Why does the supplier scorecard change nobody's decision?

Supplier evaluation runs on schedule in most quality departments: the quarter ends, scores are calculated, the scorecard is filed. Then purchasing places the same order with the same supplier. Nobody broke a rule — because there is no rule the score is attached to.

In practice, a few reasons for this keep coming back.

A scorecard is a decision, not a report

The difference between “the evaluation was done” and “the evaluation changed something” is whether the score is bound to a status. If a supplier whose class dropped keeps its exact position on the approved list, the scorecard is an opinion piece.

That binding has to be written down in advance: which class triggers which status, who approves it, and in which record the decision becomes visible. In BellaKYS the scorecard changes the supplier’s status the moment it is approved — the lowest class moves to blocked, the one above it to conditional, and the reason recorded for the change carries the period and the score. The decision is not something remembered later; it is the scorecard’s own output.

When the source of a score is invisible, the argument moves to the score

If “this score is unfair” comes up in a supplier meeting, what is being discussed is not the supplier’s performance but your calculation method. That argument does not close as long as there is no raw data underneath the score.

Raw data is the one line that shows where the score came from: how many of the shipments arrived on time, how many rejects out of how many pieces, how many of the certificates are still valid. The score is a summary of that line; when the line disappears only the summary remains, and a summary cannot be argued with.

This is also why manual intervention should not be forbidden but traceable. When a criterion score is overridden, the automatically calculated value is not deleted; it stays beside the new one, and whoever changed it writes a reason. Once a scorecard is approved the scores no longer move — if something must change, a new period is opened. Otherwise the scorecard you sent the supplier and the one in the system drift apart.

An unmeasured criterion is not a zero

Some scorecard criteria are computed from data; others are human judgement. When the second kind is left empty it pulls the total down silently: the system does not say “this criterion was not assessed”, it says zero.

That has two consequences. First, a supplier drops into a class it did not earn and trust in the scorecard goes with it. Second, the reverse happens too — if nobody fills a criterion in, everyone looks alike and the scorecard loses its power to distinguish. Checking whether the manually entered criteria were filled in before the period is closed matters more than checking the arithmetic.

No re-approval while a SCAR is open

Re-approving a supplier while a corrective action request (SCAR) raised against it is still open means closing a question before it was answered. The natural order is: the SCAR is answered, root cause and permanent action arrive, verification happens, and only then is approval discussed.

This is a rule the flow has to enforce. “We are careful about it” turns into nothing in a busy week.

A class drop is not the action itself

The “action required” flag the system raises for a supplier whose class dropped is a reminder; a human still opens the action. That distinction is worth accepting, because a record opened automatically is a record nobody owns.

The real question is who sees the drop, and when. If it is invisible to everyone on the day the scorecard is produced, the flag is invisible too. Unless the decision cycle — production, review, approval, status change — is tied to a calendar, the scorecard is an archive document.

---

What makes supplier evaluation useful is not a more detailed scoring table. It is binding the score to a status, the status to a decision, and the decision to a record. Once that chain exists, the scorecard stops being a document you defend in a meeting; it has already made its decision before the meeting starts.

Related pieces

5 August 2026 Why do corrective actions stay open? If your open CAPA count keeps climbing, the cause is usually not a lazy team but a step in the loop that nobody owns. 21 July 2026 An out-of-tolerance device: the real question is how far back you go A gauge that fails calibration gets adjusted or withdrawn; that is the easy part. The hard part is deciding how much of what it measured in the past must now be treated as suspect.

Short answers

Why does the supplier scorecard change nothing?

If the scorecard is produced as a report and filed, it feeds no decision. What separates "an evaluation was done" from "the evaluation changed something" is the decision the score is bound to.

Why do supplier meetings end up arguing about the score?

When the source of the score is invisible, the discussion inevitably moves to the score itself. Once it is visible which record produced which criterion, the conversation returns to performance.

Should an unmeasured criterion score zero?

No. An unmeasured criterion is not zero; criteria calculated from data must be kept apart from those based on human judgement, and missing data must not be treated as a zero.

Can a supplier be re-approved while a SCAR is open?

It should not be. Re-approving a supplier while a corrective action request against them is still open makes the request meaningless.