Information about the results of actions that support learning, adaptation, and improvement.
Feedback is the information that returns after an action. It tells individuals, teams, and organisations about the effects of what they have done, whether those effects were intended or unintended, visible or delayed, helpful or harmful.
In simple systems, feedback may be direct and immediate. A process change reduces defects. A customer complaint reveals a service failure. A trial produces a measurable result.
In complex systems, feedback is often slower, messier, and harder to interpret. Consequences may appear later, emerge indirectly, affect different stakeholders in unintended ways, or become visible only through patterns over time.
This makes feedback central to adaptive capacity.
Without feedback, organisations cannot learn from consequences. They may continue to act on assumptions, reports, plans, or beliefs that no longer reflect reality.
Why It Matters
Feedback connects action to learning.
An organisation may act, decide, intervene, or implement, but unless it observes and interprets the consequences of those actions, learning remains incomplete.
Feedback helps organisations ask:
- What actually happened?
- What changed?
- Who was affected?
- What did we misunderstand?
- What should we stop, continue, or change?
- What new understanding has emerged?
Feedback is therefore not just a measurement. It is part of sensemaking.
Feedback and Consequence
Feedback is the way consequences become visible.
Consequences occur whether organisations notice them or not. Feedback is the information that enables those consequences to be recognised, interpreted, and used for learning.
This is why Gemba is important. Feedback obtained solely through reports, dashboards, or remote indicators can be delayed, filtered, or distorted. Direct involvement with the actual situation helps people understand the consequences more clearly.
Feedback is most valuable when it leads to reflection, changed understanding, and improved action.
Feedback Loops
A feedback loop exists when the results of an action influence future action.
A weak feedback loop reports what happened but changes little.
A strong feedback loop changes behaviour, assumptions, processes, relationships, or decisions.
In adaptive organisations, feedback loops support:
- learning from experience
- correcting errors
- testing assumptions
- improving practice
- detecting weak signals
- adapting to changing conditions
Feedback loops become especially powerful when they support double-loop learning: not only asking whether actions worked, but also whether the assumptions behind those actions remain valid.
Red Flags
Feedback is weakened when:
- unwelcome news is hidden.
- metrics replace reality.
- dashboards become substitutes for Gemba.
- delayed consequences are ignored.
- leaders only listen to confirming evidence.
- people fear reporting problems.
- action continues even when reality says otherwise.
When feedback is blocked, distorted, or ignored, organisations become more confident yet less connected to reality.
In Summary
Feedback provides information from actual outcomes. It links actions to their consequences and consequences to learning. Without effective feedback, organisations might continue performing the same actions without learning and then adjusting. Strong feedback loops enable organisations to observe changes, challenge assumptions, refine their actions, and enhance their ability to adapt.
Related Glossary Items
- Consequence
- Consequential Intervention
- Gemba
- Double-Loop Learning
- Sensemaking
- Adaptive Capacity
- Hansei
- Line of Sight
Related Source Notes
- Chris Argyris Perspective: Defensive Routines & Double-Loop Learning
- Source Note β Consequential Intervention
- Source Note β Line of Sight
- Gemba Tests Everythingπ§ͺ
- Source Note β Learning Loops & Sensemaking
- Source Note β Action & Learning Loops
- Subject Area β Feedback
- Mindsets β The Hidden Driver of Organisational Behaviour
- Subject Area β The Learning Environment
- Subject Area β Gemba: Where Learning Becomes Real
- SECI as Learning Throughput β The Engine of Adaptive Capacity