Calibration and Conformity Assessment: Where Does Calibration Really Belong in QI?

What sounds like a technical question for specialists is, at its core, a question about the architecture of quality infrastructure (QI) - and about who gets to define it. Does calibration belong to conformity assessment? Or is it something fundamentally different? And why does the answer matter for how we build and communicate QI in developing countries?

I raised this question with two colleagues with deep expertise in metrology and QI. Their responses surprised me - not because they disagreed, but because they reframed the question entirely. And then a primary source landed on my desk that crystallised the paradox better than any theoretical argument could.

A document that settles the institutional question - and opens the conceptual one

Accreditation certificates – like those issued by the accreditation body DAkkS state, in its standard opening paragraph:

"The calibration laboratory fulfils the requirements of DIN EN ISO/IEC 17025:2018 in order to carry out the conformity assessment activities listed in this annex."

This is the standard wording in all DAkkS accreditation certificates for calibration laboratories. Germany's national accreditation body - the institution that formally recognises laboratory competence in Germany, and whose certificates carry legal weight under the European accreditation regulation - officially designates the activities of accredited calibration laboratories as “conformity assessment activities”.

At the same time, the PTB–World Bank Toolkit Ensuring Quality to Gain Access to Global Markets, one of the most widely used QI reference documents in international development, states with equal clarity:

"Calibration is considered part of metrology and not as conformity assessment."

Both statements come from authoritative sources. Both are meant seriously. They cannot both be right in the same sense - which means they are operating on different levels of reality. Understanding those levels is the point of this post.

Two definitions that anchor the debate

Before going further, two terms need fixing:

Calibration (VIM 2.39) is the "operation that, under specified conditions, in a first step, establishes a relation between the quantity values with measurement uncertainties provided by measurement standards and corresponding indications with associated measurement uncertainties, and, in a second step, uses this information to establish a relation for obtaining a measurement result from an indication" (BIPM/JCGM 2008/2012).

The result is a measured value with measurement uncertainty. No pass or fail. No conformity judgement.

Verification of a measuring instrument (VIML 2.09), by contrast, is a "conformity assessment procedure (other than type evaluation) which results in the affixing of a verification mark and/or issuing of a verification certificate."

This is the German Eichung - a legally regulated act with binary outcome and legal consequence. Calibration and verification are not the same thing, even if calibration is often the technical precondition for verification.

The conceptual view: calibration is metrology, not conformity assessment

Clemens Sanetra, a long-standing QI expert with experience at PTB and in international development cooperation, is unequivocal: The PTB–World Bank Toolkit statement is correct, and the conceptual logic is straightforward.

Conformity assessment is the evaluation of an actual state against a target state - typically a technical standard or regulation. The result is binary: compliance yes or no. This applies to accreditation, certification, inspection, testing, and verification (Eichung) alike.

Calibration produces something different: a measured value with associated measurement uncertainty. What the customer does with that result - whether they consider the instrument fit for purpose, adjust it, or replace it - is their decision. The calibration laboratory provides a metrological service. It does not issue a conformity judgement. Even where a calibration certificate includes an additional remark indicating whether a result falls within a particular accuracy class, that remark is additional information for the instrument owner - not a formal conformity finding with legal weight.

Critically: legally regulated measuring instruments are subject to verification (Eichung), not calibration. In the regulated domain – which Sanetra calls the Regulatory Infrastructure (RI) - the legal instrument is the verification certificate (Eichschein). To the extent that calibration underpins verification (as it does in OIML Recommendation 111 for weights), it is a technical input to the conformity assessment - not the conformity assessment itself. Sanetra knows of no statutory requirement that a calibration certificate must include a conformity statement; where such a statement appears, it has no legal force in the RI. It is useful information for the instrument owner, and it may matter for quality management systems - but it is not Eichung.

Why the DAkkS certificate says what it says

This makes the DAkkS wording even more interesting. If calibration is conceptually not conformity assessment, why does Germany's national accreditation body say explicitly that accredited calibration laboratories carry out conformity assessment activities?

The answer lies in the institutional framework within which accreditation operates. The accreditation system - built on the ISO/CASCO framework (ISO/IEC 17011, ISO/IEC 17025, and the ISO 17000 series) - is, by design and by mandate, a system for recognising conformity assessment bodies. The purpose of accreditation is to formally attest that a body is competent to carry out conformity assessment. When a calibration laboratory seeks DAkkS accreditation, it enters this framework and is assessed and recognised as a conformity assessment body. The certificate reflects that institutional framing - inevitably, because that is the only framing the accreditation system has.

This is not confusion or error on the DAkkS's part. It is the logical consequence of calibration laboratories participating in a regime that was architecturally designed around conformity assessment. What the certificate language attests to is the laboratory's competence to carry out conformity assessment activities based on its calibrations - meaning: when a customer requires a conformity statement, the laboratory is competent to provide one properly, with agreed and documented decision rules and appropriate treatment of measurement uncertainty, as ISO/IEC 17025 section 7.8.4.1(e) requires.

But - and this is Sanetra's core point, reinforced by the very existence of this paradox - the fact that calibration laboratories can and do operate within the conformity assessment framework does not mean that calibration as a metrological act is conformity assessment. The same laboratory, the same procedure, and the same calibration certificate may or may not contain a conformity statement, depending entirely on what the customer needs. The metrological work is identical either way.

The practical view: the customer decides what calibration becomes

Anett Matbadal, drawing on years of working in an accredited calibration laboratory and working with PTB's international cooperation activities, does not dispute the conceptual argument - but she points to what happens in practice, and why the distinction is harder to maintain than it appears.

Her key observation is that ISO/IEC 17025:2017 already provides for statements of conformity in calibration certificates - not as a universal requirement, but as a conditional one. Section 7.8.4.1(e) states that calibration certificates shall include, "where relevant, a statement of conformity with requirements or specifications." Where such a statement is included, the standard requires a documented decision rule that takes measurement uncertainty into account.

This conditional provision reflects a real gradient of customer sophistication. The International Bureau of Weights and Measures (French: Bureau international des poids et mesures, BIPM) and most National Metrology Institutes serve clients - typically other NMIs or primary research institutions - who are fully equipped to interpret a measured value and uncertainty and draw their own conformity conclusions. Many customers of commercial calibration laboratories are not even aware. Their quality management systems, contracts, or purchaser specifications require a third-party statement on whether the instrument is within specification. The specification may be a manufacturer's datasheet, an industry norm, a customer's internal requirement, or a normatively defined accuracy class.

A concrete illustration from Matbadal: OIML R 111 (for weights used as mass standards) prescribes a decision rule - measurement uncertainty U may not exceed one-third of the maximum permissible error - that enables a meaningful conformity statement to be derived from a calibration result. An accredited laboratory documents in the certificate whether the weight meets the requirements of a particular accuracy class. This is a data-based evaluation, not an opinion. Notably, OIML R 111 was developed by the legal metrology community, yet it is routinely applied in commercial calibration practice - precisely because calibration is the metrological foundation on which conformity evaluations rest.

Matbadal's broader observation is worth stating sharply: conformity assessment is not limited to the regulated domain. Buyers impose technical requirements through contracts, procurement specifications, and quality agreements. ISO/IEC 17025 draws a subtle but significant distinction between test reports ("where necessary for the interpretation of results") and calibration certificates ("where relevant") for conformity statements - a difference that merits examination. If testing is widely recognised as a conformity assessment activity, what is the principled basis for treating calibration differently?

Three areas of applications

Putting both perspectives together, a cleaner picture emerges — not a binary answer, but a map of three areas in which calibration and metrological services operate, serving different end-users with different needs:

• Regulated area: Technical regulations define mandatory compliance requirements. Legal metrology instruments are subject to Eichung (verification). The calibration that underlies verification is a technical prerequisite - indispensable, but not itself the conformity assessment act. The legal document is the verification certificate, not the calibration certificate.

• Non-regulated, market-driven domain: Buyers require conformity demonstrations through contracts and quality agreements. Calibration laboratories are asked to include conformity statements in their certificates against customer-defined specifications. This is legitimate, normatively grounded in ISO/IEC 17025, and commercially significant. The DAkkS accreditation certificate precisely reflects this use case: the calibration laboratory is recognised as competent to provide this service.

• Science, innovation, and metrology (without conformity assessment): Research, process optimisation, product development, prototype analysis, and competitive benchmarking all require reliable measurements. The calibration result - measured value with accompanying measurement uncertainty - is the end product, not the input to a binary decision. This is metrology at its purest, and it accounts for a large and economically important share of what calibration laboratories actually do. It is invisible in conformity assessment frameworks - and in most NQI diagrams.

The three areas describe the demand side: what the end-customer of a calibration service ultimately needs it for. They do not describe the supply chain of metrological traceability, which is a separate and equally important dimension.

The traceability chain: the supply side of calibration

From the supply side, calibration must be understood in the context of measurement traceability — the property of a measurement result whereby it can be related to a stated reference (a national or international standard) through an unbroken chain of calibrations, each with documented measurement uncertainty (VIM 2.41). This chain runs vertically through the entire metrological system: from working instruments on the factory floor, calibrated by a commercial calibration laboratory, whose own reference standards are calibrated by a National Metrology Institute (NMI), whose realisations are validated through international key comparisons under the CIPM Mutual Recognition Arrangement — ultimately tracing back to the International System of Units (SI).

Figure 1: Chain of traceability -hierarchy for how measurement standards disseminated in an economy from the SI

The traceability chain therefore functions as metrological infrastructure: it is the supply-side backbone that makes all three demand-side areas possible simultaneously. A single NMI calibration of a reference pressure standard may, downstream, feed into a legally regulated verification process (Area 1), a contractual conformity statement for an automotive supplier (Area 2), and an R&D measurement programme for a new sensor technology (Area 3) — without the NMI ever knowing which. The chain serves all three areas without belonging exclusively to any of them.

This is also one of the strongest arguments against absorbing calibration entirely into the conformity assessment framework: the traceability infrastructure that underpins measurement reliability across science, industry, and regulation cannot be reduced to the downstream uses some of its outputs are eventually put to. To classify it as conformity assessment because some of its customers require conformity statements is, as Sanetra suggests, to confuse the infrastructure with one of its applications.

The real problem: calibration falls through the QI communication gap

This brings us to what Matbadal identifies as the deeper issue, and the one most worth addressing. It is not who is right about a definitional boundary. It is that calibration is consistently under-represented in standard QI visualisations - and that absence has real consequences.

BIPM, NMIs, and commercial calibration laboratories technically perform the same activity: they calibrate, at different levels of accuracy and for different customers. The metrological hierarchy is continuous. But in most NQI diagrams used at government level, calibration either disappears under the heading of "conformity assessment" (see Figure 1) - absorbed and anonymised - or disappears entirely, not being mentioned under metrology as an explicit service. The DAkkS framing, paradoxically, may accelerate this disappearance: once calibration laboratories are institutionally categorised as conformity assessment bodies, their distinct metrological identity becomes harder to see.

Figure 2: Elements of quality infrastructure

Source: BMWE Infographics – (retrieved 30/03/2026).

The practical consequence is significant. Policymakers, donors, and development practitioners who encounter QI primarily through the conformity assessment lens often cannot distinguish calibration from testing, verification, or inspection. They fail to invest in calibration infrastructure as a distinct and foundational layer. Countries building their NQI frequently prioritise accreditation structures, certification bodies, and market surveillance - the visible conformity assessment apparatus - while neglecting the routine calibration services that underpin measurement reliability across science, industry, and regulation alike.

An invitation to the QI4D community

The DAkkS document shows that this debate is not purely academic: it has been resolved, at the institutional level, in favour of conformity assessment - at least for accredited laboratories. But that institutional resolution does not answer the conceptual or practical questions.

We would like to hear from practitioners:

• In your advisory experience, how do you explain the role of calibration to government stakeholders who see QI primarily through the lens of conformity assessment?

• Are there countries where the institutional absorption of calibration into the conformity assessment framework has had measurable consequences - positive or negative - for the development of metrological capacity?

• And if testing is widely recognised as a conformity assessment activity, what would it take for calibration to receive equivalent recognition without losing its distinct metrological identity and its role in the non-regulated economy?

These are not abstract questions. They shape how countries invest in QI, how donors frame technical assistance, and how practitioners communicate the value of measurement to decision-makers who will never read the Vocabulary of Metrology (VIM).

References

DAkkS (2024). Anlage zur Akkreditierungsurkunde D-K-12094-02-00 nach DIN EN ISO/IEC 17025:2018 - Horiba Europe GmbH, Oberursel. Gültig ab: 12.12.2024. Deutsche Akkreditierungsstelle, Berlin. (retrieved 30/03/2026).

BIPM/JCGM (2008/2012). International Vocabulary of Metrology - Basic and General Concepts and Associated Terms (VIM), 3rd edition. Joint Committee for Guides in Metrology.

Kellermann, M. (2019). Ensuring Quality to Gain Access to Global Markets: A Reform Toolkit. World Bank and PTB, Washington DC. DOI: 10.1596/978-1-4648-1372-6.

Metrology Asia Pacific (2024). Metrology infrastructure basics, Metrology - Enabling Developing Economies in Asia (MEDEA) Project, APLMF / APMP, (retrieved 05/04/2026).

ISO (2017). ISO/IEC 17025:2017 - General Requirements for the Competence of Testing and Calibration Laboratories, Edition 3. Geneva: ISO.

OIML (2004/2025). International Recommendation R 111-1 - Weights of Classes E1, E2, F1, F2, M1, M1–2, M2, M2–3 and M3, Edition 2004 (E). Paris: OIML.

OIML (2022). International Vocabulary of Legal Metrology (VIML), Edition 2022 (E/F). Paris: OIML.

Note: This blog post grew out of a collegial exchange with Dr.-Ing. Clemens Sanetra and Anett Matbadal, to whom the author is grateful for their expert input and critical reading.

Blog image: Torque standard measurement facility operated by the PTB. It’s the only one of its kind in the world and, for the first time, enables the measurement of very high torques.