Five Focusing Steps

Definition

The Five Focusing Steps (5FS) are TOC's core method for improving any system: (1) identify the system's constraint, (2) decide how to exploit the constraint (get the most from it without new investment), (3) subordinate everything else to that decision, (4) elevate the constraint (invest to break it) if exploitation isn't enough, and (5) if the constraint moves in Step 4, go back to Step 1 — and don't let inertia become the new constraint. Goldratt condenses all of TOC into one word: "focus," defined narrowly as "do what should be done AND don't do what should not be done."

In the Book

Chapter 1 ("Introduction to TOC—My Perspective," written by Goldratt himself) frames the whole handbook around this idea: because organizations have heavy interdependency and variability, Pareto's 80-20 rule becomes extreme — "0.1 percent of the elements dictate 99.9 percent of the result." Goldratt states the core error plainly: "the number one reason for not doing what should be done is doing what should not be done," because cost accounting rewards production everywhere, even on non-bottlenecks, past the point where more output helps. Chapter 8 walks through the 5FS applied to production operations in detail: Step 1 identifies the bottleneck resource (the weakest link relative to market demand); Step 2 exploits it by eliminating wasted capacity (bad setups, working on the wrong orders) and protecting it with a buffer; Step 3 subordinates every non-constraint resource's decisions to support the constraint's schedule; Steps 4-5 elevate capacity and then re-identify the constraint, since it may move once elevated — the "don't let inertia become the constraint" warning appears repeatedly across chapters (Ch. 7, 8, 9) because organizations often keep managing around a bottleneck long after it has moved.

Why It Matters

The 5FS reframes improvement as a repeating discovery process rather than a one-time optimization: since one element almost always governs a whole system's output, effort spent improving non-constraints is largely wasted or even harmful (an hour lost on the bottleneck is an hour lost system-wide; an hour gained on a non-bottleneck is a mirage). This generalizes far beyond factories — to any system with dependent steps and finite capacity, whenever "more effort everywhere" is competing with "find and fix the one thing actually limiting output."