The Method Behind the Method
How a single tested method reached more than a million supervisors in a few years — not the content of any one program, but the design that delivered it.
The previous section described the wartime crisis that TWI was created to solve: too many new plants, too many new and inexperienced supervisors, and no time to develop them the slow way. This section explains how TWI met that demand — not the content of any one program, but the design that let a single tested method reach more than a million supervisors in a few years. It is the part of the TWI story most worth a lean reader's attention, because TWI did to its own work exactly what it later asked Toyota's supervisors to do to theirs: it standardized it, taught it the same way everywhere, and improved it by doing.
By the end of this section, you should understand:
- how each "J" program was designed as a short, fixed, standardized course,
- what the four-step card was and why every supervisor carried one,
- the "multiplier effect" — TWI's train-the-trainer model — and why it spread reach so fast,
- the fidelity controls that kept the method from drifting as it spread,
- and why this amounts to standardized work applied to training itself.
1One course, taught the same way everywhere
TWI did not send consultants to study each plant and design a custom program for it. It built a small number of fixed courses and ran them, unchanged, in every kind of company — new plants, fast-expanding ones, and ordinary firms simply continuing peacetime work.
Each of the three "J" programs was deliberately short and identically shaped. The standard was ten hours of content delivered in five two-hour sessions, with a small group of about ten to twelve supervisors in each. The small group size was not incidental; it was required by the way the course taught (see §3 below).
TWI stated the design rules for itself plainly. A course had to have a specific, simple method the plant supervisor could follow; the method had to be stated in shop terms, not academic language; and there had to be a definite outline of exactly what would be covered, how much time each point would take, and how it would be delivered — "a recognized universal standard for the training sessions."
The reasoning behind "ten hours in five two-hour sessions" was practical, not arbitrary. Ten hours was judged the right amount of content to put across; two-hour sessions ran long enough to do real practice without needing a break, and the time lost opening and closing a meeting was no greater for a two-hour session than for a one-hour one. Some groups needed less than the full ten hours; some did not have a good grasp by the end of it. The standard held anyway, because a standard is what made the course repeatable everywhere.
2The four-step card
Every J program reduced its method to a four-step pattern, and every J program ended the same way: each supervisor was handed a wallet-sized card summarizing that program's four steps, to carry and consult on the job.
The card mattered out of proportion to its size. It meant the method left the classroom with the supervisor and stayed at hand on the shop floor, where the work actually happened. It was a reminder of how to proceed in a real situation, not a certificate filed away. The pattern was durable: decades later the same kind of card was still in use in Japan, nearly identical to the wartime original.
This is the first place a lean reader should pause. TWI did not trust a ten-hour course to "stick" on its own. It put the standard in the worker's pocket — a small, physical, always-available reminder of the agreed method. The principle is the same one that later governs a standardized-work sheet posted at the workstation: the standard has to live where the work is done.
3"Learn by doing"
TWI courses were not lectures. They were built on what TWI called "learning by doing" — a phrase it traced to Sophocles, who wrote in 445 B.C., "One must learn by doing the thing; for though you think you know it you have no certainty, until you try."
The pattern of each course followed from this. The opening session — known as the "Famous First" — presented a real problem and a deliberately bad solution to it, one designed to make a supervisor recognize something he himself might have done. Only then did the instructor introduce the four-step method and show how it produced a better result. After a session or two spent learning the method, the rest of the course was spent on real problems the supervisors brought in from their own work, analyzed and solved using the TWI method.
That is why the groups had to be small. Guided, individual practice on one's own everyday problems takes time and attention, and TWI was blunt about it: there is no substitute for practice. A method learned by doing it, on real work, with a coach watching, is a method the supervisor can actually use the next day.
The shape of a TWI session — show a flawed example first, then the method, then practice on the learner's own real problems — is itself a teaching standard. It assumes adults learn a working skill by trying it under guidance, not by being told about it. The same instinct runs straight through to Toyota's later development of supervisors on the floor.
4The multiplier effect
A handful of TWI staff could never have trained the supervisors of an entire wartime economy directly. The program's reach came instead from a deliberate train-the-trainer design that TWI called the "multiplier effect."
The multiplier effect (working definition): develop one standard method, then train people to train other people to train groups of people in that method — so reach multiplies through layers of trainers rather than depending on a few experts teaching everyone directly.
In TWI's own words, the aim was to "develop a standard method, then train people who will train other people who will train groups of people to use the method." In practice, the best performers in the ordinary sessions were selected to attend Master Institutes, where they were trained to become Institute Conductors — people able, in turn, to train more trainers. Each layer multiplied the next.
This is what turned a small federal program into a national one. The method did not have to be re-invented at each plant; it had only to be carried, faithfully, down one more layer of trainers.
Develops and holds the one standard method, then trains the conductors who carry it down.
Institute Conductors trained to teach the method faithfully — and to train still more trainers.
Trained in small ten-hour groups, learning the method by doing it on their own real problems.
Reached indirectly, as supervisors apply the same standard method where the work is done.
TWI's reach came from layered training, not direct teaching: a standard method developed once, then trainers training trainers training supervisors.
What to notice: The thing that multiplies is the standard method itself. Reach scales only because each layer passes on the same method, not a personal interpretation of it — which is exactly why fidelity control (§5) was non-negotiable.
5Holding the method steady as it spread
A multiplier amplifies whatever it copies. If the method drifted at each layer, the multiplier would have spread a different, weaker course at every plant. So TWI paired its train-the-trainer model with strict fidelity control over the method and its materials.
For the program to work in every possible circumstance, each new course was rigorously field-tested before national release — Job Instruction, the first of the three "J" courses, was trial-run in roughly seventy plants over about six months before it was let loose.
Once released, the materials were protected against drift:
- Trainers had to follow the lesson plans exactly, as detailed in their manuals. The instruction "WORK FROM THIS OUTLINE — DON'T TRUST TO MEMORY" appeared throughout the manuals, in some cases at the bottom of every page.
- The manuals carried timing marks showing the desired pace of each class, and told the trainer exactly what to write on the board and when.
- The manual was even designed to be readable from about five feet away, so a trainer could stay close to it while teaching.
- A trainer kept "active" status only by having taught a TWI course within the previous ninety days — capability was treated as perishable, kept up by use.
The intent was not bureaucratic. It was to guarantee that the supervisor in the hundredth plant received the same proven course as the supervisor in the first.
Notice what TWI standardized: not just the content of the course but the act of teaching it — the words, the timing, the board work, even the trainer's certification cycle. This is the move a lean reader will recognize. TWI treated training delivery as a process to be defined, taught, checked, and held — the same discipline it asked supervisors to bring to production work.
6Standardized work, applied to training
Step back from the parts and the larger point comes into view. TWI's method behind the method was, in substance, a standardized-work mindset applied to the work of training itself.
The course was a defined standard: fixed content, fixed time, fixed sequence. The four-step card put that standard in the worker's hands at the point of use. "Learning by doing" built capability through guided practice on real work rather than through telling. The multiplier spread the standard through layers without re-inventing it, and fidelity control kept the standard from eroding as it spread. Field-testing before release, and improving the course through repeated delivery, was the standard being refined by doing.
A lean reader does not have to stretch to see the resemblance to how a supervisor is later expected to run production: an agreed standard, kept where the work is done, taught by doing, held steady, and improved through use. TWI's lasting craft was less any one of its three programs than the disciplined way it delivered all of them.
Section summary
TWI reached more than a million supervisors not by customizing programs but by standardizing them: each "J" program was a short, fixed course — about ten hours in five two-hour sessions — taught the same way everywhere, built on "learning by doing" rather than lecture, and summarized on a four-step card the supervisor carried on the job. Reach came from the multiplier effect, a train-the-trainer model in which Institutes trained trainers who trained supervisors. And the multiplier worked only because the method and materials were tightly standardized and protected, so the course did not drift as it spread. In effect, TWI applied a standardized-work discipline to training itself.
The next section looks at what was being delivered through that machinery: the three J programs at a glance — Job Instruction, Job Relations, and Job Methods — one pass each, with pointers to their dedicated guides.