Method Validation in Laboratories: A Practical Guide
Method validation proves that a test method is fit for its purpose; here is how to plan it, run it and defend it.
Talk to a Lakshy ISO/GRC expert
Our ISO/IEC 17025 & laboratories expert will get back to you within one business day.
The short answer
Every report your laboratory issues rests on one quiet assumption: that the method behind it actually works, in your hands, on your samples. Method validation is how you prove that assumption instead of simply believing it.
Why method validation matters to your business
Customers, regulators and tender evaluators rely on your numbers to accept a batch, release a product or settle a dispute. If a method has never been properly challenged, one disputed result can put your whole reputation, and your accreditation, under question. Validation also saves money: it shows early whether a method is fit for purpose, before you invest in reagents, training and reporting formats.
The core idea
Validation answers one question: is this method fit for the job it will be used for? You decide what the job demands, run planned experiments, and compare what you observe with what you set out to achieve.
It helps to separate three situations:
- Standard method used as published: the method is already validated by its owner. You do not repeat that work, but you must verify that your people, equipment and environment can achieve the expected performance.
- Standard method modified, or used outside its stated scope: validate the changes and the extension.
- Method you developed yourself: validate it fully.
What a validation usually examines
Depending on the method, the performance characteristics you look at typically include:
- Selectivity: does the method respond only to what you want to measure?
- Working range and linearity.
- Trueness: how close results are to an accepted reference value.
- Precision: repeatability (same analyst, same day) and reproducibility (different analysts, days or instruments).
- Detection and quantification limits, where relevant.
- Robustness: how much small changes in conditions disturb the result.
- Measurement uncertainty that follows from all of the above.
Not every method needs every item. A mechanical test and a trace chemical analysis ask very different questions, so choose characteristics that match the method and the customer’s need, and note why you left any out.
What auditors look for
An assessor will rarely ask for a lecture on statistics. They want to see a clear thread:
- A written plan with acceptance criteria fixed before the experiments.
- Raw data that traces back to calibrated equipment and competent analysts.
- A conclusion signed off by an authorised person, stating the range and conditions in which the method is valid.
- Verification records for every standard method on your scope.
- Evidence that any later change to the method triggered a fresh look at its validity.
Practical steps to apply it
Start with the requirement
Write down what the customer, the product specification or the regulation needs: the analyte or property, the range, the accuracy that matters. Without this, nothing can be called “good enough”.
Plan, run and judge
Choose the performance characteristics, set the criteria, and use suitable reference materials, spiked samples or comparison with another trusted method. Run enough repeat measurements across analysts and days to see real variation. Then judge the data against your criteria including when the answer is “not yet”.
Approve, record and keep it alive
Issue the validated procedure, file the report with raw data, and link it to your quality control checks and proficiency testing so you notice drift later. Revalidate when equipment, reagents, scope or the method itself changes.
Common mistakes and how to avoid them
- Setting criteria after seeing the data. It looks like validation but proves nothing. Fix the criteria first.
- Treating verification as optional. A standard method still needs your own evidence that you can run it.
- Using one analyst on one day. That measures luck, not reproducibility.
- Validating once and forgetting. Change control should always ask, “does this affect validity?”
- Keeping conclusions but losing raw data. If the figures cannot be traced, the validation cannot be defended.
A situation many Indian laboratories will recognise
Picture a testing laboratory that adapts a published method to suit a locally sourced material, because the original sample preparation does not work well for it. The change seems small, so the team simply starts using it. At assessment, the assessor asks how the laboratory knows the adapted method still gives sound results. Without a plan, criteria and data, the laboratory has nothing to show, and the results already reported become a concern. The same change, handled through a short planned validation, would have taken modest effort and produced a clean record. Laboratories seeking NABL accreditation meet this scenario often enough to plan for it.
How Lakshy can support you
An experienced ISO/GRC expert can help you decide which methods need validation and which need verification, shape templates and plans, and prepare your team for the questions an assessor asks. Under Lakshy OS™, Design™ suits laboratories that want to implement with our guidance, while Drive™ suits those that prefer us to carry the work end to end. See our ISO/IEC 17025 laboratory guidance, a quick gap assessment of your current records, or training for your analysts and quality staff. Medical laboratories can also look at ISO 15189.
ISO 17025 in three minutes
Frequently asked questions
Do all test methods need full validation under ISO/IEC 17025?
No. Standard methods used exactly as published need verification, meaning you show your laboratory can achieve the expected performance. Full validation applies to methods you develop, standard methods you modify, and standard methods used beyond their intended scope.
Who accredits laboratories in India?
Laboratory accreditation in India is granted by NABL, the National Accreditation Board for Testing and Calibration Laboratories. A consultant such as Lakshy supports you in preparing your system and records, but it does not accredit or certify laboratories.
When should a validated method be revalidated?
Revalidate whenever something that could affect performance changes: the method steps, key equipment, reagents, scope of use or the sample type. Poor quality control or proficiency testing results are also a signal to take a fresh look.
Plan your next step
Talk to an experienced ISO/GRC expert about validating and verifying the methods on your laboratory's scope.
Related reading:
More from the ISO Certification Blog
New guides every week on ISO standards, audits and certification in India.
Start with a free ISO/IEC 17025 gap assessment.
Thirty minutes with a senior specialist, benchmarking your organisation against every requirement of the standard. No cost. No obligation.
Book My Free Session