Toy testing and CE marking basics: choosing tests that match the toy, the age group and the materials

A toy test plan should start from the toy, its age group and its materials, not from a standard test package copied from another toy. A plush toy for toddlers needs seam strength, stuffing safety and washability checks. A magnetic building set for older children needs magnet attachment, small part and breakage checks. An electronic toy needs battery compartment, sound and heat checks alongside mechanical checks. A test plan that applies the same package to all three will over test some hazards and miss others, while looking thorough on paper.

Mechanical tests examine what a child can do to a toy: pulling, twisting, dropping, biting, climbing and throwing, and what the toy does in return, including breaking, shedding parts, trapping fingers, or tipping. Chemical tests examine what a child absorbs or touches: substances in paints, plastics, textiles and fillings, especially in mouthing parts. Electronic tests examine batteries, chargers, sound, heat and connected features. A complete toy file combines the relevant families, rather than treating one family as a proxy for the others, since a toy can pass mechanical tests while failing chemical ones, and vice versa.

Age group shapes each family: mechanical tests for toddlers emphasise small parts and mouthing, while mechanical tests for older children emphasise projectiles and tools. Chemical tests emphasise mouthing parts for toddlers and prolonged skin contact for older children. Electronic tests emphasise battery compartment security for toddlers and charger and sound behaviour for older children. A test report that does not state the age group it was run for should be queried, since age determines the test conditions.

Samples for toy testing should come from production, or from a batch matching production, and should include the toy as sold with its packaging, battery and accessories. A prototype hand finished for testing, without its final paint or magnet, does not represent the toy a child receives. Reports should be read for product identification, sample source, age group, conditions, and any test skipped, failed or passed after a fix. A report that lists a failure, a fix and a retest is more credible than a report showing only passes, since it shows the toy was challenged.

Illustration: a seller of wooden trains received a report that passed every test, but the report described a train without its magnetic couplings, which were added later. The seller commissioned a new test on the complete train as sold, and the magnetic couplings then received their own attachment checks. The story is illustrative, and the lesson is to compare the report product description with the toy as sold, line by line, before filing the report as evidence.

A failed toy test should trigger a design or material change, a retest on the corrected toy, and a record of the failure, the fix and the retest together. Hiding a failure, retesting a different sample quietly, or narrowing the test to avoid the failure all destroy the value of testing and create a file that cannot be trusted. Where a failure suggests a wider issue, such as a component shared across toys, widen the check to those toys before concluding the failure was isolated, and record that wider consideration even when it concludes isolation.

Also consider whether a failed test means toys already sold need action. A failure in a current production sample may indicate a problem in toys already with children, and that question should be raised promptly with the supplier and, where needed, through corrective action, rather than waiting for complaints. Testing is not only a launch gate; it is a monitoring tool that can trigger action after sale, and the file should show that the after sale question was asked.

Toy testing should be repeated when a supplier, material, colour, component or design changes, and periodically for continuing ranges, since production drift can change a toy without changing its appearance. Keep a change log linking each change to the tests it affects and the retest result, and compare current production samples with launch samples at intervals. A toy range tested honestly at launch, and retested honestly at each change, earns confidence from retailers, parents and authorities alike, while a range whose testing stopped at launch earns only the appearance of confidence.

Think of toy testing as an ongoing conversation with a toy, not as a single exam it passes once. Each production batch, each supplier change, and each complaint adds a new question, and the test programme should answer it and record the answer. A toy whose testing continues through its life in the market is a toy whose safety is actively known, rather than historically assumed, and that active knowledge is what parents, retailers and authorities are really asking for when they ask for test evidence.