Constraint Migration
Constraint Migration is the phenomenon that gives TOC's fifth Focusing Step its necessity: every successful improvement eventually moves the constraint. The factory whose bottleneck was a specific machine, once elevated, will find its bottleneck elsewhere — another machine, a quality process, the market itself. The knowledge organization whose constraint was coordination, once the constraint is shattered, finds a new constraint: judgment, or domain knowledge, or market absorption rate. The migration is a feature of constraint theory, not a bug, and organizations that do not track it end up managing a constraint that no longer binds while the actual constraint sits unmanaged.
Migration as Designed Obsolescence — Contrarian ^ Opus
There is a parallel reading that begins not with constraint theory's elegance but with its economics: constraint migration is not a discovered feature of systems but a manufactured feature of consulting practices. The perpetual discipline Goldratt describes — the injunction to return to Step One, the necessity of continuous re-identification — is precisely what transforms TOC from a one-time framework into a permanent revenue stream. Every successful elevation creates the conditions for the next engagement.
This reading does not dispute that bottlenecks move when capacity expands. It disputes the framing that this movement is neutral, discovered, or inherent to production systems rather than to the advisory relationship. The AI transition provides the clearest contemporary example: organizations that internalized TOC principles in the coordination era now face a constraint migration that cannot be managed with existing frameworks — requiring new tools, new training, new external expertise. The migration is real. But the insistence that it requires perpetual external guidance rather than building organizational capacity to navigate migrations autonomously is a claim about business models, not systems theory. The five steps could be taught as a transferable discipline — one that organizations learn to apply independently across migrations. That they are instead positioned as requiring continuous expert intervention reveals the economic substrate beneath the conceptual framework.
In the AI Story
The mechanics of constraint migration follow directly from Goldratt's framework. When the constraint is elevated (Step Four), its capacity expands. Other resources, previously subordinated to it, now become tight. Eventually one of them becomes the new constraint. Sometimes the migration is immediate and dramatic — as in the AI transition, where the coordination constraint was not gradually elevated but abruptly shattered by the language interface, producing an immediate shift to judgment as the new constraint. Sometimes it is gradual — as in manufacturing improvement programs where incremental machine upgrades slowly shift the bottleneck across the production line.
The management implication is that TOC is a perpetual discipline, not a one-time project. An organization that implements the Five Focusing Steps, identifies its constraint, exploits it, subordinates non-constraints, and elevates capacity — and then considers the project complete — will find within months or years that its new practices have become inertia because the constraint has migrated. The fifth step's injunction to 'return to Step One' is not methodological decoration; it is the core discipline of constraint management. The question 'where is the constraint now?' must be asked continuously.
The AI transition provides the clearest contemporary example of abrupt constraint migration. For fifty years, software development systems were bounded by the coordination constraint — the translation cost between minds documented by Brooks and explicated by Goldratt's framework. All management practices — Agile, Scrum, sprint planning, velocity tracking — addressed this constraint. In December 2025, the language interface shattered it. The constraint migrated overnight to judgment. Every practice designed for the coordination era became inertia the moment the migration completed. But most organizations did not notice the migration, and continue to manage the old constraint.
Future migrations will follow. If judgment is elevated — through mentorship, structured feedback, the slow development of taste — the constraint may migrate to domain knowledge: understanding users, markets, strategic context. If domain knowledge is elevated, the constraint may migrate to market absorption: the rate at which users can integrate new capability. If market absorption is elevated, the constraint may migrate to regulatory capacity, or to the pace of legal and institutional adaptation. The sequence is not predictable in detail but is predictable in structure: each elevation moves the constraint somewhere, and the organization that manages the old constraint while the new one emerges is always wrong.
Origin
Constraint migration is implicit in Goldratt's original formulation of the Five Focusing Steps (1984) but received explicit attention in later works, particularly It's Not Luck (1994), where the Thinking Processes were developed in part to handle the ongoing re-identification work that migration demands.
Key Ideas
Every elevation moves the constraint. Successful improvement of the bottleneck shifts the bottleneck elsewhere — a feature of constraint theory, not an anomaly.
Migration can be abrupt or gradual. Technology-driven migrations (like the AI transition) happen fast; incremental-improvement migrations happen slowly; both require continuous re-identification.
The fifth Focusing Step is not methodological decoration. The injunction to return to Step One is the core discipline that prevents management practices from ossifying into inertia.
The AI transition is abrupt constraint migration at civilization scale. The coordination constraint that bounded software for fifty years has migrated to judgment, and most organizations have not tracked the migration.
Future migrations will follow. Judgment, elevated, may migrate to domain knowledge; that, elevated, to market absorption; and so on. The sequence is structurally predictable even when specific details are not.
Appears in the Orange Pill Cycle
Migration Reality, Management Choice — Arbitrator ^ Opus
The core mechanism is 100% Goldratt: constraints do migrate when elevated, and organizations that fail to track the migration manage phantoms while actual bottlenecks go unaddressed. The AI transition proves this empirically — the coordination constraint is shattered, judgment now binds, and organizations running Agile ceremonies are managing a constraint that no longer exists. This part of the entry is simply descriptive of how systems behave.
The contrarian view becomes salient at a different question: whether migration necessitates perpetual external guidance or reveals a gap in organizational capability-building. Here the weight is roughly 60/40 toward the contrarian frame. The five steps can be taught as transferable discipline — many organizations do internalize constraint thinking and navigate migrations independently. But the perpetual-engagement framing Goldratt adopted (and that TOC consulting practices reinforce) is not purely pedagogical; it is also economic. The question is whether TOC is positioned as a framework to master or a service to purchase continuously.
The synthetic frame the topic benefits from: constraint migration is a real systems phenomenon, and the perpetual discipline is genuinely necessary, but 'perpetual discipline' can mean either building internal capacity to re-identify constraints across migrations or maintaining dependency on external expertise. The choice between these is organizational and political, not technical. The migration itself is inevitable. How organizations prepare for the next one — whether they build the muscles to navigate it or outsource the navigation — is a decision about capability development that the framework itself does not determine.
Further reading
- Eliyahu M. Goldratt, It's Not Luck (North River Press, 1994)
- Eliyahu M. Goldratt, Beyond the Goal (audio lectures, 2005)
- H. William Dettmer, Goldratt's Theory of Constraints (ASQ Quality Press, 1997)