Why Clean Beauty Formulas Are Failing EU Preservation Tests — And What to Do About It

Three months before a mid-size French skincare brand was due to launch its new “preservative-free” serum across DACH markets, their safety assessor flagged a problem: the Cosmetic Product Safety Report couldn’t be finalised. The preservation efficacy test had failed. The formula — built around fermented postbiotics and a plant-derived antimicrobial blend — hadn’t cleared ISO 11930 Criteria B for yeasts and moulds. The launch slipped by seven weeks. The reformulation cost more than the original development budget.

This isn’t an unusual story. We’ve been seeing versions of it repeat with increasing frequency as more European cosmetic brands pursue natural or “free-from” positioning. The commercial instinct is understandable — consumer sentiment around certain preservatives has shifted sharply. But the regulatory and microbiological reality of the EU market doesn’t bend to marketing trends.

The Obligation That Doesn’t Disappear

Under Article 10 of EU Cosmetics Regulation 1223/2009, every cosmetic product placed on the EU market must be supported by a Cosmetic Product Safety Report (CPSR) compiled by a qualified safety assessor. Part B of that report includes an assessment of microbiological quality and — critically — antimicrobial protection. In practice, that means a verified, documented preservation system.

The reference standard for demonstrating that protection is ISO 11930:2019, Evaluation of the antimicrobial protection of a cosmetic product. It’s a challenge test: five microbial species (three bacteria, one yeast, one mould) are inoculated into the finished product at defined concentrations, and recovery is measured at days 2, 7, 14, and 28. The product must meet either Criteria A (more stringent) or Criteria B (less stringent, but still demanding), depending on its product risk profile.

Criteria A requires a minimum 2-log reduction in bacterial count at day 14, no increase from day 14 to day 28, and no increase in yeast or mould counts at any time point. Criteria B is acceptable for certain product types and allows a 2-log reduction in bacteria by day 14 with at least a 0.5-log reduction in fungi. These aren’t soft targets. A lot of formulas fail to meet even Criteria B — and a failed test means your safety assessor can’t sign off the CPSR. No CPSR, no legal market access in the EU.

Why Clean Beauty Formulations Are Struggling

The reformulation pressure has been building for years. Commission Regulation (EU) No 358/2014 tightened maximum concentration limits for propyl- and butylparaben, capping them at 0.14% each in leave-on and rinse-off products. Combined with ongoing controversy around phenoxyethanol — the SCCS restricted its use in products intended for children under three in Opinion SCCS/1638/21 (2021) — brands have been pulled toward alternatives.

But the ingredient reality is difficult. Annex V of Regulation 1223/2009 lists 57 permitted preservatives with defined maximum concentrations and use restrictions. Many of the natural alternatives that brands are drawn to — rosemary extract, certain essential oil fractions, specific neem derivatives — aren’t listed in Annex V as preservatives at all. They can appear in formulations as functional or botanical ingredients, but they cannot serve as the documented preservation system, and they frequently don’t deliver the antimicrobial performance needed to pass ISO 11930, even at concentrations that would exceed their cosmetically acceptable use levels.

Here’s an insider problem that’s harder to see from the outside: many formulators are relying on preservation challenge data supplied by ingredient vendors, data generated on simplified model matrices — typically a 5% glycerol-in-water base, sometimes with basic emulsifiers, but not the full finished formulation. Real-world product complexity changes everything. The pH of your formula, its water activity, the presence of chelating agents, the ratio of oil phase to water phase, particle size — all of it interacts with the antimicrobial system in ways a model matrix doesn’t capture. A preservative that performs in a supplier’s lab can be sequestered, deactivated, or overwhelmed in your actual emulsion.

We frequently receive samples from brands who have completed in-house development but delayed ISO 11930 testing until the final stage before submission. By that point, changing the preservation system — which may mean adjusting pH buffers, modifying emulsifier choices, or rethinking the entire aqueous phase — is genuinely disruptive. Late-stage reformulation typically costs 3 to 5 times more than catching the same problem during early development. It also burns the one resource that’s hardest to recover: time to market.

What ISO 11930 Is Actually Testing (And What It Isn’t)

One misconception worth addressing directly: ISO 11930 tests the antimicrobial protection of the finished product. It doesn’t validate that the product arrived at testing free of contamination, and it doesn’t predict performance across every environmental exposure scenario during use. A product can pass ISO 11930 and still present microbiological risks if manufacturing hygiene under ISO 22716 — the GMP standard for cosmetics — is inadequate, or if the primary packaging doesn’t provide the intended barrier to contamination.

This distinction matters for your CPSR. A safety assessor reviewing Part B will expect to see ISO 11930 data alongside the manufacturing hygiene controls documented in your GMP file. The two are complementary, not interchangeable.

The other thing ISO 11930 doesn’t reveal is the margin of safety. A formula that just scrapes through Criteria B and one that clears Criteria A by a comfortable margin both generate a “pass” result on paper — but they represent very different risk profiles in the real world. For products with complex supply chains, particularly those distributed across multiple EU climatic zones or exported to markets with higher ambient temperatures, a result sitting right on the Criteria B borderline is a commercial vulnerability. We advise clients to target Criteria A wherever the formulation allows, even for product categories where B is technically acceptable. The additional security is worth the formulation effort.

Where Brands Most Commonly Get Into Trouble

After reviewing preservation test data across a substantial number of EU cosmetic submissions, the failure patterns are consistent enough to generalise.

Water activity assumptions. Some brands reduce water activity as a core preservation strategy — valid in principle, but it needs to be measured and validated in the finished formula, not assumed from ingredient ratios. Achieving water activity below 0.75 Aw in an emulsion that still performs well on skin isn’t a straightforward engineering task.

Non-standard challenge conditions. ISO 11930:2019 specifies testing at 25°C ± 1°C. Labs with limited cosmetic experience occasionally run tests at lower temperatures. A test conducted at 20°C can meaningfully overestimate real-world antimicrobial performance — and when the product is actually stored and used at summer temperatures in southern Europe, that difference matters.

Undocumented Criteria A/B decisions. The standard permits Criteria B for certain product categories and risk profiles, but that decision needs to be explicitly justified in the CPSR, not assumed. Safety assessors are increasingly asking for that documentation. “We chose Criteria B because it was acceptable” isn’t sufficient; you need a documented risk rationale.

Rinse-off data applied to leave-on products. The microbial risk profile for a hair conditioner is fundamentally different from that of an eye cream. Contact time, skin area, proximity to mucous membranes, and application site all affect risk classification. We’ve seen submission packages where leave-on eye-area products were supported by preservation data from rinse-off cleanser testing. That’s a straightforward CPSR deficiency.

Packaging interaction overlooked. The same formula in an airless pump versus a wide-mouth jar presents entirely different contamination exposure. If you’re supporting multiple pack formats with a single ISO 11930 result, your safety assessor may require justification — or separate tests.

Building a Preservation Strategy That Actually Survives Testing

The answer isn’t to abandon natural or low-toxicity preservation approaches — it’s to test them properly, earlier, and with more realistic models. A fit-for-purpose preservation development process in 2026 typically follows this structure:

1. Early-stage screening using an abbreviated challenge test on prototype formulas, before the ingredient deck is finalised. This narrows your options cost-effectively, before you’re committed to a formulation direction.
2. System compatibility review integrated with your accelerated stability programme. Preservation efficacy can degrade significantly between T0 and 6 months under accelerated conditions (40°C/75% RH), and that data belongs in your CPSR alongside the challenge test result.
3. Full ISO 11930:2019 testing on the finished product in its commercial primary packaging, with testing on at least two production lots to demonstrate batch-to-batch consistency.
4. CPSR integration from the start. The safety assessor should be involved early — not handed a completed package at the end and asked to sign it quickly. Building the preservation data into Part B alongside the toxicological assessment produces a stronger document and catches problems before they become launch delays.

For brands developing products intended for multiple markets, it’s worth noting that the FDA’s MoCRA framework (Modernisation of Cosmetics Regulation Act, 2022) doesn’t have a mandatory CPSR equivalent, but it does require facility registration and GMP compliance. Your ISO 22716 documentation and preservation validation data carry weight in that process too. A testing programme designed from the start to satisfy both EU and US requirements doesn’t cost twice as much — it costs perhaps 30% more than designing for a single market, and it eliminates the cost of a second development cycle later.

Our partner-lab network covers EU preservation efficacy and stability testing, US and Canadian market compliance requirements, and raw material identity verification — which means we can structure a single coordinated testing programme rather than a series of disconnected projects. For brands navigating a multi-market launch under time pressure, that coordination matters more than most people expect.

---

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. Contact us

Related from our network

• Cosmetic and supplement testing to ISO 17025 and US FDA standards — Qalitex Laboratories provides ISO 11930 preservation efficacy testing and CPSR-ready analytical data for US and EU compliance programmes.
• Canadian GMP and NHP compliance for European cosmetic and supplement brands — Androxa supports European manufacturers navigating Health Canada requirements, including preservation and microbiological testing aligned with Canadian standards.