Academy › Practical guides › Laboratory accreditation
Method validation vs verification: what clause 7.2 actually expects
A standard method is not self-validating in your lab. Here is the evidence file an assessor asks for.
Updated 13 November 2026 · 8 min read · Written for working labs and quality teams
Short answer
Verification (7.2.1.3) confirms that your lab — with its own people, instruments, reagents and matrix — can properly perform a published standard method and achieve its stated performance. Validation (7.2.2) establishes, with evidence, that a non-standard, lab-developed or modified method is fit for its intended use, across the parameters relevant to that use. Verification is a confirmation exercise; validation is a demonstration exercise.
On this page
Which one applies to your method
The most quoted assessor question is simple: "show me the records that demonstrate you achieved the method's performance with your staff and your equipment, for the samples you actually report." A method SOP and a reagent catalogue are not verification records.
| Situation | What 7.2 expects | Typical evidence |
|---|---|---|
| Published standard method, used exactly as written | Verification before use | Precision, bias/trueness or recovery, and the lab's ability to meet the method's stated performance — with records |
| Standard method used outside its scope, or modified | Validation of the modification | A defined change, plus evidence that performance still meets the intended use |
| Lab-developed method, or a method from a non-standard source | Full validation | Parameter study across the intended matrix, range and concentration levels |
| Method specified by the customer | Verification that the lab can run it; clarification if it is unsuitable or outdated | Contract-review record, plus the performance evidence |
The parameters, and what they are for
- Accuracy / trueness — certified reference materials, recovery studies or comparison with a reference method; state the acceptance criterion before you start.
- Precision — repeatability and, where relevant, intermediate precision across days, operators or instruments. A single afternoon's replicates only demonstrate repeatability.
- Specificity / selectivity — can you distinguish the analyte from matrix interferences? This is where a clean-solvent validation fails a real-sample assessment.
- Linearity and range — calibrate across the range you will report, not just the middle of it; document the acceptance criteria and residuals, not only R².
- LOD / LOQ — determined for the matrix, with the approach stated (blank-based, signal-to-noise, or from the calibration).
- Robustness / ruggedness — deliberate small changes (temperature, pH, reagent lot, operator) to show the method tolerates normal variation.
- Measurement uncertainty — the validation data often supply the Type A components of the 7.6 budget, so keep the raw data, not just the summary.
Practical note. Validate the method for your scope, not the technique in the abstract. A validation file that demonstrates excellent linearity in pure solvent says nothing about reporting a pesticide in a fatty food matrix at the same limit. Assessors compare the validation range and matrices against your accredited scope — a mismatch is a finding.
The readiness file an assessor can trace in ten minutes
- Method version and source, with a documented review that it is the current edition.
- Intended use statement: analytes, matrices, concentration/measurement range, and the customer or regulatory requirement it serves.
- Verification / validation plan with parameters, acceptance criteria and sample sizes defined before the work.
- Raw data, calculations, deviations and the conclusion against each criterion, signed and dated.
- Personnel competence and authorisation for this method (6.2), plus equipment and traceability status (6.4 / 6.5).
- Link to the uncertainty budget (7.6) and to the routine QC / PT programme (7.7) that keeps the method under control.
- Change-control history: any modification triggers a documented re-verification or re-validation.
Where this goes wrong in practice
Assuming "it is a standard method" means no work is required. 7.2.1.3 still asks for verification of achievement in your own laboratory.
Validating in a clean solvent while reporting against a complex matrix, or validating a range narrower than the accredited scope.
Setting acceptance criteria after seeing the data, or reporting only R² for linearity without residual analysis.
Treating a minor method change (new instrument model, new reagent lot, shortened incubation) as administrative and never re-verifying.
Keeping the summary report but not the raw data, so the uncertainty budget's Type A components cannot be reconstructed.
Frequently asked
Does ISO/IEC 17025:2017 require validation of every method?
No. It requires verification of standard methods before use (7.2.1.3) and validation of non-standard, lab-developed and modified methods (7.2.2). In both cases the extent should be appropriate to the intended use, and records must be retained.
What is the difference between verification and confirmation in a testing lab?
Verification here means confirming your lab can properly operate a specified method and meet its performance. It is not the same as the metrological 'calibration verification' of an instrument, nor the customer's 'verification' of a product. Keep the terms distinct in your procedure.
How many replicates are needed for a validation study?
There is no universal number. The plan should be justified by the intended use, the variability and the acceptance criteria — commonly multiple days, operators and concentration levels so that intermediate precision can be estimated. Sector guidance or the method itself may prescribe a design.
When must a method be re-validated?
When it is modified, extended to a new matrix or range, moved to substantially different equipment, or when QC / PT data indicate performance has changed. Document the trigger in your change-control procedure rather than relying on a fixed calendar.
Turn the study into records
Document accuracy, precision, linearity and LOD/LOQ in one place
Enter the study data for each validation parameter and produce a structured method-validation report with the calculations shown — ready to attach to the method file an assessor will ask for.
Open the free method-validation tool →Runs in your browser. Export the report free. No sign-up needed.
LabKitPro guides are written from published standards and accreditation-body documents — references, clause numbers and paraphrased intent only, never verbatim standard text. They are for preparation and study, not a certification or accreditation decision, and not legal advice. ISO standards are trademarks of ISO; IATF 16949 is published by the International Automotive Task Force and is not an ISO standard. The current laboratory standard is ISO/IEC 17025:2017; there is no 2025 edition. Global ACI replaced IAF and ILAC on 1 January 2026.
Academy by standard · All practical guides · Free browser tools