Kaizen
Kaizen
改善Kaizen is the practice of improving a business through a steady stream of small changes, most of them suggested and carried out by the people who actually do the work, rather than through occasional big overhauls ordered from the top.
What Is Kaizen?
In Japanese, kaizen is an everyday word. It simply means improvement, the way an English speaker might say a school or a recipe had been improved. In business, it has come to mean something narrower: a habit of improving continuously, in small steps, with everyone involved, from the plant manager to the newest person on the line.
The idea runs against the instinct of most managers, which is to wait for a big fix. A kaizen company assumes that every process can be done a little better tomorrow than it is today, and that the people closest to the process are the ones most likely to spot how. One small change rarely matters much. Thousands of them a year, kept up for decades, can add up to a company that is very hard to compete with.
An American Wartime Program With a Japanese Name
Kaizen is often described as uniquely Japanese, but its business meaning has American roots. In 1940, as the United States geared up for war, the government launched Training Within Industry, or TWI, to teach supervisors how to train new workers quickly, handle people and improve their jobs. By the end of the Second World War, more than 1.6 million workers in over 16,500 plants had been certified through the program. One of its courses, Job Methods, taught supervisors to break a job into its details, question every step and look for small, practical improvements using the equipment they already had.
After the war, the U.S. occupation authorities in Japan wanted to restart Japanese industry. They contracted a Cleveland company, TWI Inc., led by former TWI instructor Lowell Mellon, which spent six months training 35 Japanese master instructors, who then spread the courses across the country. Job Methods was translated into Japanese in 1950. Japanese manufacturers took the American method of small, continuous improvement and gave it an ordinary Japanese word.
Two other American influences arrived at the same time. In the summer of 1950, the statistician W. Edwards Deming lectured hundreds of Japanese engineers and managers on statistical quality control at the invitation of the Union of Japanese Scientists and Engineers (JUSE), which created the Deming Prize in his honor. His cycle of Plan, Do, Check, Act became the working rhythm of kaizen. And in 1951, Toyota executive Eiji Toyoda visited Ford in the United States and came home impressed by its suggestion system, which rewarded workers for ideas.
What Kaizen Was Used For in Postwar Japan
Japanese industry in the late 1940s was short of almost everything except people. Capital was scarce, and the anti-inflation austerity program of 1949, known as the Dodge Line, pushed many companies into bankruptcy. Toyota itself nearly went under: a 1950 labor dispute ended with about a quarter of its workforce leaving. A company in that position could not buy its way to higher output with new machines. It had to get more from the factories, tools and workers it already had. That is exactly what kaizen does well.
In practice, it was put to work in three main ways:
- Cutting waste on the shop floor. Supervisors trained in Job Methods looked for wasted motion, waiting, excess inventory and rework, and removed it a little at a time. At Toyota this became part of the Toyota Production System developed under Taiichi Ohno.
- Fixing Japan's reputation for quality. In the early postwar years, "made in Japan" often meant cheap and unreliable. Statistical quality control, Deming's cycle and company-wide quality programs used constant small corrections to drive defects down, until Japanese cars and electronics became known for the opposite.
- Turning workers into problem-solvers. Suggestion systems and quality circles gave ordinary employees a formal way to propose and test improvements. By the end of 1962, the first year of the quality circle movement, 36 companies had registered circles with JUSE.
Toyota's 40 Million Suggestions
Toyota's suggestion system is the clearest record of kaizen at work. In its first year it drew 789 suggestions, according to Yuzo Yasuda's history of the program. Toyota's own records show the running total passing one million in 1974, then speeding up dramatically: from five million in 1981 to ten million in 1984 and twenty million in 1988. By 2011 it had passed forty million.
Cumulative suggestions in Toyota's Creative Idea Suggestion System. Points are the milestones Toyota published; the line joins them and does not show year-to-year detail. Source: Toyota Motor Corporation, 75 Years of Toyota; launch-year count from Yuzo Yasuda, 40 Years, 20 Million Ideas (1991).
Dividing the gaps between those milestones gives a rough sense of the pace. In the 23 years to 1974, Toyota averaged well under 100,000 suggestions a year. Between 1984 and 1988 it averaged about 2.4 million a year, the most intense stretch in the record. The pace slowed after that, but even the slowest recent stretch, 1995 to 2011, still averaged more than half a million ideas a year.
Average suggestions per year between each pair of published milestones, calculated by DaveManuel.com from Toyota's milestone dates. Months are approximate where Toyota gave only the year.
How Kaizen Works
Most kaizen programs follow Deming's cycle, repeated over and over on one small problem at a time:
- Plan. Go to the place where the work happens (the Japanese call it the gemba), watch the process and pick one specific problem: a step that wastes time, causes defects or makes the job harder.
- Do. Try a small, cheap change, usually suggested by the people doing the job.
- Check. Measure whether it actually helped: time per unit, defects, distance walked, cost.
- Act. If it worked, make it the new standard way of doing the job. If not, go back and try something else.
- Repeat. The new standard becomes the starting point for the next improvement. The cycle never really ends.
Kaizen vs. Innovation
Imai drew a sharp line between kaizen and the big-leap innovation that Western companies tended to prize. The two are not enemies, and the best companies do both, but they work very differently:
| Kaizen | Innovation | |
|---|---|---|
| Size of each change | Small, incremental | Large, a step change |
| Pace | Continuous, every day | Occasional |
| Who drives it | Everyone, especially frontline workers | Specialists, engineers, senior management |
| Cost | Low; mostly time and attention | Often high; new equipment or technology |
| Risk | Low; easy to reverse | Higher; a big bet can fail |
| Where it happens | On the shop floor or in the office | In the lab, boardroom or a new plant |
| What it needs to last | Discipline and standards, or gains slip back | Upkeep, or the new system decays over time |
The Same Plant, Run Two Ways: NUMMI
The strongest evidence that kaizen was a method, not something about Japanese culture, came from a car plant in California. General Motors closed its Fremont assembly plant in 1982. It was known for poor quality, absenteeism of around 20 to 25% and a backlog of more than 700 unresolved grievances. In 1984 the plant reopened as NUMMI, a joint venture between GM and Toyota, run on Toyota's system with kaizen at its core and with many of the same union workers rehired.
The change was dramatic. Absenteeism fell to about 3 to 4%. More than 70% of workers took part in the suggestion program each year in its first four years. And in John Krafcik's comparison of plants for MIT, NUMMI needed about half the hours of work per car of comparable GM plants, close to the level of Toyota's own factory in Japan.
Assembly hours per car, before adjusting for differences in the cars and plants. Source: John Krafcik (1986), as reported in Paul Adler, The "Learning Bureaucracy": New United Motor Manufacturing, Inc. GM Fremont was measured before its 1982 closure.
From Toyota City to the Stock Market
Through the 1980s, as Japanese cars and electronics took market share from American and European rivals, Western companies went looking for the secret. Imai's 1986 book gave them the vocabulary. Consultants who had trained under Taiichi Ohno began working with Western manufacturers, and American companies developed the "kaizen event" or "kaizen blitz," a focused workshop of a few days in which a team redesigns one process from top to bottom.
One of the best-known converts was Danaher. Its Danaher Business System traces its roots to 1986 at the company's Jacobs Manufacturing unit, maker of the Jake Brake, in Bloomfield, Connecticut. In early 1988 the unit brought in Yoshiki Iwata and Chihiro Nakao, two disciples of Taiichi Ohno who ran the consultancy Shingijutsu. After one all-night session redesigning a single production cell, the unit began working with them one week a month, and the approach became the core of how Danaher ran the businesses it owned and acquired.
Why Kaizen Matters to Investors
What it can do
- Lift margins year after year without large capital spending, because most improvements cost little.
- Make acquisitions more valuable, if a buyer can apply its improvement system to the companies it buys.
- Build an advantage that is hard to copy, because it lives in habits and training rather than in a patent or a machine.
What to watch for
- Companies that use the word in presentations without the results to show for it: steadily improving margins, inventory turns or quality measures over many years.
- Gains that stop once the original leaders or consultants leave, a common way kaizen programs fade.
- Small steady gains are not a substitute for innovation when a product or technology becomes obsolete.
What You Actually Need to Know
Kaizen means improvement, and in business it means improving continuously, in small steps, with everyone involved. Its methods grew out of an American wartime training program that the U.S. occupation brought to Japan, combined with Deming's quality cycle and a suggestion system borrowed from Ford. Postwar Japan used it to rebuild with little capital, fix its reputation for quality and turn factory workers into problem-solvers, and Toyota's 40 million suggestions are the clearest record of what that looks like when it is kept up for decades. For investors, the word itself proves nothing; the long record of results is what counts.
This entry is for information only and is not investment advice.
