Organisations rarely act with complete certainty.
Evidence may point towards a recurring pattern. Different perspectives may have been brought together. Leaders may have developed a credible explanation of what is sustaining the situation.
Even then, something will remain uncertain.
The explanation is necessarily partial. Some connections will be difficult to see. People and organisations may respond to change in unexpected ways. An intervention intended to improve one part of the situation may create pressure somewhere else.
This does not mean that action should be postponed until more is known.
It means the first intervention should do two things: seek to improve the situation and help the organisation test the explanation on which it is acting.
Every intervention contains an explanation
An intervention is never simply an activity.
It expresses a belief about how change will happen.
A new approval process assumes that stronger oversight will improve the quality or consistency of decisions. A leadership programme assumes that different capability or behaviour will affect organisational outcomes. A restructure assumes that changing formal boundaries, roles or accountabilities will alter how work is done.
Those assumptions may be reasonable. They are still assumptions.
Often, they remain implicit. The organisation defines what will be introduced, who will deliver it and when it should be completed. Less attention is given to why this particular action is expected to influence the outcome.
That makes it harder to learn from what happens next.
If the intended improvement does not follow, the intervention may be described as poorly implemented. People may be said to have resisted the change. More communication, oversight or time may be added.
Any of these explanations might be valid.
But another possibility is that the intervention was based on an incomplete account of what was sustaining the situation.
Making that account visible allows it to be examined rather than protected by the activity built around it.
Acting in a system
In a complicated task, it may be possible to specify the necessary steps and predict the result with reasonable confidence.
Complex organisational situations behave differently.
Outcomes are produced through interactions between people, priorities, structures, relationships, information, incentives and events beyond the organisation's control. The same intervention can produce different effects in different settings because it enters a different pattern of conditions.
The organisation also responds to the intervention.
People interpret what the change means. They adjust their behaviour. Existing pressures find new routes. Formal processes interact with informal practice. An improvement in one place may be accompanied by delay, additional work or increased risk elsewhere.
This is why an intervention cannot be understood only through whether its planned activities were delivered.
Its effects need to be observed across the system in which it was introduced.
Did the recurring outcome change? What happened to the conditions believed to be sustaining it? How did different parts of the organisation respond? Where did the consequences appear? What became visible that was previously hidden?
The response of the system is further evidence.
A proportionate first move
Treating the first intervention as a test does not mean experimenting carelessly with important services, people or outcomes.
The intervention still needs a sound evidence base, appropriate safeguards and clear accountability. Some situations require urgent or prescribed action. Questions of safety, legality or serious failure cannot always wait for a small-scale test.
But many organisational difficulties do allow a proportionate first move.
Consider an organisation in which decisions involving several teams are repeatedly delayed.
Different accounts have been explored. The people involved broadly agree that the delay is not simply a lack of decisiveness. Important decisions move through several groups, authority is unclear and people are uncertain whether a collective decision will be supported elsewhere.
The organisation could respond with a broad governance review or a leadership programme.
Alternatively, it might begin by changing the arrangements around one recurring category of decision. It could clarify who has authority, identify whose advice is required, make the relevant information available earlier and record what happens after the decision is made.
This is a real intervention. It is intended to improve the work.
It also tests the explanation.
If decisions become quicker and remain supported, that strengthens the view that authority and information were significant conditions. If decisions are still delayed, the organisation can look more closely at where and why. If speed improves but important perspectives are excluded, the apparent improvement has exposed a different risk. If the decision is made but later reversed, the source of the difficulty may sit beyond the formal process that was changed.
Each response helps refine the organisation's understanding.
Make the reasoning visible
A useful test does not require a complicated methodology.
It does require the reasoning behind the intervention to be explicit enough to examine.
Five prompts can help.
1. What do we believe is sustaining the outcome?
Describe the current working explanation.
This should connect the evidence rather than simply restate the problem. Saying that decisions are slow describes an outcome. Saying that decisions are repeatedly delayed because authority is distributed across several groups, relevant information arrives late and previous collective decisions have been reopened offers an explanation that can inform action.
The explanation should distinguish what is known, what is inferred and what remains uncertain.
2. What are we changing?
Be specific about the condition the intervention is intended to influence.
Introducing a workshop, a new role or a revised process describes an activity. The more important question is what that activity is expected to change: the flow of information, the clarity of authority, the quality of disagreement, the balance between competing priorities or something else.
This keeps attention on the connection between action and outcome.
3. What would we expect to notice?
Identify the signals that would make the explanation more credible.
These may include performance measures, but not only performance measures. Changes may first appear in how people describe the situation, where work accumulates, which decisions are escalated, how often exceptions are needed or whether the people affected experience the service differently.
Some effects will take time. Others may be visible quickly. Being clear about both helps avoid declaring success too early or dismissing an intervention before a reasonable effect could emerge.
4. Where else should we look?
An improvement in one place does not necessarily mean the wider situation has improved.
A faster decision may create more rework downstream. A reduction in one team's workload may transfer demand to another. Stronger control may improve consistency while reducing the ability to respond to legitimate variation.
Looking across boundaries helps the organisation notice displacement, unintended consequences and changes experienced by people who are not close to the intervention.
5. What would cause us to reconsider?
The purpose of a test is not to prove that the original explanation was correct.
It is to learn whether it is useful.
Leaders should decide in advance what evidence would prompt them to revise the explanation, adapt the intervention or stop. Without this discipline, organisations can continue investing in an approach while interpreting every difficulty as a reason to intensify it.
Reconsidering is not an admission that the work lacked rigour. It is part of acting responsibly in a situation that could not be fully known in advance.
More than success or failure
Organisations often review interventions through a binary question: did it work?
The answer matters, but it may conceal as much as it reveals.
An intervention can produce several effects at once. Some parts may improve while others remain unchanged. The intended outcome may occur for reasons different from those originally assumed. A beneficial short-term response may create a longer-term dependency. An intervention may fail to improve the headline measure while revealing a constraint that had not previously been understood.
A more useful review asks:
- What changed?
- What stayed the same?
- What moved elsewhere?
- What did people do in response?
- What have we learned about the original explanation?
This does not remove the need to judge results.
It improves the quality of that judgement by examining the pattern of effects rather than one measure in isolation.
Learning without losing direction
There is a risk that the language of testing and adaptation can make change sound provisional or endlessly open to revision.
People need clarity. Organisations cannot continually alter course whenever an intervention encounters difficulty. Sustained change often requires persistence through uncertainty, discomfort and an initial decline in performance.
The discipline is not constant change.
It is maintaining a clear connection between the outcome, the explanation, the intervention and the evidence emerging from its effects.
That makes it possible to distinguish between an intervention that needs time, one that needs better implementation and one that is acting on the wrong conditions.
It also creates a more honest basis for accountability. Leaders can be clear about the judgement they have made, the evidence supporting it, the uncertainty that remains and how they will respond as more becomes known.
Intervention develops capability
When the reasoning behind an intervention is visible, learning is not confined to the people who designed it.
Those doing the work can compare the expected effects with what they are experiencing. Different perspectives can reveal consequences that formal reporting misses. Leaders can see where their understanding was strong and where it was limited.
Over time, the organisation becomes better able to notice patterns, test assumptions and adapt deliberately.
This is part of developing organisational capability.
The immediate intervention may address one recurring difficulty. The deeper value lies in strengthening the organisation's ability to understand and respond to the next one.
Acting with curiosity
Complexity does not remove the need for action.
It changes the way action should be understood.
An intervention is not the final stage of a process in which uncertainty has already been resolved. It is an action within a living system. It changes conditions, prompts responses and produces new evidence.
The first intervention should therefore be deliberate, proportionate and open to learning.
Before acting, make the reasoning clear:
- We believe this outcome is being sustained by...
- We will intervene by changing...
- If our explanation is useful, we expect to notice...
- We will also watch for consequences appearing...
- We will reconsider if...
The purpose is not to predict everything correctly.
It is to make the thinking behind the intervention visible enough to test.
Act on the best explanation available.
Then pay attention to what the system does next.