How to select samples for testing and not redo the test report

Formally, sample selection is a half-hour procedure. In practice, this is where weeks are lost, when it turns out that the test report does not fit the required scheme.

What sets the rules for sample selection

What is recorded in the sampling record

• Name of the product and the designation of the modification
• Date and place of sampling, and which batch it is from
• Number of units selected
• Who selected them and in whose presence
• Conditions of transport to the laboratory

Typical mistakes

How to fit testing into the launch schedule — in the article on launching a product. What will be checked after the document is issued — in the material on inspection control.

Who selects the samples

The evidential value of the test report depends on this, and the difference is visible in any dispute.

Selecting the samples yourself is cheaper and faster, but during a check this is exactly where questions arise: what proves that the product tested is the same one you produce. The answer is given by the sampling record and careful identification of the sample.

How many samples are needed

The exact number is set by the laboratory based on the test methods, but the logic is always the same, and it is useful to understand it in advance — so as not to send one unit where six are needed.

• Destructive tests. The sample is consumed: strength, flammability, chemical analysis. A reserve is needed.
• Several groups of indicators. Microbiology, physical chemistry and toxicology are often done in different departments and need their own units.
• Arbitration sample. Part of the samples is kept in case of a dispute — this is normal practice.
• Different modifications. If the document covers several models, the typical representative is tested, and sometimes each model.
• Packaging. For products where contact with the packaging matters, the sample is supplied in consumer packaging, not loose.

Confirm the number before sending: sending more samples later extends the timeline more than it seems, especially if the product comes from abroad.

How to deliver the sample without losses

Testing starts with acceptance, and some samples are rejected at this stage.

• The packaging protects against damage: a broken sample cannot be tested
• For products with a temperature regime — keep the cold chain, otherwise microbiology will show the wrong result
• The marking on the sample is intact and readable: it is used to identify the sample
• Information on the batch, the date of manufacture and the expiry date is attached
• The sample is not opened, if the method requires intact packaging
• For imports — the samples were brought in legally, with proof of origin

What to record for the future

A year later, when an inspection control or a dispute with a buyer arrives, it is impossible to remember the details of the sampling. So right after the sampling it is worth saving:

• The sampling record with the date, place, batch number and signatures
• Photographs of the sample and its marking
• Information on which batch and shift the sample was taken from
• Who selected it and on what basis
• Where it was sent and what remained in storage

This set should be kept next to the test report: together they answer the main question of the inspector — whether the test relates to the product that lies on the shelf. How to check the laboratory itself and its accreditation scope — in a separate article.