Beyond Amazing
The Strategy Toolkit

Operations & process

Kaizen

The discipline of continuous, incremental improvement involving everyone, managers and workers alike: standardise the current best way of working, surface problems against that standard, improve in small low-cost steps, then lock each gain in as the new standard and go again.

Also known as Continuous improvement, Kaizen events (point kaizen). First set out by Masaaki Imai (codifying Toyota and wider Japanese practice) in 1986; the primary source is cited in full below.

Where this is contested

Kaizen is presented as Japanese management philosophy, but its improvement method descends substantially from the American Training Within Industry programmes taught in occupied Japan from the late 1940s; the TWI Job Methods course, 'improvement in four steps', was translated as kaizen long before Imai's 1986 book named and codified the practice for the West.

Format
Process / loop
Level
Team · Business unit
Best for
Plan execution
Decision stage
Execute · Review
Difficulty
Introductory
Time to apply
A team routine can start within a week; the point is that it never finishes. Individual kaizen events typically run three to five days.

Plate · The model

Standardisethe currentbest waySurfaceproblems andwasteImprove insmall stepsMake theimprovementthe newstandard
The 4 stages of Kaizen, run as a continuous clockwise loop.
I

The components

1

Standardise the current best way

The current best-known method is agreed, documented and followed, providing the stable baseline that makes problems visible and gains holdable.

Signals of strength
Standards written with, and owned by, the people doing the work · Deviation from standard is noticeable within the shift, not at month end · Standards describe the current best method, not an aspiration

2

Surface problems and waste

Gaps between standard and reality, and the classic wastes, are made visible where the work happens, and treated as material for improvement rather than grounds for blame.

Signals of strength
Managers regularly at the gemba observing, not just reviewing reports · Visible daily measures the team maintains itself · People raise problems without fear, and problems raised get responses

3

Improve in small steps

Small, low-cost countermeasures are trialled continuously through PDCA, by the people closest to the work, at both daily point-kaizen scale and event scale.

Signals of strength
Many small changes per month rather than one large project per year · Experiments checked against a measure before being kept · Suggestions acted on quickly, with feedback to the suggester

4

Make the improvement the new standard

Proven countermeasures are written into the standard and trained, so each gain becomes the floor for the next cycle instead of decaying back to the old way.

Signals of strength
Standards revised frequently and dated · Training follows every standard change · Audited gains still present ninety days later

II

When it earns its keep

  • A process is stable enough to have a standard, and you want it to get a little better every week through the people who run it rather than through occasional big projects.
  • Quality, delivery or cost problems keep recurring and being worked around. Kaizen turns each recurrence into an improvement rather than a rescue.
  • You want improvement capability in the workforce itself, so that suggestions, small experiments and problem-solving become normal work, not a specialist function.
  • Capital is scarce. Kaizen's bias is towards many small, cheap, reversible changes over few large investments.

And when it doesn't

  • The process or business model is fundamentally wrong. Incremental refinement of the wrong thing entrenches it; step-change redesign or innovation is a different activity, and Imai himself distinguishes kaizen from innovation.
  • There is no standard and no stability. If the process runs differently every day, standardise first; you cannot improve what you cannot repeat.
  • Management wants a one-off cost-cutting exercise. Kaizen run as a headcount tool destroys the trust that makes people surface problems at all.
  • The change needed is structural, for example a site closure or a system replacement. Kaizen works within a design, not on decisions above it.
III

How to run it

Before starting, gather the inputs the analysis depends on:

  • A current standard for the work: the present best-known method, documented simply enough that deviation from it is visible.
  • Measures the team can see daily, such as defects, rework, delays or output against plan, so problems announce themselves.
  • Time and permission for the people who do the work to stop, study problems and try countermeasures, with management attention when they do.
  • A simple improvement mechanism: suggestion handling, short team huddles, or structured kaizen events for bigger targets.
  1. 1

    Standardise the current best way

    Agree and document how the work is done now, at the level of the people doing it. Imai is blunt on this point: there can be no improvement where there are no standards, because without a standard there is nothing to improve from and no way to hold a gain. Standards in kaizen are the current best method, owned by the team, not a compliance document.

  2. 2

    Surface problems and waste

    Make gaps visible: defects, rework, waiting, overproduction, excess motion and inventory, and any deviation from standard. Go to the gemba, the place where the work happens, and observe directly. A workplace where no problems are visible is hiding them, and kaizen treats every problem as an improvement opportunity rather than an occasion for blame.

  3. 3

    Improve in small steps

    Attack one problem at a time with small, cheap, quick experiments, using PDCA as the improvement engine: plan a countermeasure, try it, check the result against the measure, act on what you learn. Prefer ideas from the people who do the work. This runs at two scales: daily point kaizen on single processes, and focused kaizen events where a team improves a target area intensively over a few days.

  4. 4

    Make the improvement the new standard

    When a countermeasure works, write it into the standard, train everyone to it, and make the new level the baseline the next improvement starts from. Imai pairs this maintenance role with improvement itself: management's job is to hold standards and raise them. Skipping this step is why improvements fade; the cycle then repeats from the new standard, indefinitely.

IV

Reading the result

Not a document but a running capability: a stream of small implemented improvements, each held by a revised standard, with visible measures showing the trend. Kaizen events additionally produce before-and-after results for a target area and a follow-up list.

  • Judge kaizen by trend and retention, not by any single improvement: are the measures moving, and are last quarter's gains still in place?
  • Watch the ratio of ideas implemented to ideas suggested, and the speed of response. A suggestion scheme with a six-week queue is a suppression scheme.
  • If results spike after events and decay between them, the standardise step is failing, and the system is running episodic improvement, not kaizen.
V

A worked example

A commercial laundry serving hotels builds a daily kaizen routine

A commercial laundry processes 40 tonnes of hotel linen a week on two shifts. Rewash rate is 6 percent, trolleys of sorted linen queue in front of the ironer line, and hotel complaints about short deliveries arrive weekly. Rather than commission a project, the operations director establishes a kaizen routine with the shift teams, starting on the sorting-to-ironer flow.

Standardise the current best way
The teams find sorting is done four different ways across the shifts, so they agree one standard: sort categories, trolley loading pattern, and an ironer changeover sequence between sheet and duvet-cover runs, photographed and posted at each station. Simply agreeing the standard exposes that half the rewash comes from mixed-category trolleys reaching the ironer.
Surface problems and waste
A whiteboard at the ironer line tracks rewash, trolley queue length and short deliveries each shift, filled in by the team. Ten-minute morning huddles review it. Within a fortnight the board shows the pattern: rewash spikes on Mondays, when weekend agency staff have been sorting, and the trolley queue peaks whenever the ironer changeover exceeds twenty minutes.
Improve in small steps
Small PDCA cycles follow: colour-coded trolley cards by linen category (rewash from mis-sorting halves in three weeks); a laminated one-page sorting guide for agency staff, trialled on one weekend and kept; changeover tooling staged on a shadow board beside the ironer, cutting changeover from 22 to 9 minutes. One idea, pre-folding towels at sorting, fails its check, adding handling time for no downstream gain, and is dropped without recrimination.
Make the improvement the new standard
Each successful countermeasure is written into the posted standards and covered in the five-minute start-of-shift brief; the agency induction now includes the sorting guide. After four months, rewash sits at 3.1 percent, short-delivery complaints are down two thirds, and the team is choosing its next target, the folding area, from the board.

The read. Nothing in this required capital or consultants; it required standards, a visible board, ten minutes a day and management that responded to what the board showed. The honest reading is that the gains are only as durable as the routine: when a maternity cover manager skipped huddles for three weeks, the queue and the rewash rate both began drifting back, which is exactly what the kaizen-sustainability literature predicts.

VI

Pitfalls

  • Running kaizen as a programme of events with nothing between them. Events without daily kaizen and standard-keeping produce results that fade within months.
  • Skipping standardisation because it feels bureaucratic. Without the standard, gains have nowhere to live and every improvement is temporary.
  • Harvesting savings as immediate headcount cuts. People will not surface waste that eliminates their own jobs; credible kaizen redeploys freed capacity.
  • Confusing a suggestion box with kaizen. The work is management responding, coaching and closing the loop, not collecting ideas.
  • Using kaizen where redesign is needed, polishing a process the strategy is about to make irrelevant.
VII

What the critics say

Empirical study of kaizen events shows their results frequently decay: in a field study of 65 events across eight organisations, sustained outcomes depended on factors such as continued management attention, learning and goal clarity, and many organisations fail to hold gains after the event ends.

Glover, W. J., Farris, J. A., Van Aken, E. M. and Doolen, T. L. (2011) 'Critical success factors for the sustainability of Kaizen event human resource outcomes: an empirical study', International Journal of Production Economics, 132(2), pp. 197-213.

The reengineering school argued that incremental improvement is structurally incapable of delivering the step changes competitive conditions sometimes require, and can absorb energy that should go into redesigning the process wholesale.

Hammer, M. and Champy, J. (1993) Reengineering the Corporation: A Manifesto for Business Revolution. New York: HarperBusiness.

Labour-process critics contend that continuous improvement can function as work intensification, with workers surfacing slack that is then removed from their own jobs, a tension the sustainability literature echoes in findings on trust and employee outcomes.

VIII

Sources and further reading

  • Imai, M. (1986) Kaizen: The Key to Japan's Competitive Success. New York: McGraw-Hill. ↗
  • Imai, M. (1997) Gemba Kaizen: A Commonsense, Low-Cost Approach to Management. New York: McGraw-Hill.
  • Schroeder, D. M. and Robinson, A. G. (1991) 'America's most successful export to Japan: continuous improvement programs', Sloan Management Review, 32(3), pp. 67-81.
  • Huntzinger, J. (2002) 'The roots of lean: Training Within Industry, the origin of Japanese management and kaizen', Target, 18(1). ↗

Pairs well with PDCA Cycle·Value Stream Mapping·Five Whys·DMAIC·Theory of Constraints·compare side by side

Near neighbours (computed from shared tags)·ADKAR Model·AI Risk Management Framework (NIST AI RMF)·AIDA