The Goal

The model

Eliyahu Goldratt was an Israeli physicist who became interested in manufacturing in the 1970s, when he wrote software to schedule a friend’s factory. He found that the scheduling problem could not be solved by optimising every machine, and that a few machines determined the output of the whole plant. In 1984 he and Jeff Cox published The Goal, a novel about a plant manager named Alex Rogo whose factory has ninety days to improve or be closed. It sold millions of copies and became one of the few business novels taught in operations courses.

The model at its centre, later called the theory of constraints, rests on three ideas.

First, the goal. A business’s goal is to make money now and in the future. Every activity should be judged by its effect on that goal, and Goldratt proposes three measures to do so. Throughput is the rate at which the system generates money through sales. Inventory is all the money the system has invested in things it intends to sell. Operating expense is all the money the system spends turning inventory into throughput. The aim is to increase throughput while reducing inventory and operating expense, and of the three, throughput matters most.

Second, constraints. In any system, one or a few resources limit output. The output of the whole can never exceed the output of its constraint. An hour lost at the bottleneck is an hour lost for the entire system; an hour saved at a non-bottleneck is a mirage, because it only produces extra inventory waiting in front of the constraint.

Third, a process for improvement, which Goldratt later summarised as five focusing steps. Identify the constraint. Decide how to exploit it, getting the most from it, for instance by never letting it sit idle or work on defective parts. Subordinate everything else to that decision, so that other resources work at the pace of the constraint rather than at their own maximum. Elevate the constraint, adding capacity if it is still limiting. And when the constraint is broken, go back to the first step, because a new constraint will have appeared, and do not let inertia become the constraint.

Seeing with it

The novel’s best-known scene is a hike. Alex takes his son’s scout troop on a long walk and watches the line spread out. The slowest boy, Herbie, sets the pace for the whole group; the boys in front race ahead and the gaps grow. Alex’s solution is to put Herbie at the front, so that nobody can outrun him, and then to lighten Herbie’s pack by distributing its contents among the others. The troop moves together and faster. The scene captures nearly the whole theory.

Back at the plant, Alex discovers that his two bottleneck machines are idle during breaks, work on parts that are not needed, and process defective parts that are scrapped later. Fixing these problems increases throughput substantially without new investment. He also learns that the efficiency measures his company uses, which reward every machine and worker for being busy, are causing the problem by flooding the plant with inventory that the bottlenecks cannot process.

The lesson reaches well beyond factories. In software teams, a single reviewer, test environment or approval step often limits delivery. Making developers faster only lengthens the queue. The model also explains why local efficiency metrics can be harmful wherever work flows through several stages.

Where the model stops

The theory is clearest in systems with a stable, identifiable constraint, such as a production line. In complex organisations the constraint can shift frequently, or lie in policies, habits and markets rather than in a machine. Goldratt himself argued that most constraints are policy constraints, and developed tools for analysing them, which he called the thinking processes. These are harder to apply and less widely used.

Empirical studies of the theory of constraints have generally reported improvements in output, lead times and inventory in organisations that applied it, though many studies are case reports rather than controlled comparisons, a limitation that reviewers such as Victoria Mabin and Steven Balderstone noted in 2003.

Since it was published

Goldratt wrote several sequels and applied the theory to project management, in Critical Chain (1997), and to distribution and retail. He died in 2011. The Goal’s influence on software has been large. Gene Kim, Kevin Behr and George Spafford modelled The Phoenix Project (2013) explicitly on it, retelling the story in an IT department, and the DevOps movement drew heavily on its ideas about flow and constraints. Kanban systems that limit work in progress apply the same principle that the pace of the whole should be set by its slowest stage.

Questions it lets you ask

What is the one step that limits how much we deliver? Is it ever idle, or working on things nobody needs? Are other parts of the system working at their own pace, building queues in front of it? Which of our measures rewards local busyness at the expense of throughput? And once we fix this constraint, where will the next one appear?