Technical reference
doesn't make food safe. It the damage when something has gone wrong, by shrinking what must be held or to what is really affected, and by showing what happened. Its quality is set while the work is being done, not on the day of the question.
A wide trace is a statistical problem as much as a record problem: the more batches that share an ingredient lot, the more food one positive result can implicate. Smaller lots, cleaned breaks between runs and a record of each changeover all narrow what has to be held.
Parameters
| Minimum reach | One step back, one step forward; plus the links inside your own business |
|---|---|
| Identity | Lot or date on each container, including part-used and decanted food |
| Test, BRCGS | At least annually, forward and back, with ; 4-hour target |
| Test, SQF | At least annually, across shifts and shared materials |
| Record retention, Codex | Longer than the product's shelf life, or as the authority sets |
| If a link is missing | Treat everything that could be affected as affected |
Records that prove it
- Delivery lots and suppliers
- Lots used in each batch, including rework and packaging
- Who received each batch, when, and how much
- Trace file: the question, the timeline, the conclusion, the answer
- Trace test, with mass balance
CCP or prerequisite?
That depends on the for your process. The number, if it has one, belongs to your plan.