Preservative-Free Doesn't Mean Risk-Free: What EU Cosmetics Regulation 1223/2009 Requires From Clean Beauty Brands

The “no preservatives” claim is everywhere in European cosmetics right now. It’s on shelf talkers in Paris pharmacies, embedded in brand stories across Instagram, printed in pale font on kraft-paper packaging from Stockholm to Milan. It sells well. And it’s creating a quiet compliance problem that’s now landing on the desks of safety assessors and Responsible Persons across the continent.

The issue isn’t the marketing claim itself — though that carries its own labeling implications. The problem is what happens at the formulation and documentation level when brands design for “clean” and forget that EU Cosmetics Regulation 1223/2009 has no section called “except for natural products.”

What Annex V Actually Says — And What It Doesn’t

Most formulators know the basics: Annex V of Regulation 1223/2009 is the positive list of permitted preservatives for EU cosmetics. It contains 57 numbered entries — specific substances, each with defined maximum concentrations, applicable product types, and in some cases restrictions on use in products for children or near mucous membranes.

If a substance appears on Annex V and you’re using it to inhibit microbial growth, you’re using it as a preservative. The permitted maximum for phenoxyethanol, for instance, is 1.0% — and the Scientific Committee on Consumer Safety (SCCS) reinforced that ceiling in Opinion SCCS/1575/16, including tighter guidance for use in products applied to the nappy area of children under three years old.

Here’s what Annex V does not say: it doesn’t say that products formulated without an Annex V preservative are automatically safe, microbiologically stable, or exempt from assessment. That’s the gap many clean beauty brands step into without realising it.

The multifunctional ingredients that brands typically substitute — ethanol at 10–15%, propanediol, levulinic acid, sodium anisate, rosemary extract — do carry antimicrobial properties. But when used for preservation purposes, they need to be assessed as such in Part A of the Cosmetic Product Safety Report (CPSR). If your safety assessor simply notes “no Annex V preservatives used” and moves on, you’re missing a required element of the technical dossier under Article 10.

Article 3 Doesn’t Negotiate With Aesthetics

Article 3 of Regulation 1223/2009 is deceptively brief: cosmetics placed on the EU market must be safe for human health under normal and reasonably foreseeable conditions of use.

“Safe” in microbiological terms means controlled microbial load throughout the product’s shelf life — from the manufacturing line to the last use by the consumer. A face cream that passes microbial testing on Day 1 but supports Pseudomonas aeruginosa growth by month six isn’t compliant. It doesn’t matter what the label says.

The Ph. Eur. 5.1.3 antimicrobial efficacy test — widely called the challenge test or preservation efficacy test (PET) — is the method the industry uses to demonstrate this. Five test microorganisms are inoculated into the finished formula at defined concentrations:

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

For Category A products — which covers most leave-on skin preparations — the acceptance criteria require at least a 2-log reduction in bacterial counts by Day 14, with no increase through Day 28. Fungal counts must show no increase at Day 14 or Day 28. These criteria apply whether your formula contains phenoxyethanol, a glycol-and-acid blend, or nothing beyond a water-in-oil emulsifier and a cold-pressed botanical. The microbiology doesn’t adjust for brand positioning.

Where Clean-Formulated Products Actually Fail

We see a predictable set of failure modes when reviewing preservation work on clean-positioned cosmetics.

Under-assessed water activity. Anhydrous formulas — balms, oil serums, solid bars — are genuinely harder to contaminate because water activity (Aw) below 0.60 doesn’t support most microbial growth. But emulsions with natural wetting agents, micellar toners, and hydrosol-based serums all contain sufficient free water to be genuinely at risk. Aw measurement is underused; many brands assume that because a formula “feels light” it won’t support growth. That assumption has no regulatory basis.

Botanical extracts introducing contamination load. Plant-derived extracts — particularly unsterilised liquid extracts in hydroglycolic bases — carry their own microbial populations. Bacillus spores survive most preservation systems. If your brief calls for a 5% raw botanical water extract and your preservation system was designed around a low-challenge formula, you’ve changed the problem without changing the testing scope.

Non-standard challenge testing. Challenge testing is only meaningful if properly executed. The Ph. Eur. 5.1.3 protocol requires bacterial inoculation at approximately 10⁵ to 10⁶ colony-forming units per milliliter (CFU/mL). We’ve reviewed technical dossiers where the challenge test was conducted by unaccredited laboratories using inoculation concentrations well below that threshold. A test that’s easy to pass tells you nothing useful — and the Responsible Person carries the liability when it later fails in the market.

Confusing accelerated stability with microbial stability. Stability studies at 40°C and 75% relative humidity for 3 months confirm physical and chemical performance. They are not a substitute for challenge testing. The two tests answer entirely different questions. Elevated temperature can destroy a preservation system as reliably as it reveals emulsion breakdown — and the CPSR must contain both assessments, not one standing in for the other.

What a Compliant “Clean” CPSR Needs to Show

The compliance pathway for preservative-free formulations is clear, if more demanding than simply listing Annex V entries.

First, identify and document your preservation strategy explicitly. Whether it’s water activity control, an ethanol system, a combination of chelators and organic acids, or packaging engineering — it needs to appear in Part A of the CPSR as a functional antimicrobial rationale, not an absence. Your EU-qualified safety assessor needs to sign off on why this formula is microbiologically safe, not just confirm it doesn’t contain parabens.

Second, commission a properly conducted challenge test at an ISO 17025-accredited laboratory. Full Ph. Eur. 5.1.3 protocol, all five organisms, Category A criteria for leave-on products. A 28-day test is the minimum for a complete assessment. Many brands run a 14-day preliminary screen at 70–80% of target formula first — if results are marginal, there’s still room to adjust without rebuilding the entire safety dossier.

Third, account for real-world use. Leave-on products applied with fingers are re-inoculated at every use. Packaging matters here: pump dispensers with one-way valves provide meaningfully better microbiological protection than open jars. Your risk assessment under Article 10 should note this, particularly for products targeting sensitive skin or compromised skin barrier conditions.

Fourth, plan for shelf-life verification. Challenge testing at development is a snapshot. If you reformulate, change a raw material supplier, or alter any process parameter — including the water source — retest. Article 10 requires that the safety assessment reflects the product as actually manufactured, not the version from the initial launch dossier.

The Responsible Person Cannot Delegate This Liability

One thing worth being direct about: under Regulation 1223/2009, the Responsible Person (RP) established in the EU bears full legal accountability for the safety of every product they place on the market. That accountability includes microbiological safety.

If a product causes a serious adverse reaction — a contaminated moisturiser infecting a compromised skin barrier, a baby product causing an opportunistic infection — the RP faces reporting obligations under Article 23, potential market withdrawal, and action by the relevant national competent authority. In France, that’s the DGCCRF. In Germany, the relevant Landesbehörde. In Italy, the Ministry of Health.

The SCCS Guidance on the Safety Assessment of Cosmetic Products (11th revision) is explicit on this: the safety assessment must include a microbiological quality assessment and a documented justification for the preservation strategy adopted. “We didn’t add any preservatives” is not a justification. It’s the beginning of a technical question that requires a rigorous, documented answer.

A Note for Non-EU Brands Entering the European Market

We work regularly with brands from North America and Asia who are preparing EU market entry, and the clean beauty preservation question comes up in almost every engagement. The regulatory framework is the same regardless of where a product was formulated or manufactured — compliance with Regulation 1223/2009 is required to access the EU market, full stop.

The CPSR must be prepared by a qualified safety assessor recognised under EU law. The challenge test must be conducted to Ph. Eur. 5.1.3 standards, ideally at an accredited laboratory. If you’re entering through a French or German distributor acting as your Responsible Person, they will ask for this documentation. If they don’t, that’s a signal worth paying attention to.

---

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

• ISO 17025-accredited antimicrobial efficacy and preservative testing for US cosmetics — Qalitex Laboratories provides Ph. Eur.- and USP-method challenge testing for brands distributing in the US market
• Microbiological and preservative efficacy testing for Canadian cosmetics and NHPs — Androxa supports European manufacturers with Health Canada-compliant microbiological testing for Canadian market entry