Kaizen Tracker

What this tool does

CM8-261 is a working pipeline for continuous-improvement ideas. Each row is one idea: who raised it, what waste it attacks, how big the fix is, what it might be worth, and where it stands — from logged, through review and implementation, to a verified benefit. The tiles show whether the system is alive (ideas coming in, people participating, ideas getting done), the charts show where the shop floor sees waste and whether estimates hold up, and the two tables are your standing meeting agenda: what is waiting, and what got done.

Everything runs inside this single file. There is no account, no upload and no network request of any kind, so your ideas, names and numbers never leave the computer you are using.

What kaizen actually is

Kaizen is continuous improvement in small steps, made by the people who do the work. The person who carries the part 40 metres is the person who knows the bench is in the wrong place — no consultant will find that faster, and no manager will feel the ache in the same shoulders. A kaizen system is simply a reliable way for that knowledge to become action: the observation gets written down, somebody answers it quickly, the small ones get done immediately, and the results are visible to everyone.

What kaizen is not is a suggestion box. A suggestion box is where ideas go to be ignored: anonymous slips, a committee that meets quarterly, no feedback, no names on wins, and eventually no slips. The differences that matter are speed of response, visibility of outcomes, and recognition — which is why this tracker is built around ages, statuses and named raisers rather than a savings ticker.

Nor is kaizen only about big money. Moving a bench, labelling a valve, raising a packing bench so backs stop aching — most improvements are individually small. Their value compounds: a workplace that fixes fifty small irritations a year is measurably faster, safer and easier to work in than one that waits for a big project, and — more importantly — it has fifty pieces of evidence that raising a problem leads to something.

Participation is the health metric

The tile to watch is not the money — it is participation: how many different named people have raised at least one idea. A total savings figure can be propped up by one engineer and a spreadsheet; a rising count of distinct contributors cannot be faked, and it is the leading indicator for everything else. When participation is broad, ideas keep coming after the founders lose interest, problems surface while they are still cheap, and the improvements stick because the people affected designed them. When participation is three names out of sixty employees, the money column is telling you about three people's energy, not about your organisation — and the fix is to make raising an idea easier and the response faster, not to chase a bigger number.

The eight wastes

The classic memory aid is TIMWOOD, plus an eighth that matters more than the other seven:

• Transport — moving product around: pallets shuttled between buildings because the processes are laid out in the wrong order.
• Inventory — stock waiting: a month of brackets on the floor tying up cash and hiding problems underneath.
• Motion — people moving unnecessarily: fifty trips a shift to a bench that could sit three metres away.
• Waiting — idle time: an operator watching a machine, a pallet queue at the one stretch wrapper at 15:00.
• Overproduction — making too much, too soon: running 500 because setups are painful, when the customer ordered 200.
• Over-processing — doing more than needed: polishing a face nobody sees; a paper traveller re-keyed into three systems.
• Defects — errors and rework: mis-keyed goods-in quantities surfacing days later as shortages and wrong picks.
• Underused talent — ideas and skills wasted: the fitter who has known the fix for two years and was never asked. Every empty week in this tracker is this waste, measured.

The waste donut shows where your people see friction. It does not need to be balanced — a plant with a layout problem should show a pile of motion and transport — but a category that never appears is worth a walk-around.

Effort triage — “just do it” first

Every idea gets one of three effort sizes, and the point of the smallest is in its name. A “just do it” — under an hour — should not enter a queue at all: the team leader says yes on the spot, it happens this week, and the tracker records that it happened. In a healthy system these are the bulk of the volume, because most of what irritates people daily is small.

If everything in your pipeline needs approval — if the “just do it” count is near zero and every idea waits for a meeting — the system is dead, whatever the savings total says. People learn in about a month that raising an idea means paperwork, and they stop. Reserve real review for days and weeks items, where money or coordination genuinely warrant it, and let the small ones run.

Estimating and verifying benefits

Put an honest annual estimate on ideas that have one — time saved priced at a loaded rate, scrap avoided, stock reduced — and put zero on the safety and morale wins rather than inventing a number. Then, after implementation, measure what actually happened and enter it as the verified benefit, even when the honest answer is less than the estimate, or nothing.

The realisation chart exists because estimates inflate. It is nobody's fault — people are enthusiastic about their own ideas — but a savings total built from unverified estimates drifts steadily away from reality, and eventually a finance director checks one number, finds it soft, and stops believing all of them. A verified 1,800 beats an estimated 2,500 in every argument that matters. If your realisation buckets pile up below 50%, halve the next round of estimates; if they pile up above 110%, your people under-promise, which is a nicer problem but still worth knowing.

Names on wins

The raised by field is required, and it is deliberate. Recognition is most of the wage kaizen pays: the wins table prints the raiser's name next to each implemented idea, and reading three of those out at a team meeting does more for next month's pipeline than any poster. Credit the observation, not just the execution — the person who spotted the waste matters as much as the person who fixed it, and they are often not the same person. If your culture allows one further step, let people present their own implemented idea in two minutes at the weekly meeting. Pride is renewable.

The weekly 15-minute review

The pipeline table is a standing agenda. Once a week, same time, fifteen minutes, standing up:

1. New ideas since last week — read each aloud, thank the raiser by name, and triage: “just do it” items get a yes now; bigger items get an owner and go to review.
2. Anything marked stale — ideas sitting at “idea” or “in review” beyond your flag window get an answer today: approve it, decline it with a reason, or name what information is missing and who is getting it.
3. In progress — one sentence each: still moving, or stuck on what?
4. Done since last week — say the names. Book a measurement date for anything with an estimate worth verifying.

That is the whole method. The tool's job is to make the agenda build itself.

Declining ideas without killing the flow

Some ideas should be declined — too expensive for the benefit, solving a problem that is already being fixed another way, or moving waste rather than removing it. Decline them kindly, with the reason written down, and fast. The sample pipeline keeps a declined idea on the register with the reason in the idea text: the second pallet wrapper was a real observation about a real queue, answered honestly within days — the queue was fixed by staggering despatch instead.

A fast, reasoned no keeps the raiser raising. A slow no — the idea aging silently in review for two months — teaches everyone watching that the pipeline is where ideas go to die, and it costs you not one idea but all the ideas that person and their neighbours never raise afterwards. If you find declining awkward, remember the alternative is worse: nothing kills participation like silence.

FAQ

Who should be able to log ideas? Everyone who does the work, in their own name. If the tracker lives on one supervised machine, agree that anyone can dictate an idea to whoever runs it — the barrier to entry should be a sentence, not a form.

Do declined ideas stay on the register? Yes. They carry their reason, they show the system answers everything, and this year's decline is sometimes next year's approval when the volumes change.

What if an idea saves nothing measurable? Log it with a zero estimate and the benefit type that fits — safety and morale wins are often the most valuable rows on the register precisely because no money column can hold them.

Should we pay per idea? Most experienced operations avoid paying per idea — it buys volume, not improvement, and it turns colleagues into competitors. Recognition, speed and visible results are the durable rewards. If you share gains, share them with the whole team on verified results.

One idea or one project per row? One idea. If an idea grows into a project with five parts, log the parts as their own rows — small rows get done; big rows stall.

Saving your work

Ideas, settings and the report header are written to this browser's local storage as you type, and the toolbar shows the time of the last save. That storage belongs to one browser on one computer: another browser, a private window, a second machine or a clean-up tool that clears site data will not have it.

Treat Export .json as the real save — one file containing everything, which Import .json restores anywhere. Export CSV gives you the register for spreadsheet work. Reset asks twice, then erases everything this tool has stored. There is no undo.

Accuracy & disclaimer

The arithmetic here is simple and the tool does it faithfully: ages in days, verified ÷ estimated, counts by month and by waste. Everything that matters sits underneath: whether the estimates were honest, whether anyone measured after implementation, and whether the pipeline gets answered fast enough to keep people contributing. Benefits are estimates until measured, and the totals shown are only as sound as the verified figures entered.

This is a record-keeping and prioritisation aid for an improvement programme, not accounting, not a cost statement, and not financial advice. The tracker's real product is participation — the money column is there to rank effort, not to be believed unaudited.