite action;
- stated values contradicted by executive behaviour;
- interventions aimed at attitudes when workflow redesign is available;
- one generic change plan used across materially different roles.
A particularly important signal is the intention-action gap. If people consistently say the right thing and do something else, the problem may sit in the system rather than in their values.
Questions for the Leadership Team
- Which three specific behaviours are critical to the success of our current strategy?
- What evidence tells us why each behaviour is or is not occurring?
- Are we trying to change beliefs when changing the process would be more reliable?
- What do our incentives and leadership reactions teach people to do in practice?
- Which desired behaviour is currently harder than the behaviour we are trying to replace?
- How will we measure actual behavioural change rather than awareness or sentiment?
Closing Perspective
People do not resist strategies in the abstract. They encounter moments of action.
The strategic discipline is to identify those moments and design them deliberately.
When leaders move from "change the culture" to "change this behaviour, in this context, for these reasons", intervention quality improves. The organisation can test assumptions, remove friction and learn what actually changes performance.
Behaviour is specific. Change design should be too.
Source basis: This article is an original ERANORTH synthesis principally informed by Emil Juvan and Sara Dolnicar (2017), Drivers of pro-environmental tourist behaviours are not universal, Journal of Cleaner Production, 166, 879–890.
About EraNorth Insights
EraNorth Insights publishes practical analysis on strategy, projects, operations, transformation and decision intelligence for professional and organisational use. About EraNorth.