Reference10 · 02 · 20266 min read

Lot Numbers: What A Traceable Record Has To Contain

A lot number identifies a defined quantity of material produced under one set of conditions, so that every test result, record and label relating to that quantity can be found again. Traceability is the property of the record system rather than of the code: a lot number is useful exactly to the degree that the documents behind it exist and match. A printed code with nothing retrievable behind it is decoration.

The formal definitions are narrow and worth using precisely. A batch is a quantity of material produced in a single process such that it is expected to be homogeneous. A lot is a batch, or a defined portion of a batch, that is uniform in character and quality. A lot number, also written as a batch number, is the distinctive combination of numbers or letters that identifies that quantity and permits its history to be determined.

Nothing in those definitions requires the code to be meaningful. Some producers encode date and line into it, for example 20260814-A for a lot completed on 14 August 2026; others issue sequential integers. Both satisfy the requirement equally, because the requirement is retrievability rather than legibility. Reading a date out of a lot code is a convenience for whoever printed it and an assumption for everybody else.

What has to exist is the chain. Every test result should name the lot it was measured on, every container label should carry the lot it holds, and every step where material was divided, repacked or relabelled should record which incoming lot became which outgoing lot.

What the record behind a lot number has to contain

  • The identity of the material, as a name and a sequence or formula, not a trade shorthand.
  • The quantity produced, in grams or in units, with the fill quantity per container stated separately.
  • The date the lot was completed, independent of whatever the code encodes.
  • Every analytical result measured on that lot, with the method, the instrument conditions and the acceptance criterion applied.
  • The identity of the incoming lots consumed, so a result can be traced backwards as well as forwards.
  • A retained sample of the lot, held under defined conditions, with the quantity retained recorded.
  • The disposition decision, and who made it.

Record retention periods are set by the framework a producer works under rather than by preference. ICH Q7 sets retention of batch records for at least one year past the expiry date assigned to the batch, and for material with a retest date rather than an expiry date, at least three years after the batch is fully distributed. Retained samples are held for comparable periods and in quantities at least twice what a full specification test would consume.

IdentifierIdentifiesChanges when
CAS registry numberA chemical substance, across all producersNever, for a given substance
Catalogue or item numberA product listing, including its fill quantityThe listing changes
Synthesis or bulk lot numberOne quantity of material from one production runEach production run
Packaging lot numberOne filling run from one bulk lotEach filling run, which may be several per bulk lot
The four codes that appear on peptide packaging, and what each one actually identifies.

The distinction in the last two rows is where most confusion sits. A CAS registry number identifies the substance and is shared by every producer of it, so it is not an identifier of material in the sense a lot number is.

Forward and backward traceability

Traceability runs in two directions, and a system can be good at one and useless at the other. Backward traceability answers what went into this lot: which bulk material, which vials, which stoppers, which analytical reference standard. Forward traceability answers where this lot went: which packaging lots were filled from it, which containers were shipped, to which consignees.

ISO 9001:2015 clause 8.5.2 states the requirement as identification and traceability: outputs are to be identified where necessary to ensure conformity, and the unique identification of outputs is to be controlled when traceability is a requirement, with documented information retained. The clause says nothing about code format, which is the point.

The failure mode: the chain breaking at repackaging

The specific thing that goes wrong is a repackaging step that issues a new lot code without recording the incoming one. Bulk material arrives under one lot number, is divided across several hundred vials, and the vials are labelled with a fresh code generated at the filling step. If the record linking the two codes was not written, the certificate of analysis measured on the bulk can no longer be matched to any particular vial.

It shows up as a certificate whose lot number does not appear anywhere on the container, or a container whose lot number appears on no certificate. Both are visible from the outside, and both are worth resolving before the material is used for anything, because the analytical record and the physical material have been separated and cannot be rejoined after the fact.

A second, quieter version is a pooled fill: two bulk lots combined into one packaging lot to make up a shortfall. The packaging lot then has two parents, and unless both are recorded, every purity figure attached to it is attached to an average nobody measured.

What Aurum's lot codes refer to, precisely

The lot code printed on a vial label identifies the packaging lot: one filling run, from one bulk lot, on one date. The analytical results we publish were measured on the bulk material before filling, not on the individual vial in hand. Test vials individually is not among the specifications Aurum publishes, and no purity figure we publish was measured on the specific container it accompanies.

That is the ordinary arrangement across the category, and it is still a limit worth stating rather than leaving to be discovered. It means a published result describes a population the container was drawn from, on the assumption the bulk was homogeneous and the filling run did not change it. We also do not perform or claim sterility or endotoxin testing on any lot, so no lot record of ours contains those results.

Label durability matters here too, because a lot code that has abraded off a vial has taken the traceability with it.

Common questions

Can a date be read out of a lot number?

Sometimes, and never reliably. Encoding a date is a convention a producer may or may not follow, and two producers using similar-looking eight-digit codes may mean entirely different things by them. The date belongs on the record, where it is stated rather than inferred.

Is a catalogue number a lot number?

No. A catalogue number identifies a listing and stays constant across every lot ever sold under it. Two containers with the same catalogue number and different lot numbers came from different material.

How many certificates should one lot have?

One per set of tests performed on it, each naming the lot and the method. A certificate that names no lot number cannot be tied to material and is not a record of anything.

What does a retained sample accomplish?

It allows a result to be checked later against the same material rather than against a fresh production run. Without retention, a question raised six months after shipping has nothing left to test.

References

  1. 01International Council for Harmonisation Q7: Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients. ICH, 2000.
  2. 02International Council for Harmonisation Q6A: Specifications - Test Procedures and Acceptance Criteria for New Drug Substances and New Drug Products. ICH, 1999.
  3. 03International Council for Harmonisation Q10: Pharmaceutical Quality System. ICH, 2008.
  4. 04International Organization for Standardization ISO 9001:2015 Quality management systems - Requirements. ISO, 2015.
  5. 05United States Pharmacopeia General Chapter <1163> Quality Assurance in Pharmaceutical Compounding. USP-NF.
  6. 06European Directorate for the Quality of Medicines General Notices, Section 1.1: General Statements. European Pharmacopoeia.

Every citation links out to the paper on PubMed. Identifiers are omitted deliberately rather than reproduced from memory, so where we do not hold a verified PMID or DOI the link is a PubMed search for that exact title — it resolves to the paper without anything being invented.

FOR RESEARCH USE ONLY · NOT INTENDED FOR HUMAN CONSUMPTION. This article describes compounds and the research literature in which they appear. Nothing here is a recommendation, protocol, or statement of effect.

More from the journal

Keep reading.