ISO 11930 Preservative Efficacy Testing: The Criterion A vs. Criterion B Mistake Most EU Cosmetic Brands Make

About a third of the challenge test failures our team reviews come down to one decision made long before any microorganism touches the product: which acceptance criterion to apply.

Under ISO 11930:2019, the EU’s reference standard for evaluating antimicrobial protection in cosmetic products, brands must choose between Criterion A and Criterion B. Choose wrong — typically by defaulting to the less demanding Criterion B when Criterion A is actually required — and you’ve built compliance documentation on a foundation that won’t survive regulatory scrutiny. The product that “passed” may not be adequately preserved at all.

It’s a quiet but consequential error. And it’s remarkably common, particularly among smaller brands working with laboratories that report test outcomes without advising on criterion selection.

What ISO 11930 Actually Tests — and What It Doesn’t

The standard assesses whether a formulation can adequately limit microbial growth under conditions representative of normal consumer use. It isn’t a sterility test, and it doesn’t measure contamination present in the product as manufactured. Instead, it introduces deliberate, quantified contamination with five reference microorganisms and tracks what happens across 28 days:

• Pseudomonas aeruginosa (ATCC 9027)
• Staphylococcus aureus (ATCC 6538)
• Escherichia coli (ATCC 8739)
• Candida albicans (ATCC 10231)
• Aspergillus brasiliensis (ATCC 16404)

The three bacterial strains are inoculated and tracked separately from the two fungal strains. Viable counts are taken at four time points — day 2, day 7, day 14, and day 28 — and compared against the initial inoculum using log₁₀ reductions or increases.

What the test cannot tell you is whether your preservative system will hold up against every organism a consumer might introduce through use. These five strains are selected for their representative risk profiles — they’re not the only microorganisms that matter. The test also doesn’t account for product dilution with water during consumer use, a real-world variable that can drop effective preservative concentrations significantly. These are known limitations, acknowledged within the standard itself.

What it can tell you is whether the preservative concentration and combination you’ve chosen is doing its job at the formulation level. That’s precisely the claim that Annex I of Regulation (EC) No 1223/2009 requires you to support in your Product Information File — documented evidence of microbiological quality, including adequate preservation under intended use conditions.

Criterion A vs. Criterion B: The Classification Decision That Shapes Everything

Here’s where most misclassifications happen. ISO 11930 defines two sets of acceptance criteria, and which one applies depends on how and where the product is used — not on a product category label.

Criterion A is mandatory when any of the following conditions apply:

• Products applied near or around the eyes (mascaras, eyeliners, eye shadows, eyebrow products)
• Products applied to mucous membranes (lip products, toothpaste, mouthwash, intimate hygiene products)
• Products intended for use on broken or damaged skin (post-procedure care, wound-adjacent applications)
• Products specifically designed for children under 3 years of age (baby creams, bath foams, lotions)
• Any product where the relevant exposure conditions present an elevated microbiological risk

Criterion B applies to everything else — but with a critical qualification that brands frequently overlook: only when the responsible person can demonstrate and document why Criterion A does not apply.

That documentation step is not optional. It forms part of the microbiological quality justification in the Cosmetic Product Safety Report (CPSR).

The numeric thresholds between the two criteria differ substantially, and understanding them makes it obvious why misclassification matters.

For bacteria, Criterion A requires a reduction of at least 2 log₁₀ from the initial inoculum count by day 2, no increase from that reduced count at day 7, no increase at day 14, and no increase from the initial count at day 28. In practical terms: your preservative system must demonstrate rapid, aggressive kill within the first 48 hours.

Criterion B for bacteria requires only that there is no increase from the initial count at day 14, maintained at day 28. There is no day-2 or day-7 requirement at all.

For fungi, Criterion A requires no increase from the initial count at day 14 and no increase at day 28. Criterion B permits up to a 1 log₁₀ increase at day 14, provided the count hasn’t climbed further by day 28.

A 2-log-reduction requirement within 48 hours versus a hold-the-line requirement at two weeks — these are not interchangeable thresholds applied to the same underlying data. A formulation that achieves Criterion B hasn’t been evaluated against the Criterion A benchmarks at all. Applying Criterion B to an eye product, then citing the passing result in your safety report, is a gap that a qualified safety assessor reviewing your PIF will identify immediately.

Three Common Failure Patterns — and What They Mean for Your Formulation

When a preservative efficacy test fails, the failure mode almost always points directly to the formulation problem. Here are the patterns we see most frequently.

Late bacterial recovery in aqueous systems

Bacterial counts drop acceptably in the first week, then climb back above the initial count by day 28. This typically indicates that the preservative has been depleted or inactivated over time — usually through interaction with an emulsifier, polymer, or high-molecular-weight ingredient that sequesters the active in a bound, antimicrobially unavailable form.

Phenoxyethanol, one of the most commonly used cosmetic preservatives in Europe, is well documented to partition preferentially into oil phases and into surfactant micelles. In formulations with a substantial oil fraction or high-surfactant content, the effective aqueous-phase concentration of phenoxyethanol may be considerably lower than the nominal inclusion level. The fix isn’t automatically to increase the total preservative concentration — it’s often to address the solubilisation system so that adequate free-phase activity is achieved.

Fungal failure linked to undocumented pH justifications

Products with pH below 4 or above 9 are sometimes assigned Criterion B classification on the implicit assumption that extreme pH itself provides antimicrobial protection. ISO 11930 doesn’t grant Criterion B status based on pH alone. The justification must be documented with supporting data, and the test must still be conducted. When that documentation step is skipped — common when labs simply run the test without advising on classification — the Criterion B assignment is indefensible to a notified body reviewing the PIF.

Implausibly high early kill rates caused by neutralisation failures

ISO 11930 requires a validated neutralisation step to confirm that any preservative carry-over to the recovery medium is inactivated before colony counting. When the neutralisation protocol isn’t correctly validated for a specific preservative system, what appears as a passing result at day 2 may reflect preservative-mediated kill on the agar plate rather than within the product sample.

This doesn’t flag as a numerical failure in the raw data — it shows up as suspiciously near-complete kill at day 2, which should prompt careful review of the neutralisation controls. Many contract laboratories report numeric outcomes without flagging this pattern. Catching it requires knowing what the data should look like at each time point, not just whether the final numbers meet the criterion.

Connecting Your ISO 11930 Results to the EU Product Information File

Under Article 11 of Regulation (EC) No 1223/2009, the PIF must contain a cosmetic product safety report (CPSR) prepared by a qualified safety assessor. Part B of the CPSR covers the safety assessment, including microbiological quality — and this is where your ISO 11930 data must be integrated, not merely appended.

The assessor cannot attach the lab report and move on. The CPSR must include a reasoned evaluation of whether the selected criterion was appropriate, whether the test was conducted on a formulation representative of the finished product at commercial preservative levels, and whether the results demonstrate adequate protection for the product’s intended use and packaging configuration.

Packaging matters more than brands typically expect. A product tested in bulk but sold in a wide-mouth jar faces a meaningfully different real-world contamination profile than the same formulation in an airless pump. Some safety assessors require packaging-specific justification, particularly for Criterion A products where the margin for error is narrow.

Three documentation gaps that most commonly generate queries from safety assessors and notified bodies:

1. ISO 11930 results without a recorded and justified criterion selection
2. Criterion B classification applied to products carrying an eye-area claim or child-directed marketing — both of which mandate Criterion A regardless of the formulation’s physical chemistry
3. Testing conducted on a prototype or over-preserved development batch rather than the finished commercial formulation

If your brand also exports to North America, one further complication is worth understanding clearly: ISO 11930 and USP <51> Antimicrobial Effectiveness Testing are not equivalent. The five reference organisms overlap substantially, but the acceptance criteria, time points, and category structure differ. A product that passes ISO 11930 Criterion A for bacteria may not satisfy USP <51> Category 2 at equivalent time points, and vice versa. Labs capable of running both protocols simultaneously, with a single sample submission and cross-protocol interpretation, reduce the compliance burden for dual-market launches considerably. That capability is genuinely uncommon in Europe.

Getting the Foundation Right Before You Test

The preservative efficacy test result is only as reliable as the decisions made before the sample arrives at the laboratory. Before commissioning your next ISO 11930 study, confirm three things:

First, that you’ve selected and documented the correct criterion for each product in your range — not as a blanket decision by product type, but per product, per intended use, per target consumer group. A body lotion with a “safe for babies” marketing claim and a body lotion without one are not the same product from a criterion-selection perspective.

Second, that the submitted sample is the commercial formulation at the commercial preservative concentration — not a stability prototype, not a batch with a topped-up preservative level, and not a formulation that will change before launch.

Third, that your laboratory can provide validation data for the neutralisation step in your specific preservative system. If they can’t, the test results are not interpretable with confidence.

A clean ISO 11930 result, obtained against the correct criterion, with validated neutralisation and representative formulation, is a genuinely useful piece of evidence for your safety case. Anything less than that is a number that looks like compliance documentation but isn’t.

---

Written by Nour Abochama, Quality & Regulatory Advisor, Care Europe | VP Operations, Qalitex. Learn more about our team

Talk to our team about EU market entry and cosmetic compliance documentation. Contact us

Related from our network

• ISO 17025-accredited cosmetic and supplement testing services — Qalitex Laboratories provides accredited challenge testing, microbiology, and safety testing for brands preparing EU and North American compliance dossiers.
• Health Canada-compliant cosmetic and NHP testing for European brands — Androxa specialises in Canadian market testing requirements, including antimicrobial effectiveness testing to USP <51> alongside EU protocols.