Certificate of Analysis, Conformance and Origin: Three Different Documents

Three documents travel with materials, all three are abbreviated to something beginning with “certificate”, and only one of them contains measurements. Being able to tell them apart at a glance is most of the value, because the differences are not stylistic — they are differences in what is being asserted and by whom.

Certificate of analysis

A certificate of analysis reports results. Its defining feature is that each attribute appears with a value, the method used to obtain it, and the criterion it was judged against — the three-part structure described in how a specification limit gets set.

It is tied to a specific lot, because the results were obtained from that lot. It is reproducible in principle: another laboratory applying the same method to the same material should land in the same region, within the spread covered in measurement uncertainty. And it is falsifiable, which is what gives it its weight — a claim that can be checked is a different kind of claim from one that cannot.

Certificate of conformance

A certificate of conformance — sometimes conformity, sometimes compliance — asserts that a lot meets its specification. It usually carries the lot number, the specification reference, a statement of conformity and a signature. What it does not carry is data.

That is not deception; it is a different instrument, and there are contexts where it is the appropriate one. A conformance statement transmits a conclusion without disclosing the underlying results, which matters where the method itself is proprietary or where a distributor is passing on an assurance received from upstream.

The practical consequence is simple. A conformance certificate cannot be evaluated. There is no chromatogram to look at, no integration to question in the sense of peak integration, and no value to compare against another lot. It asks to be believed, and its usefulness therefore rests entirely on the credibility of whoever signed it.

Certificate of origin

A certificate of origin states where goods were manufactured. It is a customs and trade document, not a quality document, and it is normally issued or endorsed by a chamber of commerce or a comparable body rather than by a laboratory.

It says nothing about composition, purity, or whether anything was tested. Its relevance to a buyer is duty treatment and import documentation, which sits alongside the considerations in importing research chemicals into Canada. Seeing one presented as evidence of quality is a category error, and occasionally a deliberate one.

The hybrid that causes most of the confusion

Many documents in circulation are headed “Certificate of Analysis” and are structurally conformance statements: a table of attributes, a column of limits, and a column reading “Conforms”, “Pass” or “Complies” where the values should be.

A few entries of this kind are unremarkable — some tests genuinely produce a verdict rather than a number, and a visual appearance check is one of them. A document where every row is a verdict is a conformance certificate with an analysis heading. It is worth noticing, not because it is necessarily wrong, but because the reader who takes it for a data report is drawing conclusions the document never supported.

Who signed it, and on what basis

The issuer matters more than the format. Four situations look similar on paper:

• The testing laboratory issues it. The signer performed the work. This is the strongest position, and the scope question in ISO/IEC 17025 accreditation applies to it.
• The manufacturer issues it from its own testing. Common and legitimate; the incentive structure is discussed in third-party versus in-house testing.
• A distributor reissues an upstream certificate under its own letterhead. Here the chain matters: the lot identity has to connect the document to the container in hand, which is the point of batch and lot numbering.
• A distributor issues a conformance statement based on an upstream certificate it has not verified. This is an assertion about an assertion, and it is the weakest of the four.

What a certificate of any kind never covers

All three documents describe the lot at the moment it was assessed. None of them describe what happened to the container afterwards — the shipping conditions, the temperature excursions covered in cold chain and excursions, the storage since arrival, or the number of times the vial has been opened.

A certificate is a photograph with a date on it. Treating it as a description of the material currently in the freezer is the most common misreading of all three document types, and the one with the most practical consequences.

Reading the three quickly

Look for numbers. If each attribute has a value and a method, it is an analysis. If the attributes have verdicts, it is a conformance statement whatever the heading says. If it names a country and no attributes at all, it is a trade document.

Then look for the lot number, and check that it matches the container. A document that does not connect to the material in front of you is not evidence about that material, regardless of which of the three it is — the connection that chain of custody exists to preserve.