The Tea Ledger
Testing & Lab Reports · 2026-11-18

What Is A Third-Party Tea Test? Eurofins, NSF And Independent Labs

third-party testing tea lab test ISO 17025 COA pesticide residues
Quiet laboratory corridor with a lab coat on a hook.

A supplier tells you their tea is "tested". That word sounds reassuring, but on its own it means very little. Tested by whom? For which pesticides? At what detection level? And on which bag from which lot?

A third-party tea test is simply a laboratory analysis carried out by an organisation independent of both the grower and the buyer. It is one of the few tools that turns a claim into evidence. This article explains what these labs do, what their reports actually tell you, and where the limits of a test lie.

What "third-party" really means

The key idea is independence. A first-party test is done by the seller themselves. A second-party test is arranged by the buyer. A third-party test is performed by a separate laboratory with no stake in whether the tea passes or fails.

That independence matters because a residue result is a commercial fact. It can decide whether a container is accepted, a batch is rejected, or a contract is honoured. Having an outside lab produce the number reduces the temptation to massage it.

It does not make the lab infallible. A third-party result is still only as good as the sample, the method and the competence of the laboratory. Those are the three things to examine.

Who are these labs?

Several well-known testing companies handle food and tea analysis around the world. Common names you may encounter include Eurofins, SGS, Intertek, Bureau Veritas, ALS and NSF.

These are real, established testing companies, and they are mentioned here only as examples of the kind of independent laboratory you might see on a report. Naming a laboratory does not mean that laboratory certifies any particular tea brand, and you should be wary of any seller who implies otherwise. A lab tests samples; it does not endorse a product.

You are also free to use a smaller local laboratory. What matters is not the size of the name but whether the lab is competent and whether the report is clear.

The accreditation to look for: ISO/IEC 17025

The international benchmark for testing laboratories is ISO/IEC 17025. It is the standard for the competence of testing and calibration laboratories. When a lab is accredited to it, an independent body has assessed the lab's methods, equipment, staff and quality system and found them competent to perform specific tests.

Accreditation is not a mood or a marketing badge — it is a formal status with a scope. A lab may be accredited for pesticide analysis but not for heavy metals, or for certain matrices but not tea. So check two things:

  • that the laboratory holds ISO/IEC 17025 accreditation;
  • that the specific test you need falls within its accredited scope.

If a seller cannot produce a report from an accredited lab, or cannot tell you the scope, you are entitled to ask questions before you trust the result.

What should be on the report

A useful Certificate of Analysis (COA) or test report is more than a single number. A good report identifies:

  • the laboratory, including its name and accreditation details;
  • the sample or lot ID, so you can tie the result to a specific batch;
  • the date the sample was taken and the date the report was issued;
  • the analytes tested — which substances were actually looked for;
  • the method used;
  • the result and its unit, such as mg/kg;
  • the LOQ and LOD for each analyte;
  • a pass or fail against a stated limit.

If any of these are missing, the report is weaker than it looks. A result without an analyte list tells you nothing about what was screened. A result without a lot ID cannot be connected to the tea in front of you.

LOD, LOQ and the trap of "not detected"

Two terms cause endless confusion, so it is worth defining them plainly.

LOD means limit of detection — the lowest amount of a substance the method can detect at all.

LOQ means limit of quantification — the lowest amount the method can measure with acceptable precision and accuracy.

Detection and quantification are not the same thing. A lab may see a signal at the LOD but not be able to put a reliable number on it until the level reaches the LOQ.

This is why "not detected" (often written ND) does not mean "zero". It means the substance was below the method's detection capability. A well-run report will state the LOD or LOQ alongside the ND so you can judge how low the detection actually went.

For tea entering the EU, this links directly to the law. Where no specific MRL exists for a substance, Regulation 396/2005 applies a default of 0.01 mg/kg at the level of quantification. If your lab's LOQ is above that figure, an ND result is not meaningful evidence of compliance — the method simply was not sensitive enough.

Sampling is where most tests go wrong

Here is the uncomfortable truth: testing one bag from a lot does not prove that every bag in the lot complies. Tea is a bulk agricultural product, and residues are not distributed perfectly evenly through it.

Sampling is statistical. The general principle, reflected in standards such as ISO 2859 and normal sampling practice, is that a small number of samples can only give a certain level of confidence about the whole lot. This is not a magic number you can look up once and apply everywhere; the right plan depends on the lot and the risk. What you should take away is the concept: the result describes the sample, and only by inference the lot.

That has practical consequences:

  • Ask how many samples were taken and how they were drawn from the lot.
  • Be cautious of a single "representative" sample with no method behind it.
  • Remember that a clean COA is evidence about a moment and a batch, not a permanent property of a tea.

A laboratory can run a flawless analysis on a poorly chosen sample and still give you a misleading answer.

How to use a third-party test as a buyer

Bring curiosity, not blind faith. Reasonable questions to put to a supplier include:

  • Which laboratory did the test, and are they ISO/IEC 17025 accredited for this analysis?
  • Can I see the full COA, not just a summary?
  • Which pesticides were screened, and what were the LOQs?
  • Which lot does this report refer to, and when was it sampled?
  • How were the samples taken from the lot?

A supplier who welcomes these questions is usually worth listening to. One who deflects them, or offers only a glossy certificate with no analyte list, is telling you something too.

Third-party testing is a tool for managing risk, not a guarantee that can replace your own judgement. Used properly — accredited lab, clear report, sensible sampling — it is the best evidence available that a tea meets a stated standard.

Bottom line

A third-party tea test is an independent lab analysis, and it is valuable precisely because the lab has no stake in the outcome. Look for ISO/IEC 17025 accreditation within the right scope, read the full COA, understand that ND means "below detection" rather than "zero", and remember that a result is only as strong as the sampling behind it. The lab gives you evidence; it is still your job to read it carefully.

Sources & further reading

Related reading