Batch and Lot Numbers: What They Tie a Certificate To

A batch number, also called a lot number or control number, is a unique identifier assigned to one specific quantity of material. On a certificate of analysis it names the lot that the reported results describe, and it is the field that connects the document to physical material. The certificate describes that one lot and nothing else. Without the identifier, the result is a statement about material that cannot be pointed to.

This is a short field and an easy one to skip. It deserves more attention than the purity figure beside it, because a purity figure with no lot behind it has no subject. What follows is what the terms mean, where the number comes from, what it links back to, and how to check that the lot in your hand is the lot on the paper.

What a batch is

Regulatory frameworks define a batch with some care, because the whole idea of quality control depends on knowing exactly what quantity of material a result refers to.

In the language of pharmaceutical good manufacturing practice, a batch is a specific quantity of material produced in a process or series of processes so that it is expected to be homogeneous within specified limits. The United States drug manufacturing regulations at 21 CFR 210.3 define it in similar terms, as a specific quantity of a drug or other material intended to have uniform character and quality, produced according to a single manufacturing order during the same cycle of manufacture. The ICH Q7 guideline on active pharmaceutical ingredients includes a further point for continuous production: a batch may correspond to a defined fraction of the output, sized either by a fixed quantity or by the amount produced in a fixed time interval.

Three ideas sit inside that definition.

• It is a defined quantity. A batch has edges. Material inside them belongs to it. Material outside them does not.
• It comes from one process, or one defined series of processes. Two separate synthesis runs are two batches, even if the same person ran them with the same reagents on consecutive weeks.
• It is expected to be uniform within limits. The word is expected. Uniformity is a design intention that testing then checks, not something a definition can guarantee.

Batch versus lot

The two words overlap, and the overlap is where most confusion starts. A lot is a batch, or a specific identified portion of a batch, having uniform character and quality within specified limits. A lot can therefore be the whole batch or a subdivision of it.

The usual example is a single synthesis run that is later divided into several containers or filled on different days. The run is one batch. Each filling or portion may be assigned its own lot identifier. Some frameworks treat the two terms as cross-references for the same thing. Others keep them distinct. A reader does not need to resolve this. What matters is that whichever term appears, it must identify one defined quantity of material, and the same identifier must be findable on the product.

This also explains why the identifier is sometimes called a control number. The name emphasises its job. It controls what is being talked about.

What the identifier is for

Regulatory definitions describe a lot number, control number or batch number as a distinctive combination of letters, numbers or symbols from which the complete history of the manufacture, processing, packing, holding and distribution of the batch can be determined. The ICH Q7 glossary gives a closely similar definition: a unique combination of numbers, letters or symbols that identifies a batch and from which the production and distribution history can be determined.

The important point is that the history can be determined from the number. The number is not decoration. It is an index. Its value is that it points to a set of records.

Those records, in a GMP setting, are commonly called batch records. In general terms they document what materials went in, which equipment and procedures were used, what was observed along the way, which tests were run on the result and what the results were, and where the material went afterwards. The exact contents differ between frameworks and products. The principle does not. A batch number is the key that opens that file.

A number that opens nothing is only a label. The identifier earns its meaning from the records behind it.

Where the number comes from

A batch number is assigned by whoever makes, fills or releases the material. It is not assigned by the testing laboratory, and it is not derived from the test. This is a point people often reverse.

The format is the manufacturer's choice. Many encode a date, a site, a product code or a running count. Others are simply sequential. None of these conventions is universal, and the number alone does not tell a reader anything about quality. A longer or more elaborate-looking code is not a better one.

When a sample reaches an analytical laboratory, the laboratory records the identifier supplied with it and reports it back on the certificate. It may also assign its own internal sample reference for tracking inside the laboratory. That reference is separate. The two numbers answer different questions: the batch number says which material the sample came from, and the laboratory reference says which job the sample belongs to.

One consequence is worth stating plainly. A laboratory can only report the identifier it was given. It cannot independently establish that the sample really came from the lot named on the submission form. This is one of the reasons a certificate is read as a description of the sample submitted, and why the connection between sample and lot sits partly outside the laboratory's knowledge.

A certificate describes one lot and nothing else

This is the central limit, and it follows directly from how the identifier works.

A result measured on one lot is a fact about that lot. It is not a fact about the next lot, the previous lot, or the product line in general. Separate batches come from separate runs, and synthesis, purification and drying all introduce run-to-run variation. A lot that tested well gives no guarantee about one made a month later with the same supplier and the same specification.

That is why a certificate dated some months ago, carrying a lot number that does not appear on the current vial, has no bearing on the current vial. It is not that the earlier result was wrong. It is that it was about something else.

The limit also runs in a second direction, inside the lot. A certificate reports on the portion that was submitted. Whether that portion represents the whole lot depends on how it was drawn, and a single unit tested from a large lot is evidence about that unit. We have written separately on single vial versus batch representative testing. The two limits together can be put in one sentence: a certificate describes the sample submitted, and the sample belongs to one lot.

How to check that the lot in your hand matches the paper

The check is simple and takes about a minute. Almost nobody does it.

• Find the lot marking on the product. Look at the vial label and the outer packaging. It is often labelled "Lot", "Batch" or "LOT No.".
• Find the identifier on the certificate. It should be a stated field, not a phrase in a footer.
• Compare them character by character. Letter case, leading zeros and similar-looking characters such as the letter O and the digit 0 are easy to misread. A near match is not a match.
• Check the compound name and the date. The same lot identifier can be reused by a supplier across different compounds, so confirm the compound also matches. Check that the test date is plausible for the age of the lot.
• Confirm the certificate at source. A matching number on a document proves little if the document itself is unverified. Use the laboratory's register, as described in verifying a certificate of analysis, and check the client name as well as the result.

If any of these steps fails, the honest reading is that the certificate has not been shown to describe this material. That does not mean the material is bad. It means the evidence does not reach it.

What a certificate with no lot identifier cannot show

A certificate with no batch or lot identifier cannot be tied to physical material at all. The document may be genuine. The test may have been carried out carefully. But there is nothing on the page that connects the result to any particular container, so it can accompany any stock for as long as anyone chooses to attach it.

Several specific things become impossible.

• No one can match it to a vial. Not the buyer, not the supplier, not the laboratory.
• No one can tell whether it is current. A certificate from an old run and one from a recent run look identical.
• It cannot be checked against records. The index that would lead back to the batch history is missing.
• It cannot be challenged. A statement that can never be matched to anything can never be shown to be wrong.

That last point is the uncomfortable one. A document that cannot be contradicted is not strong evidence. It is unfalsifiable, which is a different property. For the full list of fields and what each absence implies, see how to read a peptide certificate of analysis.

Related cases deserve the same scrutiny. A certificate that lists a lot identifier but the vial carries none cannot be matched. A certificate that names a lot of a raw ingredient, presented as covering a finished product, describes the ingredient. One lot of material can sit behind more than one presentation, and the scope of the certificate should say which it covers.

What a matching lot number does not prove

It is worth ending on the limits, because a matching number can feel more conclusive than it is.

A match shows that the certificate names the same lot as the product. It does not show that the lot is safe, because purity is a measure of composition, not of safety. It does not show that the sample was representative of the lot. It does not show that storage since testing has preserved the material. And it does not establish that the number on the vial was assigned honestly. A lot identifier is a claim made by whoever labelled the product, and a laboratory can only report it.

What a match does give is a defined subject. The result is about this lot, measured on a sample submitted for testing, by the methods stated, on the date given. That is a narrow statement. It is also the only kind of statement a certificate is able to make.

Novagen is a third-party analytical laboratory. It reports what was measured in the sample submitted, and it publishes a verifiable register of the certificates it issues at novagenlaboratories.com. Current quoted turnaround is 21 days from receipt of samples. Questions can be sent to admin@novagenlaboratories.com. To check a certificate against the register, use the verification page.