You just expanded your scope, and your assessor wants proof that your results hold up. Your first thought is finding proficiency testing (example of PT provider), and there is not a suitable one to be found. Now what? An interlaboratory comparison ISO 17025 is the most common and arguably the best way to verify your lab’s ability to perform a testing or calibration method. In an Interlaboratory Comparison (ILT), you measure the same device another accredited lab already measured, then compare the two results with using an error formula that takes both labs uncertainty into account. This guide shows you how to run that comparison, calculate the number that decides pass or fail, and package the evidence your assessor expects.
If you want a PDF version of this guide click here.
Why ISO 17025 Wants you to Compare your Results
Clause 7.7 of ISO/IEC 17025:2017 asks you to ensure the validity of your results. The ISO 17025 standard requires this because it’s industry’s best practice that accredited lab results agree with one another when they are using the same methodology for testing/calibration. This is great for customers of accredited labs, because they can be rest assured that ISO 17025 accredited labs have had to provide evidence for the validity of their testing/calibration results.
Many labs quickly turn to proficiency testing to meet this requirement. Which is a great approach, and often we recommend this. However, there are times when those tests are not readily available or available at all. This is where labs can employ Interlaboratory Comparison or (ILTs).
Read that clause closely, because it settles a common worry. Proficiency testing is not the only accepted way to comply with clause 7.7. An interlaboratory comparison satisfies the requirement as well.
- Important Note: You must get your Proficiency Testing and ILT plan approved by your accreditation body (e.g., PJLA, A2LA, NVLAB, ANAB). They typically want to see Proficiency testing as the first course of action. If ILTs are needed, you must provide a reason and get it approved by your accreditation body.
When an Interlaboratory Comparison for ISO 17025 Makes Sense
An interlaboratory comparison under ISO 17025 fits several situations well. Below are a few reasons why you may want to do ILTs:
- You added a parameter and need interim verification before the next proficiency testing round
- PT not available
- PT is too expensive
- PT does not cover the range of the scope of measurement so ILT may be better
How to Run an Interlaboratory Testing – Step by Step
The workflow for Interlaboratory testing stays the same across most measurement disciplines. This example uses a balance, since scales show the method clearly.
- Pick your Device (DUT). Choose an item with a recent calibration certificate from an accredited laboratory. A scale you already own works well, because you hold its certificate from a prior calibration provider.
- Measure the same item yourself. Run your own calibration on that Device (DUT). Capture your as-found and as-left data at each nominal point you plan to compare.
- Line up the shared points. Put both labs’ readings side by side. Match them by nominal value, such as 50 g, 100 g, and 200 g.
- Record each expanded uncertainty. Pull the expanded uncertainty for each lab at each point.
- Calculate the normalized error. Compute the En number for every matched point.
- Check Compliance for Each Point. Accept the point when the absolute value of En stays at or below 1.
Build this as a short interlaboratory comparison report. One column holds your readings. The next column holds the other lab’s readings. The following columns hold each lab’s uncertainty and the resulting En number. That layout reads cleanly, and it maps straight onto what your assessor wants to see.
The En Number, Explained
The En number, or normalized error, tells you whether two results agree once you account for their uncertainties. The formula is straightforward:
- x₁ is the measurement result from lab 1
- x₂ is the measurement result from lab 2
- U₁ is the expanded uncertainty at that measurement point (x₁) from lab 1
- U₂ is the expanded uncertainty at that measurement point (x₂) from lab 2
- The numerator captures the raw difference between the labs.
- The denominator captures how much difference the combined uncertainty already allows.
- Divide one by the other, and the result comes out unitless.
You would use this formula for every nominal or measurement point that you would want to compare. You will need to calculate an En value for every measurement point (range or nominal) that you want to compare.
That structure explains the acceptance rule. When the difference between labs stays smaller than the combined uncertainty, En lands between −1 and +1, and the point passes. When the difference outgrows the uncertainty, En climbs past 1, and the point fails. So the En number rewards agreement, and it forgives small gaps that your stated uncertainty already predicts.
Two details protect your math. First, use expanded uncertainties, not standard uncertainties, on both sides. Second, confirm both labs report at the same coverage factor before you compare. Mix a k = 2 value with a k = 1 value, and your En number may not represent how close your data points are. If a lab uses a different coverage factor than you, the easiest thing to do is to expand your uncertainty to the same coverage that the other lab used.
Stop guessing what your assessor will expect.
Get practical guidance for ISO/IEC 17025 or ISO 9001 implementation, documentation, audit preparation, measurement uncertainty, and laboratory accreditation.
A Worked Example – Interlaboratory Testing for Mass Measurement
Suppose your lab reads 99.9998 g at the 100 g point with an expanded uncertainty of 0.20 mg. The other lab reads 99.9999 g with an expanded uncertainty of 0.05 mg. The difference is 0.1 mg. The combined uncertainty is √(0.20² + 0.05²), roughly 0.21 mg. En comes out near 0.5. That point passes with room to spare. See the table below for a summary of this example.
What if you are a Testing Lab?
Testing laboratories can also perform interlaboratory comparisons. Instead of calibrating a shared artifact, each laboratory tests the same sample for one or more analytes. In this example, two laboratories receive portions of the same water sample and use ion chromatography with conductivity detection to determine fluoride, chloride, and bromide concentrations. The laboratories then compare their reported results and expanded uncertainties using normalized error.
Testing Laboratory Interlaboratory Comparison Example
Two laboratories analyzed the same water sample using ion chromatography with conductivity detection. Results are reported in milligrams per liter.
| Analyte |
Lab 1 Result mg/L |
Lab 1 Expanded Uncertainty, U mg/L |
Lab 2 Result mg/L |
Lab 2 Expanded Uncertainty, U mg/L |
Difference mg/L |
Combined Uncertainty mg/L |
En | Evaluation |
|---|---|---|---|---|---|---|---|---|
| Fluoride | 1.02 | 0.06 | 0.99 | 0.05 | 0.03 | 0.078 | 0.38 | Satisfactory |
| Chloride | 25.4 | 0.80 | 25.0 | 0.70 | 0.40 | 1.063 | 0.38 | Satisfactory |
| Bromide | 0.48 | 0.04 | 0.50 | 0.05 | -0.02 | 0.064 | -0.31 | Satisfactory |
How to Handle Gaps and Mismatched Test Points
Labs rarely run the exact same nominal points. One lab jumps from 10 g to 50 g. Another skips 150 g. You cannot force an En number where one side has no reading.
Handle these gaps cleanly. Leave the cell blank where no comparison data exists, then drop that row from the final report. Alternatively, keep the row and add a note that reads, “No interlaboratory comparison data available at this nominal value.” Either choice works. The point is to never invent a reading and never let a blank field feed a broken calculation. A short line in the special notes keeps the record honest and easy to audit.
Creating an Interlaboratory Testing Report for the Auditor
Your evidence package matters as much as your results do. Assemble it in a form the auditor can easily follow without having to ask you questions. A typical final report for Interlaboratory Comparison contains the following:
- A cover page that includes (reason for the report: method verification, the date of the report, both performing labs information)
- A data summary page that shows each of the test points, each labs readings and their uncertainties, the En Value, and a status of Pass or Fail
- The two calibration certificates used in the interlaboratory comparison attached as evidence
Example Cover Page
Example Data Page
Compile these two pages into a report and give it to your assessor. It shows evidence for verification of your measurement method with an outside source, a recognized statistic, and a clear pass decision, all traceable to source certificates.
Save time and Download our Template
Calculate En Values using this FREE document template. Just populate the Excel sheet with the data from your reports.
Put Interlaboratory Testing to Practice
An interlaboratory comparison for ISO 17025 is simpler than it sounds. Anchor it in Clause 7.7.2, which accepts comparison with other labs as valid monitoring. Measure a shared device, align the points, and compute the En number for each. Accept every point that lands within −1 and +1. Then attach both certificates, add a clean cover report, and note your conclusion for the assessor. That sequence gives you real evidence of competence while you wait on a proficiency testing round.
Want a faster start? Precision ISO builds uncertainty budgets and reporting tools that already carry the En calculation and the interlaboratory comparison layout. Tailor them to your scope and your methods, and you turn the steps above into a report your assessor will recognize as sound.
Why Partner with Precision ISO
Precision ISO supports laboratories and quality teams across pharmaceutical, medical device, aerospace, and calibration service environments. We help organizations evaluate calibration vendors, review scopes of accreditation, build internal qualification procedures, and prepare for ISO 17025 and ISO 9001 audits. Whether you are selecting an outside lab for the first time or rebuilding your supplier qualification process, our consulting engagements give you a clear, defensible path forward. Visit www.precisioniso.com to schedule a free 30-minute consultation.
Stop guessing what your assessor will expect.
Get practical guidance for ISO/IEC 17025 or ISO 9001 implementation, documentation, audit preparation, measurement uncertainty, and laboratory accreditation.
