A kaizen event is a scheduled, multi-day improvement sprint where a cross-functional team pulls off their normal jobs to fix one process; daily kaizen is the everyday habit of operators spotting and fixing small problems as part of the work itself. Events deliver big step-changes on a calendar; daily kaizen delivers a steady drip of small ones. A plant needs both, and each covers the other's weakness.
Most plants have heard of the event and run a few a year. Far fewer have a real daily-kaizen habit, and that imbalance is the quiet reason their event gains keep fading. The word kaizen, popularized for Western readers by Masaaki Imai's 1986 book Kaizen: The Key to Japan's Competitive Success originally meant the daily, everybody, everywhere kind, continuous incremental improvement as a way of working. The multi-day event is largely a Western adaptation: a way to manufacture improvement momentum in organizations that do not yet have the daily habit. The Lean Enterprise Institute makes the same point in its own terms, warning that organizations should not rely solely on kaizen events but foster continuous improvement through daily engagement across every level (LEI, Kaizen). Understanding the difference is the difference between improvement that compounds and improvement that resets. Both are core tools of lean manufacturing.
What is the difference between a kaizen event and daily kaizen?
The difference is cadence, scope, and who does it. A kaizen event is a discrete, scheduled thing, typically three to five days, with a chartered team, a target, and a report-out. Daily kaizen has no start date; it is the accumulated small improvements operators make continuously, one problem at a time, inside their own work. One is an expedition; the other is how you walk every day. The table below lays the two side by side.
| Dimension | Kaizen event | Daily kaizen |
|---|---|---|
| Cadence | Scheduled, a few times a year | Continuous, every shift |
| Duration | Three to five days, full-time | Minutes, woven into normal work |
| Scope | One big cross-functional process | Small problems in one's own area |
| Who | Chartered cross-functional team | Operators and their team leader |
| Problem type | Structural, needs everyone at once | Local, visible, fixable now |
| Output | A step-change plus new standard | Many small gains that compound |
| Main risk | Gains fade after the team disbands | Never cracks the big problems |
Neither column is better; they are aimed at different targets. The mistake is treating them as competitors, or worse, treating the event as the plant's entire improvement system and expecting a few weeks a year to carry twelve months of progress.
When should you run a kaizen event instead of daily kaizen?
Run an event when the problem is bigger than any one person's area and needs several functions solving it together in a compressed window. Reserve the event for problems that pass a simple test: does fixing it require maintenance, engineering, quality, and operators redesigning something in the same room? Is the current state genuinely unknown and in need of mapping? Is there a step-change worth pulling a team off the line for a week to capture? A full kaizen event run over three to five days, is the right instrument for a 52-minute changeover that should be 30, or a material flow that crosses three departments.
Reach for daily kaizen when the problem is local, visible, and fixable now: a tool that lives two steps too far away, a label that peels, a form with a field nobody uses. These never justify a chartered event, and if they wait for one, they never get fixed. Daily kaizen is also where the culture actually lives, because it is the mechanism that puts improvement in every operator's hands every shift, not just on the calendar.
How do events and daily kaizen reinforce each other?
They form a cycle: the event creates a new standard, and daily kaizen holds and extends it. When a kaizen event ends, it hands the area a better process and a new standard work. Daily kaizen is what keeps that standard alive, operators running it, spotting the small drifts, and pushing it a little further every week, so the next event starts from a higher baseline instead of re-fixing the last one. Run the other direction, daily kaizen surfaces the patterns and pain points that tell you where the next event should aim. The two are a ratchet: the event lifts, the daily habit holds the gain and inches it up, the next event lifts again.
Break the cycle and both halves suffer. Events with no daily habit to catch them produce the classic sawtooth: a sharp gain during the week, then a slow slide back as the old pressures return and no one is minding the standard day to day. Daily kaizen with no events plateaus, because the big structural problems never get the concentrated cross-functional attention they need. The plants that compound improvement are the ones running both loops on purpose.
How do you build a daily kaizen habit?
You build it by making small improvement a normal, low-friction, visible part of the shift, not a special project. It does not appear by announcement. Here is the sequence that tends to work:
- Give people an easy way to raise a problem. A card, a board slot, a tablet field, whatever removes the friction between noticing a problem and recording it. If raising an issue takes ten minutes and a manager's ear, nobody does it.
- Make the problems and their status visible. Post them where the team stands, using visual management so the whole crew sees what is raised, what is in progress, and what is done. Invisible problems do not get solved and invisible progress does not motivate.
- Talk about them daily. Fold a two-minute improvement item into the shift huddle or gemba walk. A standing daily moment is what turns intention into habit.
- Push authority and fixing down to the floor. Let the team make the small fix without a work order. A daily-kaizen system where every idea waits on approval dies of latency.
- Close the loop, visibly. When a problem is fixed, mark it done and thank the person. Nothing kills a daily habit faster than ideas that vanish into a manager's inbox. Track open items the way a kaizen newspaper does after an event: who does what by when.
How much of each should a plant run?
As a rule of thumb, daily kaizen should be doing the majority of the work, and events should be rare, deliberate, and reserved for problems that genuinely need them. A plant that runs a dozen events a year and has no daily habit is almost certainly using events to paper over the absence of a system, and it will feel busy while standing still. A plant with a strong daily habit can often go months between events, because most problems get solved as they appear, and the events it does run are aimed at the few structural problems the daily loop cannot reach.
A simple filter helps route a problem to the right loop. Ask three questions in order: Can the operator or team leader fix it this week without a work order? If yes, it is daily kaizen, do it now. Does it need two or more functions redesigning something together? If yes, it is a candidate for an event. Is the current state genuinely unknown and worth mapping? If yes, an event earns its cost. Anything that clears the first question should never wait for an event, and anything that trips the second should not be forced into a hurried daily fix that treats a symptom.
The daily loop also depends on the plant's everyday communication working. A daily-kaizen item raised on first shift only helps if second and third shift hear about it, which is why a disciplined shift handover process and a daily huddle are the connective tissue that keeps small improvements from dying at the end of a shift.
Why do event-only plants keep re-fixing the same things?
Because without a daily habit, nobody holds the standard between events, so it drifts, and next year's event re-solves last year's problem. The pattern is easy to spot: a plant that runs impressive events, celebrates real gains on Friday, and then finds the same area on next year's event schedule. The gain was never held, because holding it is daily work and the plant only improves on the calendar. This is the same reason kaizen event gains tend to evaporate within about 90 days, covered in depth in our kaizen events guide.
The common root cause is measurement. Daily kaizen needs the team to see how the process is running today but if the numbers live on paper and surface at month end, there is nothing to react to day to day, and the daily loop never gets traction. Plants that sustain both loops capture production data where the work happens and put it on a live board, so operators see the trend on their own shift and a drift shows up in days rather than at the next event. That live feedback is what a daily habit runs on, and it is the shift CLS made from paper logs read the next morning to production data visible during the shift itself. Events give you the step-changes. A daily habit, fed by real-time visibility, is what keeps you from paying for the same step-change twice.