Who certifies the number a claim rests on

A commercial laboratory sells a figure that other people's disclosures depend on, and an entire international structure exists to make that figure mean the same thing everywhere.

Pending review

A precision balance with its glass door slid open and a shallow dish of pale powder centred on the pan, a spatula lying alongside. — Illustrative artwork: a laboratory balance of the kind used to certify analytical results. Not a facility, equipment or material connected to this project. · Illustration · Osmond Hub

The short version

Among the firms selling know-how to this industry, the commercial laboratory is the one whose output travels furthest: an assay result becomes a public disclosure, and other people act on it. The structure holding that up is a standard, ISO/IEC 17025:2017, 'General requirements for the competence of testing and calibration laboratories'(opens in a new tab), whose stated scope is the general requirements for the competence, impartiality and consistent operation of laboratories(opens in a new tab). Accreditation against it is not a quality badge. It is the mechanism by which a number obtained in one country is accepted in another.

Competence is a thing somebody assesses

The word doing the work is competence, and in this system it has a technical meaning. UK government guidance defines accreditation as ensuring that those who carry out activities such as testing, calibration, certification and inspection 'are competent to do so'(opens in a new tab).

That is different from certification, and conflating the two is the commonest error in this area. An accreditation body puts it plainly: accreditation 'represents a formal and authoritative recognition that an organisation is technically competent to perform specific activities most commonly in the areas of testing, calibration, inspection, and related technical services'(opens in a new tab), while certification 'is a formal confirmation that an organisation, product, process, or individual meets the requirements of a specific standard'(opens in a new tab). One is a judgement about capability; the other is a confirmation of conformity to a standard's requirements. A laboratory can be certified to a management-system standard and still not be accredited to perform a particular test.

Why a result crosses borders at all

The reason accreditation matters commercially is that its recognition is reciprocal by arrangement. ILAC describes itself as the international organisation for accreditation bodies operating in accordance with ISO/IEC 17011 and involved in the accreditation of conformity assessment bodies including calibration and testing laboratories using ISO/IEC 17025(opens in a new tab).

The arrangement's own text is the load-bearing part. Each signatory recognises the accreditation of a conformity assessment body by other signatories 'as being equivalent to an accreditation by its own accreditation body'(opens in a new tab), and accepts endorsed certificates or reports issued by bodies accredited by other signatories 'on the same basis as it accepts endorsed certificates or reports issued by its own accredited CABs'(opens in a new tab).

Equivalence, declared in advance. That is why a sample taken in one jurisdiction, assayed in a laboratory in a second, can support a disclosure filed on an exchange in a third without anyone re-running the analysis.

The laboratory does not sell a measurement. It sells a measurement that somebody else's regulator will accept without repeating it.

What the reporting codes ask of it

The demand side of this is written into the mining reporting codes, which require that the analytical work be described rather than assumed. The JORC Code's disclosure table asks a reporting company to include reference to measures taken to ensure sample representivity and the appropriate calibration of any measurement tools or systems used(opens in a new tab) — calibration, named explicitly, in a code about public reporting rather than about laboratory practice.

The practical content of that obligation is set out more fully in professional-institution guidance. One such document states that quality assurance and quality control programmes 'should be planned and implemented as integral components of an exploration program', and 'should include submission of external blanks, certified reference materials, and duplicate samples, and regular check sampling by a third-party analytical laboratory'(opens in a new tab). Note the composition: something known to be barren, something of known value, a repeat of a real sample, and a second laboratory. Each is inserted to catch a different failure — contamination, drift, imprecision, and systematic bias in the primary laboratory.

How often they should be inserted is left to judgement: the same guidance says only that they should appear frequently enough to provide statistical confidence in the results(opens in a new tab), and gives no rate. Sample security is treated as part of the same system, with the guidance stating that the security of samples from acquisition to analysis 'is a vital component of the sampling and analysis process'(opens in a new tab).

One caveat about this article's own sourcing, in keeping with the subject. The JORC criterion rows dealing specifically with sub-sampling and with assay quality could not be retrieved from the code's published PDF across repeated attempts, and nothing is quoted from them here. The substance above comes from the sampling criterion, which was retrieved, and from institution guidance that was.

The other knowledge businesses, briefly

Laboratories are the clearest case because their product is a number, but the category is broader — process licensors, consultancies, software vendors — and their business models differ in ways that matter to adoption. On licensing, an intergovernmental training manual describes royalty income as coming usually and primarily 'in the form of an ongoing incremental income stream on the basis of the success of the product in the marketplace'(opens in a new tab), with a lump-sum, fully paid-up licence as the alternative structure.

What a process licensor guarantees — recovery, yield, reagent consumption — is the question that follows, and no citable source for it was obtained. It is named here as a gap rather than filled, which is the same discipline the rest of this article is about.

Related

Equipment Suppliers — the physical-plant side of the same market
Research Organisations — where methods originate before anyone licenses them
Mining Companies — the disclosures these numbers end up inside