Fable
Two ledgers
Factory Director Tong ran a food processing plant. Every day the quality department filled out forms: raw material temperature, disinfection duration, workshop humidity, pest and rodent control. Each form was signed and stamped, and bound into volumes at month-end.
The district supervisory office came once a quarter, flipped through the booklet, ticked the boxes and left. Three years passed without incident, and Factory Director Tong assumed this record system was already reliable enough.
In July a batch of canned goods had a problem: consumers found foreign objects in them. The investigation team had to trace back: which production line, which shift, which batch of raw materials, which step went wrong.
The quality department flipped through forms all morning and saw only 'temperature up to standard' and 'cleaning completed.' There were no specific values, no machine numbers, no operation times. The investigation team asked why there were only conclusions. Factory Director Tong said the supervisory office only required conclusions to be written.
Unable to narrow the scope, they could only recall products from all production lines and all shifts that day. The recall cost was dozens of times the cost of finding the problem. Director Tong then realized that he had mixed the forms for passing inspections and the process records for accountability into one ledger.
He redesigned two sets of records. The external inspection book continued to record metrics and conclusions; the internal traceability book recorded equipment parameters, employee numbers, exact times, raw material batches, and production line bindings, and could not be changed or deleted at will. When an anomaly occurred again half a year later, the problem was traced to a four-second fluctuation in one wok, and the recall scope was much smaller.
When the new system was first introduced, the quality department found it troublesome. Director Tong had them look back at that recall: it was not that no one filled out the forms, but that the forms only proved "it looks compliant" and could not help the company find the real breakpoints.
Later, external inspections still looked at the scorecard, while internal investigations pulled the process ledger. The two accounts served different questions and also bore different requirements. The factory finally understood that records are not better the more there are; rather, when questions arise, you must be able to produce materials that can answer them. Later, when newcomers started training, the old master no longer began with abstract terms. Instead, he had them review that earlier failure: where intuition had been relied on, why the first fix was insufficient, and how the new arrangement kept risk in a place that could be checked. Only when they could explain these details clearly, and point out when to keep using the approach and when to upgrade it, were they considered to truly understand the method. Later, they also used this method to drill newcomers: first have the newcomer do it the old way and see how errors arise; then redo it under the new rules and see which step was blocked in advance and which step left evidence. If a newcomer could only recite names, they would still misuse it in similar scenarios; only those who could state the applicable boundaries, the costs, and the remedies after failure were allowed to take on work independently.