Continuous Improvement Register

What this tool does

CM8-85 is a continuous improvement (kaizen) register. You record improvement ideas as they are raised, track each one through evaluation, approval and implementation, and — this is the part most registers get wrong — record what the idea actually saved once somebody has measured it, kept strictly separate from what it was claimed to save when it was proposed.

The tool works out payback for every verified idea, totals the verified and claimed savings as two separate figures that are never combined, flags open ideas that have slipped past their target date, and prints a report suitable for a management review. Everything runs inside this single file: no account, no upload, no network request of any kind.

Recording an idea

An idea needs a date, a title, an area and a category to be worth recording; everything else can be added as it becomes known. Write the description so that someone who was not in the conversation understands what the idea is and where the money or benefit comes from — "fewer changeover minutes on press 2" beats "efficiency improvement".

• Implementation cost — the one-off cost of putting the idea in place: materials, contractor time, internal labour if you cost it, and any downtime taken to do the work.
• Annual saving claimed — the estimate made when the idea was raised or costed. It is allowed to be rough; it is not allowed to masquerade as a result.
• Annual saving verified — leave this blank until somebody has measured the saving. Enter 0 if verification found no saving at all; blank means "not yet measured", 0 means "measured, and there wasn't one". The distinction matters.

Not every idea has a cash figure, and that is fine. A guarding improvement raised on risk alone belongs in the register with no claimed saving — the register exists to make good ideas happen, not to make every idea produce a number.

The life of an idea

The status field carries an idea through seven stages: Idea (raised, not yet looked at), Evaluating (being costed), Approved, In progress, Implemented (done, but the saving has not been measured), Verified (measured and confirmed) and Rejected.

Two of these do the heavy lifting. "Implemented" is not the finish line — an idea sits there until its saving has been checked against reality. And "Rejected" is a respectable outcome: an honest register records the ideas that did not stack up, with the reason, so the same proposal is not re-evaluated from scratch every year. Rejected ideas are excluded from every savings total.

Claimed vs verified — the rule this tool is built on

Improvement programmes lose credibility one way above all others: someone adds up the claimed savings, presents the total as money saved, and finance cannot find it in any budget line. This register makes that mistake structurally impossible. Claimed and verified savings are held in separate columns, totalled separately, charted separately, and never added into one number anywhere in the tool.

Note what the second total deliberately leaves out: once an idea is verified, its claim drops out of the claimed total — the measured figure replaces the estimate rather than sitting alongside it. Rejected ideas' claims are excluded because the idea is not happening. Both figures are annual rates, in your currency, at the point the saving was measured or estimated.

The claimed total is still useful — it is the size of the prize if everything in the pipeline delivers as hoped. Quote it that way: "a further X claimed, not yet verified", never "X saved".

Payback

Payback answers one question: how many months of the measured saving does it take to earn back the implementation cost?

The register shows "—" for every idea that is not yet verified, and the help you are reading is the explanation: a payback computed from a claimed saving is an estimate divided into a cost, and printing it in the same column as measured paybacks would launder it into looking like a result. If you want an indicative payback before verification, the entry form shows the arithmetic as a preview while you type — clearly labelled — but it is never stored, totalled or charted.

Ideas verified at zero saving are excluded from the average (their payback is not a number, and including them as anything would distort it) but they remain in the register, visibly, as verified ideas that saved nothing. The payback target in the settings is a marker only: paybacks at or under it are shown in green, over it in amber. It approves nothing and blocks nothing.

Verifying a saving

A verification is a measurement, not a signature. Compare something real from before the change with the same thing after it: metered energy, invoiced prices, changeover logs, scrap weights, downtime records. Measure for long enough to smooth out noise — a few weeks at least, longer for anything seasonal — then scale to an annual rate and enter it.

The "verification required above" setting drives a flag: any implemented idea whose claimed saving exceeds the threshold is marked "Verify saving" in the register until it is verified. Set the threshold at the level where an unverified claim could distort your reporting; small claims below it can be accepted on judgement, and verifying a claim of 200 rarely justifies the effort of measuring it.

Expect verified figures to differ from claims in both directions — press 2's changeover saving in the sample data verified at 20% below claim, while the packaging change verified above it. A register whose verified figures always exactly match the claims has not really been verified.

Categories

Ideas are classified under the six headings common in improvement programmes: Safety, Quality, Cost, Delivery, Morale and Environment. Pick the primary benefit — many ideas touch several. The category charts show where your verified savings actually come from, which is frequently not where the claims said they would: a register can be full of cost claims while the money that materialises is in delivery.

Watch the safety and morale categories in particular. If they are empty, the programme has drifted into being a cost-cutting scheme, and people stop raising the ideas that make the place better to work in.

Target dates and the overdue flag

Every idea can carry a target date for completing implementation. An idea in an open status — Idea, Evaluating, Approved or In progress — whose target date has passed is flagged Overdue in the register and listed in the "Open ideas past their target date" table with the days late. Implemented ideas past their target are not flagged as overdue (the work is done); their outstanding debt is verification, which has its own flag.

Nothing kills a suggestion scheme faster than ideas disappearing into a list. The overdue table is the page to bring to every improvement meeting: it is the list of promises the programme has made and not yet kept.

Reading the charts

Ideas by status shows the shape of the pipeline. A healthy register has ideas at every stage; a pile-up at "Evaluating" means ideas outrun the capacity to assess them, and a pile-up at "Implemented" means nobody is doing the measuring.

Ideas raised per month tracks whether ideas keep coming. The flow usually dries up not because people run out of ideas but because earlier ideas went nowhere.

Verified savings by category contains measured figures only, and is empty until your first verification — deliberately so.

Claimed vs verified by category shows both figures side by side as separate segments. Resist reading a bar's full length as a total: the segments are different kinds of number, and the chart exists to show the gap between them, not their sum.

Printing and sharing

Print Report produces a report from whatever the current filter shows: headline figures with verified and claimed totals clearly separated, all four charts, the top verified savers with their original claims alongside, the overdue list, the full register and your notes. Print to PDF to circulate it. The scope line states the filter in force — clear the filters before issuing anything described as covering the whole programme.

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 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 and includes every filtered record. Reset asks twice, then erases everything this tool has stored. There is no undo.

Accuracy & disclaimer

This tool calculates from what you enter. It keeps claimed and verified savings separate and computes payback from verified figures only, but it cannot tell whether a verification was done honestly, whether the implementation cost is complete, or whether an annual rate measured over six weeks will hold for a year. Savings that depend on volumes, energy prices or exchange rates move when those move.

It is a record-keeping and calculation aid for an internal improvement programme — not a financial reporting tool, not an investment appraisal (there is no discounting; payback here is simple payback), and not a substitute for finance checking any figure that reaches a budget or a business case.