ISO 17025 Method Verification: Requirements, Steps, and Examples

ISO 17025 method verification is the process a laboratory uses to demonstrate, with objective evidence, that it can properly perform an established method before releasing routine customer results. When a lab adopts a standard or previously validated method, it generally does not need to repeat the method’s complete original validation. However, the laboratory must still confirm that the required performance can be achieved using its own personnel, equipment, facilities, reagents, sample matrices, and operating conditions.

Method verification demonstrates that a laboratory can achieve the required performance of an established method under its own operating conditions.

Understanding ISO 17025 method verification vs. validation is essential. Method verification confirms that your laboratory can successfully implement an established method within its intended scope. Method validation is normally required for a nonstandard, laboratory-developed, significantly modified, or out-of-scope method. Confusing these two activities can lead to unnecessary work—or, more importantly, inadequate evidence that a method is suitable for routine use.

The ISO 17025 method verification requirements do not prescribe one identical study for every laboratory method. Instead, the verification should focus on the performance characteristics that matter for the method’s intended use. Depending on the method, this may include trueness or recovery, precision, selectivity, working range, detection capability, quantification capability, or measurement uncertainty. These results should be compared with documented method verification acceptance criteria established before the study begins.

A properly planned verification gives your laboratory confidence that the method performs as expected under everyday conditions. It also provides assessors with clear evidence showing why the method was approved for routine testing or calibration—without repeating validation work that has already been completed by the method developer or standards organization.

Key Takeaways

• Method verification shows your lab can carry out a standard method that’s already been validated elsewhere
• You need verification before you start using any standard method in your lab
• Verification just confirms you meet the method’s performance criteria—it’s not a deep dive like validation

What Is ISO 17025 Method Verification?

Method verification under ISO/IEC 17025:2017 is how you check that a standard test method actually works in your lab. If you’re following a published method exactly, there’s no need for full validation. Instead, you just need to show you can achieve the same performance the original method claims.

Verification is about showing you can reproduce expected results. It’s less data-intensive than validation. You’re not building new performance stats—you’re just checking if what worked for someone else also works for you.

Clause 7.2 of ISO/IEC 17025 says you’ve got to prove methods are fit for what you want to use them for. Verification does that for established methods like EPA or ASTM procedures.

When You Need Verification

• Using a standard method as published
• Following established protocols without any tweaks
• Staying within the method’s original scope

What You Verify

Document these checks:

• Accuracy – Are you getting the right answer?
• Precision – Are your results repeatable?
• Detection limits – Can you measure at the levels you need?
• Recovery rates – Do your samples give expected results?

Validation and verification aren’t the same. With verification, you’re just making sure your people, tools, and setup can hit standards already set by someone else. You’re not proving the method for the first time—just that you can use it as intended.

Keep clear records of your verification. Auditors will look for this during assessments.

Why a Validated Standard Method Still Needs Laboratory Verification

Organizations like AOAC, USP, or ASTM validate their standard methods thoroughly before publishing. That validation proves the method works—at least, it worked for them, under their conditions. But does it automatically work in your lab? Not necessarily.

Lab verification checks that your setup can actually reproduce the published performance. Your instruments, analysts, reagents, and even the air in your lab might be different. All those things can change how the method behaves.

ISO/IEC 17025 says you need to prove you can get the same results the method promises. You need to show your instruments are calibrated, your analysts know what they’re doing, and your lab environment supports accurate testing.

Think of verification as proof that the method transfers to your world. Maybe the method was validated on a different brand of instrument, or with purer water, or in a different climate. Those little differences can matter.

What should you check?

• Precision – Can you get consistent results?
• Accuracy – Do you match expected values with reference materials?
• Calibration linearity – Does your instrument’s response line up with the published curve?
• LOD and LOQ – Are you hitting the detection and quantification limits?

Verification isn’t just a box to tick. It’s your proof that the method works for you, not just for the people who wrote it years ago in a totally different lab.

When Is ISO 17025 Method Verification Required?

Clause 7.2 of ISO/IEC 17025:2017 calls for method verification when you use a standard method that’s already been validated by someone else or published by a standards body.

You’ve got to verify a method before you start using it for routine testing. This covers methods you pull from places like ASTM, EPA, or any recognized group. The whole point is to make sure your lab can run the method correctly with your own tools, people, and setup.

Common times you’ll need verification:

• First time you use a published standard method
• Adopting validated methods from another lab
• Minor tweaks to equipment or instruments
• Switching reagent suppliers or lots
• Training new staff on existing methods
• Moving methods to a new lab location

Any big change to your testing environment—new gear, facility moves, or updates to the procedure—means you should verify the method again.

If you’re using someone else’s validated method, you verify. If you’re building your own or making big changes, that’s when you need full validation.

Your lab has to show, through verification, that you can meet the performance the original method lays out. That’s what keeps your results reliable and your accreditation safe under Clause 7.2.

ISO 17025 Method Verification vs. Method Validation

Verification and validation aren’t interchangeable in ISO 17025. Knowing which to use (and when) keeps you on the right side of accreditation.

Method validation is for new or changed methods—non-standard, lab-developed, or anything you’ve modified. Validation proves a new or altered method actually works for what you want to do.

Set up a validation plan before you start. That plan outlines which parameters (like accuracy, precision, detection limits, measurement uncertainty) you’ll test and how you’ll do it.

Method verification is what you do when you’re using standard, already validated methods. You’re just confirming the method works in your lab, with your equipment and people.

Here’s how they stack up:

Aspect	Validation	Verification
When Used	New, modified, or non-standard methods	Standard published methods
Scope	Extensive testing of all parameters	Limited testing to confirm performance
Documentation	Validation protocol, validation report, detailed validation records	Simpler records showing local performance
Effort	Time-intensive and comprehensive	Faster with fewer tests

Validation techniques and records will depend on the method and its purpose. You’ll pull everything together into a validation report as evidence you’ve met requirements.

Whenever you change things—equipment, sample types, procedures—expect to revisit validation or at least run a re-verification. ISO 17025 method validation keeps your testing on track over time.

A Simple Verification-or-Validation Decision Tree

Not sure if you need validation or verification? Honestly, it can get confusing. A quick decision tree can help.

First: Is this a brand-new or seriously modified method?

If you answered yes, you’re looking at full validation. If not, keep going.

Now: Are you using a method that someone else already validated?

If yes, verification should cover you. That goes for standard methods from EPA, ASTM, or similar groups.

Finally: Are you using the method exactly as written, with similar samples and equipment?

If yes, then a verification study should do the trick. If you’re planning big changes to the process, equipment, or sample types, you probably need validation instead.

Here’s a cheat sheet:

Situation	Required Process
Brand new method you developed	Validation
Modified standard method with significant changes	Validation
Standard method used as written	Verification
Previously validated method in a new lab	Verification
Method applied to different matrix than validated	Validation or extended verification

Your verification study should show that you can hit the key performance marks—like precision and accuracy—using your own lab setup. If you’re unsure, it’s smart to check the original method docs or ask your accreditation body. Better safe than sorry, right?

Method Verification Is Not the Same as Analyst Competency

Method verification shows a test procedure actually works in your lab, with your stuff, your quirks, and your conditions. Analyst competency, on the other hand, is all about whether a person can run that procedure the right way. ISO 17025 treats these as separate boxes to check—you can’t just tick one and call it good for both.

When you verify a method, you’re really sizing up the procedure itself. You look at accuracy, precision, detection limits, and other performance traits. This tells you if the method does what you need it to, in your hands, not just in theory.

Evaluating analyst competency means you’re checking if your staff can actually pull off that method. That includes their technical chops, their understanding of the steps, and their ability to get reliable results, not just once, but consistently.

Here’s where some labs mess up: they think if an analyst joins in method verification, that’s enough to prove competency. Not quite.

ISO 17025 expects you to prove competency with real activities like:

• Initial training and supervised practice
• Directly watching their technique
• Reviewing their actual work
• Written or hands-on exams
• Proficiency testing (PT) or interlaboratory comparisons

Proficiency tests are great for this. When an analyst takes part in a PT or interlab comparison, you see how their results stack up against others using the same method. If they bomb the PT, you’ve got a competency issue, even if your method verification data looks fine.

Method Verification Is Not the Same as Equipment Qualification or Calibration

Method verification and equipment calibration aren’t interchangeable. If you want to keep your ISO 17025 compliance intact, you really need to get the difference.

Calibration is about comparing your device to a standard. It shows you how much your instrument has drifted from where it should be. When you send something out for calibration, the lab traces it back to international standards—think NIST and all that. That’s your traceability.

Method verification is something else entirely. You’re confirming a specific testing method works as it should in your lab. That means checking your procedures deliver accurate, reliable results, not just in theory, but with your actual setup.

Here’s a quick breakdown:

Aspect	Calibration	Method Verification
Focus	Equipment accuracy	Procedure performance
What’s tested	Individual instruments	Complete testing methods
Frequency	Set intervals (often yearly)	When adopting new methods
Output	Calibration certificate	Validation data

Equipment qualification checks if your instruments meet the manufacturer’s specs when you first set them up. Calibration keeps them accurate over time. Method verification? That’s about your whole testing and calibration process working as a system.

Let’s say you calibrate a spectrophotometer so the readings are right. But method verification checks if your entire analytical process—calibration curve, sample prep, analysis steps—delivers correct results. It’s not just about the machine.

Don’t mix these up. Sure, you need calibrated equipment for method verification, but calibration by itself doesn’t prove your methods work. You really need both if you want to check all the ISO 17025 boxes. Calibration looks after your instruments; method verification makes sure your procedures actually give you good data.

Start by Defining the Method’s Intended Use

Before you even think about verifying a method, you’ve got to nail down what you’re going to use it for. This is the foundation—skip it and the rest of your process is shaky at best.

Fitness for purpose starts with knowing your own testing needs. Write down what you’re measuring, the range you expect, and how accurate you need to be. If you’re testing water for lead, for instance, specify the detection range and the precision you need for compliance. Don’t just wing it.

Your choice of method hangs entirely on this definition. Make sure you jot down:

• The analyte or parameter you want to measure
• The sample type (matrix)
• The expected measurement range
• Detection limits you need
• Accuracy and precision you have to hit
• Any regulatory or customer specs

This kind of documentation saves your bacon during audits. Plus, it helps you decide if you need verification or full-blown validation—ISO 17025 treats those differently, depending on how you use the method.

Think about what your clients actually want. A screening method might not cut it for quantitative analysis. The more specific you get now, the easier it’ll be to show the method works when someone asks.

Keep this document handy. You’ll come back to it over and over as you check if the method meets your requirements.

Frequently Asked Questions

Method verification under ISO 17025 brings up a bunch of real-world questions about performance, documentation, and whether you can lean on published data. Here are some answers to the practical choices labs face when they’re trying to make sure established methods actually work for them.

What is the difference between method validation and method verification in an accredited laboratory?

Method validation is what you do when you’re building a new method, tweaking an old one, or using a standard method in a way it wasn’t intended. You’ve got to collect solid experimental data to prove the method works for your purpose.

If you’re adopting a published standard method exactly as written, you’re looking at method verification. Here, you’re just confirming your lab can hit the performance marks the method developer already published. It’s less work than full validation, but you still need formal tests and records.

Bottom line: validation means you own the performance claims. Verification means you prove you can match what someone else already showed.

Which performance characteristics should be evaluated during method verification for routine testing?

Accuracy? You’ve got to check it with recovery studies—use certified reference materials or spike your samples at relevant concentrations. Your recovery should land within the method’s expected range, usually 80-110% for most analytical work.

For precision, you need to look at both repeatability and intermediate precision. Repeatability is how consistent you are in one run, same analyst. Intermediate precision is about variability across days, analysts, or instruments.

Detection limit and limit of quantification need a check, too. Make sure your LOD and LOQ line up with the published values. Your LOQ should be below your spec limits by a comfortable margin.

Specificity and selectivity matter—can your method measure the target analyte without interference from your sample matrix? If your matrix is different from the original developer’s, double-check this one.

For linearity, run calibration standards across your working range and make sure your correlation coefficient meets the published standard—usually r² ≥ 0.999. Range verification is about confirming these performance traits hold across the whole concentration interval you plan to use.

How should acceptance criteria be defined and justified for a method verification study?

Start with the performance characteristics from the original method. If they list specific values for accuracy, precision, linearity, or detection limits, those are your targets for verification.

If the published criteria are missing or vague, base your acceptance criteria on what you actually need. Your LOQ should be lower than your spec limit. Your precision should be tight enough that uncertainty doesn’t make borderline results a guessing game.

For recovery, 80-110% is typical for chemical methods. For precision, you’re usually looking at a relative standard deviation under 5% for instrumental methods, but if your matrix is tricky, you might need to allow more wiggle room.

Document why you chose every criterion. Point to the original method, regulatory guidance, or technical papers. Assessors will absolutely ask why you picked those numbers.

What is the minimum number of replicates, concentration levels, and days needed to demonstrate precision and accuracy?

For repeatability, run at least six replicates in a single batch, under the same conditions. That gives you enough data to calculate a meaningful RSD.

Intermediate precision takes at least three non-consecutive days. Each day, do multiple replicates so you can separate within-day from between-day variability. Some accrediting bodies want you to use at least two analysts for this.

For accuracy, cover at least three concentration levels across your working range—one near the LOQ, one in the middle, and one at the top. Each level needs several replicates for both accuracy and precision calculations.

To check detection limits, analyze at least seven replicates of a low-level sample. Use the standard deviation from those to calculate your LOD and LOQ.

ISO/IEC 17025 does not prescribe one universal verification study, replicate count, or list of performance characteristics for every method. The extent of the work should reflect the method’s intended use, published performance claims, laboratory conditions, matrix, measurement range, and the risk associated with an incorrect result. For more detailed technical guidance on planning verification studies and determining how much work is appropriate, laboratories can consult the Eurachem guidance on method verification and validation.

How should method verification be documented to satisfy accreditation assessment expectations?

You need a written protocol before you start, laying out which performance characteristics you’ll test, your acceptance criteria, and your experimental plan.

Keep all your raw data traceable and complete—instrument printouts, spreadsheets, analyst initials, dates, sample IDs, reference material certificates. Assessors will want to see the original data, not just your tidy summary tables.

Your summary report should show results for each performance characteristic, the acceptance criteria, and a clear pass/fail for each one. Include tables with recovery percentages, precision numbers, linearity coefficients, detection limits—the works.

You’ve got to estimate and document measurement uncertainty, too. Your uncertainty budget should cover all significant sources of variability. ISO 17025:2017 Clause 7.6 doesn’t let you skip this.

Store all your verification records where staff can get to them fast. If an assessor asks for a specific method’s verification on a certain matrix, you should be able to pull it up in a few minutes, not go on a wild goose chase.

When can published standards or manufacturer claims be used to reduce the scope of verification, and what evidence is still required?

If you’re using published standard methods from AOAC, ASTM, EPA, or similar, they usually come with documented performance data. You can use those values as your acceptance criteria, but you still have to prove your lab can achieve them.

Manufacturer claims about instrument performance or kit specs aren’t enough. If they say their ELISA kit detects down to 0.1 ng/mL, you still need to show that’s true in your lab, with your analysts.

You can lean on published collaborative study data to set your precision targets, but you still have to show intermediate precision under your own conditions. Published repeatability values don’t get you out of testing across days and analysts.

Certified reference materials are useful for accuracy checks. You don’t need to re-certify them, but you do need to prove your method can recover those values accurately.

If the published method only has limited validation data—like if they only tested water, but you’re running botanical extracts—you’ll have to expand your verification. Matrix-specific checks are non-negotiable, no matter what the publication says.

Conclusion

Method verification really keeps your lab’s reputation intact and helps your test results hit the mark your clients expect. When you verify a standard method, you’re basically showing your lab can pull off the performance specs the method’s original developer intended.

This kind of verification work shows assessors you know what you’re doing and gives everyone a little more confidence in your testing. It’s important to document your verification studies thoroughly and keep those records handy for when someone wants to take a look.

Key points to remember:

• Verification covers standard methods as long as you’re not tweaking them
• You need to check accuracy, precision, and whatever other parameters make sense
• Your documentation needs to spell out both the acceptance criteria and your actual results
• If verification doesn’t work out, you’ve got to fix the issues before using the method

Putting real effort into method verification can save you headaches—fewer mistakes, happier customers, and a much easier time during accreditation assessments. Your lab gets to show it can actually run these methods right, under your own unique circumstances.

Honestly, verification isn’t something you do once and forget about. If you swap out equipment, bring in new people, or change up materials, you might need to revisit your verification. Keeping an eye on things with regular quality control samples helps you stay on top of method performance in the long run.