CAPA Tracker

What this tool does

CM8-228 is a working CAPA register. You raise corrective and preventive actions, record the containment, the root cause, the action plan and the owner, and track each one through to a verified close. The tool shows you the health of the register — how many are open, how old they are, where they come from, whether closed ones actually worked — and prints a report fit for a management review or an external audit.

Everything runs inside this single file. There is no account, no upload and no network request of any kind, so complaints, audit findings and supplier problems never leave the computer you are using.

What a CAPA is

CAPA stands for corrective and preventive action. It is the mechanism a quality system uses to make sure problems get permanently fixed rather than repeatedly patched. Every serious quality standard — the ISO 9001 family, its medical and automotive relatives, and most industry schemes — expects some version of it: when something goes wrong, or when you can see it about to, you record it, find out why, act on the cause, and then prove the action worked.

The register matters because memory does not scale. Without one, the same problem gets "fixed" by three different people in three different years, each time with a reminder at the shift briefing, and each time it comes back. The register is how an organisation notices that it is going in circles.

Corrective vs preventive — the distinction everyone gets wrong

The two words sound interchangeable and they are not. A corrective action responds to a problem that has already happened: a customer complained, an audit found a gap, a batch failed. The trigger is an event. A preventive action responds to a risk — nothing has gone wrong yet, but data, a trend, a near miss elsewhere or a hard look at your own process says it will. The trigger is foresight.

Crisp examples: a customer returns flaking paint and you fix the degreaser bath — corrective. Your scrap trend is creeping up month on month, nothing has failed yet, and you introduce a preventive tool-change interval before it does — preventive. Reading about a competitor's product recall and checking whether the same failure could happen in your process is preventive. Fixing your process after your own recall is corrective, and it is also several years too late to be proud of.

The ratio between them is one of the most honest health signals a quality system has. A register that is 100% corrective belongs to an organisation that only ever reacts. It is fire-fighting with paperwork. Even one or two genuine preventive actions per quarter is evidence that somebody is looking ahead — and the corrective : preventive tile on the front screen will tell you bluntly if nobody is.

The CAPA lifecycle this tool models

Each record moves through the same sequence, and the status field follows it:

• Raise — describe the problem or the risk in plain words, date it, give it an owner and a severity. This is the "open" status.
• Contain — for corrective CAPAs, record what you did immediately to protect the customer while the real fix was found. Preventive CAPAs usually have nothing to contain.
• Find the root cause — keep asking why until you reach something you can change. "Operator error" is where the analysis stops, not where it finishes.
• Act — the action plan addresses the cause, not the symptom. While it is being worked the status is "in progress".
• Verify effectiveness — when the action is done, the status becomes "implemented — awaiting effectiveness check". The CAPA is not closed. It waits, deliberately, until the check falls due and is carried out.
• Close — "verified effective" if the check passed. If it failed, close it as "ineffective" and raise a new CAPA that goes after the cause the first one missed. The sample data contains exactly such a pair.

The tool enforces the two transitions people try to shortcut: it will not let a CAPA be marked implemented without a root cause and an action recorded, and it will not let one close as effective without an effectiveness check method.

Containment, correction, corrective action — three different things

These get merged in everyday speech and an auditor will unmerge them for you, so it is worth being precise:

• Containment stops the problem reaching the customer right now: quarantine the batch, recall the pallet, put a second person on the check. It buys time and fixes nothing.
• Correction fixes the instance: rework the parts, relabel the pallet, reissue the invoice. The defective thing is now good; the process that produced it is unchanged.
• Corrective action changes the process so the problem cannot easily happen again: fix the leak rather than mop the floor, interlock the step rather than remind people about it.

A register full of containment and correction but empty of corrective action is the signature of a shop that is permanently busy and never improving. If your action plan reads like your containment did, the analysis is not finished.

Effectiveness checks — the step everyone skips

Most CAPA registers die at the same point: the action is done, everyone is relieved, and the record is closed. Whether the action worked is never established. Six months later the problem returns and the register says it was fixed.

A good effectiveness check has three properties. It is measurable — a count, a rate, a test result, not "seems better". It is time-bound — "over the next 60 days", long enough for the problem to recur if it is going to. And it uses the same signal that raised the CAPA — if a complaint raised it, complaints verify it; if a scrap trend raised it, the scrap rate verifies it. "Team briefed and everyone agrees it is fixed" fails all three tests.

This is why "implemented" is a separate status here rather than a synonym for closed. The effectiveness checks table lists everything waiting for verification with its method and due date, and flags the checks that are themselves overdue — a quiet number worth watching, because an overdue check is a fix nobody has proved.

How the aging table drives review meetings

The aging table buckets every open CAPA by days since it was raised: 0–30, 31–60, 61–90 and over 90, split corrective against preventive, with the oldest reference named in each bucket. It is built to run a review meeting from, in order: start with the over-90 bucket and make each item justify its continued existence — is it genuinely hard, or merely unowned? Then the overdue actions, then the overdue effectiveness checks. A meeting run this way takes twenty minutes and touches everything that matters; a meeting run through the register in date order takes two hours and touches nothing twice.

Old CAPAs are not automatically a scandal — qualifying a second supplier legitimately takes months. But an old CAPA whose status has not moved and whose owner cannot say what happens next is not an action, it is a decoration, and the oldest-reference column exists so it gets asked about by name.

Formulas

The effective rate divides only by closed CAPAs — open ones have not had the chance to succeed or fail yet, so they belong in neither the numerator nor the denominator. With very few closures the rate will swing wildly; treat it as meaningful once you have perhaps ten closed records, not before. The average days open covers open items only, so closing your oldest CAPAs makes it fall — which is the behaviour you want a metric to reward.

FAQ

Is a CAPA the same as a non-conformance? No. A non-conformance is the event — the thing that failed to meet a requirement. A CAPA is the response to it. Not every non-conformance needs a CAPA: a one-off slip with a clear correction can be recorded and closed as just that. Raise a CAPA when the problem is significant, repeating, or systemic.

Should every complaint become a CAPA? If it did, the register would drown and nothing would get real analysis. Filter by severity and repetition: a critical or repeating complaint, yes; a one-off minor irritation, usually no. The decision itself is worth recording somewhere.

What if the effectiveness check fails? Close the CAPA as "closed ineffective" and raise a new one. Do not quietly reopen and re-edit the old record — the failed attempt is valuable evidence, and the new CAPA should ask why the first analysis missed the real cause. The sample data shows this pattern: a "remind the operators" fix fails its check and the replacement goes after the workstation layout that made the error easy.

Who should own a CAPA? Someone with the authority to change the process it touches — not automatically the quality manager. A register where quality owns everything is a register the rest of the organisation has opted out of.

Do preventive actions need effectiveness checks too? Yes. "Risk removed" is a claim like any other. If you qualified a second supplier, the check is a live order actually fulfilled by them — not the approval paperwork.

Saving your work

CAPAs, 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 and includes every filtered record, not only those drawn on screen. Reset asks twice, then erases everything this tool has stored. There is no undo.

Accuracy & disclaimer

This tool records what you enter and calculates from it. It cannot tell whether a root cause is the real one, whether an action addresses it, or whether an effectiveness check was honestly carried out — and the value of every figure on the report depends on all three. It is a register and a calculation aid, not a certified quality management system and not a guarantee of conformity with any standard.

What a quality standard requires — how quickly CAPAs must be processed, what records must be kept and for how long, and what an assessor will accept as verification — depends on the standard, the industry and the certification body. Auditors judge the quality of the analysis, not the neatness of the log. Establish what applies to you and treat this register as the working record behind it.