How to Brief a Laboratory for Product Testing
A badly briefed test produces a number you cannot interpret, which is worse than no test at all. Specify the method, state the limit you are testing against, and say what question you want answered. Then the result means something.
Why Briefing Matters
This guide is for supplement, Ayurvedic and herbal brand owners commissioning independent testing in India.
A laboratory will test what you ask for by whichever method it uses as standard. If you have not specified the method, the result may not be comparable with your manufacturer’s figure. If you have not stated the limit, the report tells you a number without telling you whether it passes. And if you have not said what you are trying to establish, the laboratory cannot tell you that a different test would answer it better.
The consequence is a report that creates uncertainty rather than resolving it, which then prompts an unnecessary dispute with a manufacturer or an unwarranted reassurance.
A good brief takes a few minutes and makes the result actionable.
What a Good Brief Contains
Include all of these.
- The product, its form and a brief description.
- The batch number and the date it was received by you.
- Each parameter you want tested, named specifically.
- The method for each parameter, by name, matching your specification.
- The specification limit for each parameter, so the report states pass or fail against it.
- What question you are trying to answer, in one line.
- How the sample was taken, stored and transported.
- Whether you want a retained portion held by the laboratory.
- The reporting format you need, including whether limits of detection must be stated.
- Any context that affects interpretation, such as a complaint about the batch.
Common Briefing Failures
Each produces a result you cannot use.
| Failure | Consequence |
|---|---|
| No method specified | Result not comparable with the manufacturer’s figure |
| No limit stated | A number with no pass or fail context |
| Wrong parameter requested | A test that does not answer your question |
| No batch number | A result that cannot be tied to anything |
| Sample taken poorly | A result reflecting handling rather than the product |
| No limit of detection reported | A “not detected” result of unknown sensitivity |
| No retained portion | No ability to retest if the result is unexpected |
| No context given | The laboratory cannot flag that another test fits better |
The limit of detection point matters for contaminant testing. A result reported as not detected means little without knowing the sensitivity at which it was not detected.
Asking the Right Test
Framing the question lets the laboratory help.
Verifying a Declaration
Say so. The laboratory will use a method appropriate for a quantitative check against a stated figure.
Investigating a Complaint
Say what the complaint was. A different panel may be appropriate from routine verification.
Checking Authenticity
Say so explicitly, since authenticity screening uses different techniques from quality parameters.
Establishing a Baseline
Say so, and ask what parameters are worth including beyond the obvious ones.
A laboratory told what you are trying to establish will frequently suggest a better or cheaper route to the answer. One given only a parameter list cannot.
Interpreting the Result
Three cautions.
A single result is a single result. Natural material varies, and a figure near a limit is not necessarily a problem while a figure far outside one usually is. Compare against your batch log rather than reacting to one number.
An unexpected result warrants verification before escalation: confirm the method matched your specification, check the sampling and storage, and retest using your retained portion, ideally at a second laboratory.
And a result on your sample tells you about your sample. If the sampling was poor, the result reflects that rather than the batch, which is why how the sample was taken belongs in the brief.
Checklist
Before sending any sample.
- Confirm the laboratory’s accreditation scope covers your tests.
- Specify each parameter and its method by name.
- State the specification limit for each parameter.
- State in one line what question you want answered.
- Include the batch number and receipt date.
- Take the sample from received stock, properly and promptly.
- Record how the sample was taken, stored and sent.
- Retain a portion yourself.
- Ask the laboratory to retain a portion where possible.
- Require limits of detection on contaminant results.
- Give any relevant context, such as a complaint.
- Compare the result against your batch log rather than in isolation.
- Verify any unexpected result before raising it with your manufacturer.
- File the report with the batch record in your product file.
Want Documentation Ready for Verification?
Share your product and our team will provide batch documentation with methods and limits stated, ready for independent testing.
Frequently Asked Questions
What are the essentials of a testing brief?
The method for each parameter by name, the specification limit you are testing against, and one line stating what question you want answered. Without those three, the result is a number you cannot interpret.
Why specify the method?
Because a laboratory will otherwise use its own standard method, and results from different methods are not comparable. If your manufacturer reports by one method and the laboratory tests by another, a difference may be a method difference rather than a problem.
Why does the limit of detection matter?
Because a contaminant result reported as not detected means little without knowing the sensitivity at which it was not detected. Require limits of detection on contaminant results so the report can be interpreted.
Why tell the laboratory what I am trying to establish?
Because a laboratory told the question will frequently suggest a better or cheaper route to the answer, or flag that a different test fits. One given only a parameter list cannot help in that way.
What should I do with an unexpected result?
Verify before escalating: confirm the method matched your specification, check the sampling and storage, and retest using your retained portion, ideally at a second laboratory. Compare against your batch log rather than reacting to one figure.