Organisations do not choose interventions at random.
They reach for approaches that are familiar, available and within their authority to commission.
A performance concern leads to tighter governance. A collaboration problem leads to team development. Inconsistent leadership leads to a development programme. A difficult transformation leads to restructuring or a new programme office.
Each response may be reasonable. It may also be directed at the wrong level.
A sensible intervention applied at the wrong level can absorb considerable time and effort while leaving the conditions sustaining the problem largely untouched.
Before deciding what to do, it is worth asking a more fundamental question:
Where is this outcome being produced?
A reasonable response can still miss the problem
Problems are often defined by where they become visible.
If a decision is delayed, attention turns to the individual or group responsible for making it. If teams are in conflict, the problem is located in their relationships. If performance varies, the response focuses on capability or behaviour.
But the place where a problem appears is not necessarily the place from which it originates.
Consider an organisation in which important decisions are consistently slow.
The immediate explanation might be that leaders need to become more decisive. A leadership intervention follows.
Yet the delay may also reflect unclear strategic priorities. Leaders cannot make a clear choice because the organisation has not agreed what matters most.
Decision rights may be poorly defined, with several groups believing they must be consulted or give approval.
Relationships may lack sufficient trust for disagreement to be expressed openly. Concerns are discussed outside the meeting rather than resolved within it.
Previous decisions may have been challenged or reversed, teaching people that caution is safer than commitment.
The organisation may also be dependent on partners, regulators or other parts of a wider system whose timescales it cannot control.
What first appeared to be an individual leadership problem is being produced through a combination of strategic, structural, relational, behavioural and systemic conditions.
Leadership development may still help. On its own, it is unlikely to resolve the delay.
Five connected levels of intervention
HAWS uses five connected levels to understand where support might have the greatest impact.
- 01
Structural
Governance, accountability, roles, operating models and decision rights.
Structural questions include who has authority, how work is organised and whether formal arrangements help or hinder delivery.
- 02
Relational
Collaboration, trust, team dynamics and cross-boundary working.
Relational conditions influence what can be discussed, how disagreement is handled and whether people can work through uncertainty together.
- 03
Behavioural
Leadership behaviours, culture, norms and psychological safety.
Behaviour is not simply a matter of personal style. It develops in response to the surrounding environment.
- 04
Strategic
Alignment, priorities, clarity of purpose and organisational direction.
When strategic direction is unclear or contradictory, other parts of the organisation are left to interpret priorities for themselves.
- 05
Systemic
Interdependencies, unintended consequences and wider system dynamics.
Systemic problems often emerge through interaction. No single organisation or individual controls the outcome, even though many contribute to producing it.
The levels are not separate boxes
The five levels are not a hierarchy. Systemic does not sit above the others as the most sophisticated or important explanation.
All five are aspects of a system.
Here, “systemic” refers specifically to patterns created through wider interdependencies, particularly those extending across organisational boundaries. Structural, relational, behavioural and strategic conditions remain systemic in the broader sense because they interact and influence one another.
The purpose of distinguishing the levels is not to divide a complex situation into neat categories. It is to prevent one familiar explanation from dominating too quickly.
A structural change will affect relationships and behaviour. Strategic ambiguity may produce structural duplication. Low trust can encourage additional governance, which may further slow decisions and reinforce mistrust.
Why organisations intervene at the wrong level
Several pressures make this more likely.
Some interventions are easier to commission than others. A workshop, restructure or leadership programme is visible and definable. Changing the conditions that repeatedly produce a problem may require a more difficult conversation about power, priorities or accountability.
The way a problem is described also narrows the available response. Once something has been labelled a culture problem, a leadership failure or a governance issue, alternative explanations receive less attention.
Sponsors may focus on the part of the situation that falls within their remit. The problem is then shaped around the authority they already hold rather than the conditions that need to change.
Responsibility can also be pushed downwards. Organisational problems become translated into expectations that individual employees should become more resilient, collaborative or accountable.
This does not mean that behaviour is irrelevant. It means behaviour should be understood in context.
People often respond rationally to the environments in which they work. If the environment remains unchanged, asking them to behave differently may have little lasting effect.
Finding a useful point of intervention
Complex situations rarely offer a single root cause. Trying to change everything at once would be neither practical nor desirable.
The task is to identify a useful point of intervention: somewhere a proportionate change could influence the wider pattern.
That begins by describing the recurring outcome without building an assumed cause into the description.
Instead of saying, “Leaders are not taking accountability,” ask, “What is happening to decisions that need clear ownership?”
Instead of saying, “Teams are working in silos,” ask, “What makes working across these boundaries difficult?”
The inquiry can then move across the five levels:
- What strategic choices or tensions shape the situation?
- What structures, processes or accountabilities reinforce it?
- How do relationships affect what is possible?
- What behaviours have become normal, and what might they be responding to?
- What wider interdependencies or feedback processes help sustain the pattern?
The answers will not all carry equal weight. Some conditions will be consequences rather than causes. Some will be difficult to influence. Others may offer a practical point at which a relatively small change could alter the wider pattern.
The first intervention should therefore be treated as an informed choice, not a final answer. Its effects need to be observed. Assumptions need to be tested. The organisation should remain alert to consequences appearing elsewhere.
Proportionate, not exhaustive
Not every organisational problem requires an extensive multi-level assessment.
A broken process may need to be fixed. A role may be genuinely unclear. Two people may simply need a better conversation.
The five levels become particularly valuable when a problem:
- Keeps returning despite repeated action
- Appears in several parts of the organisation
- Crosses team or organisational boundaries
- Produces different explanations from different groups
- Has resisted interventions that appeared sensible
- Improves in one place but creates consequences elsewhere
These are signs that the outcome may be produced through interactions rather than one isolated cause.
The aim is not to make every problem appear more complicated. It is to avoid making a complex problem artificially simple.
Choosing the right intervention is not primarily about selecting the best method.
It is about understanding the conditions producing the outcome, identifying where change could alter the pattern and acting with enough curiosity to learn what happens next.
Do not begin with the intervention you know.
Begin with the outcome you need to understand.