Organisations rarely fail because they lack experience. They fail because they fail to retain it.
Every major initiative surfaces lessons: decision rights were unclear, incentives were misaligned, portfolio load was excessive, escalations were softened too long. Post-implementation reviews capture these insights. Reports are written. Recommendations are made. And then the system moves on.
Why learning evaporates
Learning is often treated as reflection rather than redesign. The organisation acknowledges what went wrong. It may even agree. But if structural elements remain unchanged — authority design, incentive alignment, prioritisation discipline — the same patterns re-emerge in the next initiative. Behaviour follows structure more reliably than it follows memory.1
Institutional memory that does not alter architecture becomes narrative rather than correction.
Experience creates confidence. Leaders believe they know what to watch for. Practitioners recognise early signals. Sponsors anticipate resistance. But if the operating model, governance boundaries, and incentive systems remain unchanged, prior awareness does not prevent recurrence. It simply makes the instability feel familiar. That familiarity can create complacency.2
Recognition without redesign does not alter outcome.
How repetition becomes embedded
Over time, recurring instability becomes normalised. Escalation cycles are expected, adoption dips are anticipated, and mid-stream realignment becomes routine. The organisation builds muscle memory for correction rather than prevention. That muscle memory is adaptive — but it consumes energy.3
Repeated instability reduces confidence quietly. Employees begin to assume that initiatives will wobble before they stabilise. That assumption shapes engagement.
When learning is acknowledged but not embedded structurally, the same authority ambiguities recur, the same incentive contradictions surface, and the same misdiagnoses appear. Each cycle reinforces the perception that change is inherently unstable. In reality, instability was structurally reproducible.
Without embedding lessons into governance and operating design, learning becomes anecdotal. And anecdote does not alter architecture.
What embedding actually requires
Embedding learning means asking: what structural conditions produced this instability? Have we altered decision rights accordingly? Have we adjusted portfolio discipline? Have we realigned performance metrics?4
Institutional memory must manifest in structure. Otherwise, the system resets. Integration becomes episodic rather than cumulative.
Organisations often believe experience accumulates automatically. It does not. Only structural redesign accumulates. Without embedding architectural correction, change efforts repeat familiar instability patterns.
This is one way of understanding how integration, governance design, and practitioner stance determine whether learning stabilises the system or simply documents its repetition.
-
Argyris, C., & Schön, D. A. (1978). Organizational Learning: A Theory of Action Perspective. Addison-Wesley. Argyris and Schön distinguish single-loop learning — error correction within an existing frame of governing values — from double-loop learning, which requires questioning the governing values themselves. Post-implementation reviews that acknowledge what went wrong without altering structural design are single-loop: they correct the error but leave the conditions that produced it intact. The same patterns recur not because the organisation failed to learn but because single-loop learning was applied to a double-loop problem. The structural conditions remain the governing attractor; behaviour follows them in the next cycle exactly as in the previous one. ↩︎
-
Weick, K. E., & Sutcliffe, K. M. (2007). Managing the Unexpected: Resilient Performance in an Age of Uncertainty (2nd ed.). Jossey-Bass. Weick and Sutcliffe’s analysis of normalisation of deviation describes precisely how familiarity with anomalies becomes a liability rather than an asset: organisations that have experienced a pattern before assign it to the “expected” category and reduce their vigilance accordingly. Prior awareness of instability patterns does not sharpen the response; it attenuates it. The instability feels familiar, and familiarity — the sense that “we’ve seen this before and managed it” — creates exactly the complacency that prevents the structural recalibration that would actually interrupt the pattern. ↩︎
-
Sterman, J. D. (2000). Business Dynamics: Systems Thinking and Modeling for a Complex World. McGraw-Hill. Sterman’s systems analysis of policy resistance explains why organisations accumulate capability in symptomatic correction rather than structural redesign. When the structural attractor remains intact, symptomatic interventions must be repeated each time the attractor reasserts itself. Each repetition builds competence in the symptomatic response — the organisation becomes faster and smoother at correction cycles — while the structural condition continues to generate the need for correction. “Muscle memory for correction rather than prevention” is the inevitable product of applying symptomatic solutions to problems with unresolved structural drivers. ↩︎
-
Schein, E. H. (1999). Process Consultation Revisited: Building the Helping Relationship. Addison-Wesley. Schein argues that the most important diagnostic question after a difficult episode is not “what happened?” but “what structural conditions allowed this to happen?” — because it is structural conditions, not events, that recur. Embedding learning means making structural conditions the object of inquiry: decision rights, prioritisation discipline, performance metrics. Until these are examined and adjusted, post-implementation learning remains what Schein calls “content knowledge” — knowledge about what happened — rather than “process knowledge” — knowledge about how the structural conditions that produced it can be altered. ↩︎