Cosmetic Standards & Regulations Explained for Eye Care

Key Technical Parameters #

Eye care products sit at one of the trickiest intersections in cosmetics compliance: a thin-skinned, mucosa-adjacent application zone that regulators in Brussels, Washington, and Beijing each treat differently. The challenge isn’t just knowing each market’s rules in isolation — it’s understanding where those rules conflict, which creates real SKU-level decisions about concentration caps, labelling claims, and preservative systems. Brand developers who’ve only built one market’s portfolio consistently hit delays when they try to expand because the qualification burden they assumed was universal turns out to be regional. Our eye-care formulation team spends a disproportionate amount of project time on exactly this translation problem, and what follows is how we actually navigate it.

When the Same Formula Fails Three Regulatory Reviews #

We had a client — a mid-size European brand expanding into China — bring us an eye contour gel that had sailed through CE compliance with zero issues. Sodium hyaluronate at 2%, niacinamide at 3%, a phenoxyethanol/ethylhexylglycerin preservative system at 0.85%, and a cucumber extract listed under its INCI name. Clean label, no prohibited substances, passed ISO 11930 efficacy testing at category 3. By every metric we use internally, a straightforward brief.

The NMPA review came back with three flags. First, the preservative concentration. Second, the botanical extract required a documented country of origin and safety dossier that doesn’t exist in the EU submission format. Third, the product claim — “soothes periorbital fatigue” — pushed it toward a quasi-drug classification under China’s 2021 cosmetic regulation framework, which triggers a separate review pathway entirely. The brand had budgeted eight weeks for China registration. It took eleven months.

This is the failure mode nobody documents cleanly. It isn’t that the formula was wrong. It’s that the same formula carries a different compliance burden depending on where you submit it, what you call it, and which test reports you can produce on request. Three markets. Three different interpretations of what “eye care cosmetic” means.

The Parameters That Actually Determine Your Compliance Burden #

Let’s be specific about what drives the regulatory split across the three primary export markets we formulate for: EU, US, and China.

pH and mucous membrane proximity. EU EU Cosmetics Regulation 1223/2009 sets no universal pH floor for eye area cosmetics, but SCCS opinions on specific ingredients do carry pH-dependent validity. Annex III restrictions for certain preservatives, for example, assume typical formulation pH. Go below pH 4.8 in an eye cream and you need to re-examine whether those concentration limits still apply as written — because they were assessed at a different ionisation state. We flag this in every brief that involves an acidic active system.

Preservative selection under ISO 11930. ISO 11930 preservative efficacy testing is mandatory as a supporting document in EU, expected (though not always formally required) in the US, and now practically unavoidable in China post-2021. The standard defines five categories based on reduction criteria. For eye-area products, we target category 2 or better — a ≥2 log reduction at 14 days for bacteria, no increase for yeast/mould. The issue is that many preservative systems that pass easily in a rinse-off or body product will underperform in a low-water-activity eye gel, particularly if you’re running a natural-positioned formula with a reduced preservative palette.

INCI naming and its invisible regulatory consequences. Under PCPC Guidelines, INCI names are standardised for US labelling. EU follows the same INCI convention under Annex VI/VII logic. China does not. NMPA requires Chinese INCI equivalents listed in the Chinese cosmetic ingredient database (CIRS-aligned), and not every botanical translates 1:1. We’ve had cases where an ingredient listed under one INCI name in the EU dossier required a different identifier for China submission — same molecule, different bureaucratic identity, separate safety substantiation.

The ISO 16128 natural index. Voluntary in all markets, but commercially increasingly mandatory if you’re selling to retailers that enforce clean beauty criteria. ISO 16128 gives a calculation methodology for natural and organic origin indices. For eye care, brands often request a “95% natural origin” call-out. The number is achievable in some formats but requires careful emulsifier and preservative selection — and some of the synthetic film formers that make eye serums feel elegant will pull that index below threshold fast.

FDA OTC drug status. The FDA Cosmetics Guidelines draw a clear line between cosmetics and OTC drugs. For eye care, the relevant boundary is anything with an active ingredient claim — “reduces eye puffiness” stays cosmetic, “treats periorbital oedema” does not. Most brands understand this in theory. In practice, we push back on claim language almost every project. The problem is marketing teams write their claim language before regulatory review, and sometimes they’ve already printed sales materials.

Here’s where the three markets diverge most sharply in our experience:

The most commonly overlooked parameter, in our observation across several hundred briefs, is the IECIC positive list requirement for China. Brands familiar with the EU model assume that any ingredient not on a prohibited list is permissible. China works on a dual-list logic: you need to confirm the ingredient is either on the permitted (positive) list or has gone through a new ingredient registration. Several popular peptides used in eye care — some well-established in EU and US markets — are not yet in the IECIC database. This effectively blocks them from China SKUs until a new ingredient application is filed, which currently takes 12 to 18 months.

The Conditional Logic of Multi-Market Eye Care Development #

If you’re building a single global SKU, the formulation constraints stack on top of each other, and you end up designing to the most restrictive intersection. In practice, that usually means EU + China as the tightest combined target, with US falling naturally within those limits.

If your preservative system needs to pass EU Annex V limits and appear on the IECIC positive list, you’re working with a relatively constrained palette: phenoxyethanol ≤1.0%, ethylhexylglycerin as booster (not restricted but not a standalone), sodium benzoate acceptable in acidic systems, and a few others. We’ve found that a dual-preservative system at sub-threshold concentrations — phenoxyethanol at 0.7% combined with ethylhexylglycerin at 0.3% — reliably passes ISO 11930 category 2 for eye gels with a water activity above 0.92. Below that, you usually need to adjust.

If you’re launching EU-only first and planning China expansion later, the calculation changes. You can use actives that aren’t yet in IECIC and flag them early for new ingredient registration while the EU SKU is in market. But this only works if you’ve budgeted the registration timeline and cost. One peptide our formulation team tracks — palmitoyl tripeptide-38, commonly used in eye contour creams — had a roughly 14-month new ingredient registration cycle in China as of our last project using it. Brands that hadn’t planned for this ended up reformulating for China, which means two stability packages, two PIF dossiers, and two batch validation runs. The cost delta adds up quickly.

If your brief includes a claim that touches drug territory — and we see this most often with vascular-targeting actives for dark circles — the entire regulatory category shifts in China. A product positioned as “correcting” dark circles through a physiological mechanism can be reviewed as a special-category cosmetic, not a general cosmetic, with a longer and more expensive registration pathway. Our advice on this is direct: build the claim language after the regulatory category is confirmed, not before. Marketing teams resist this. But we’ve seen enough delayed launches to know which choice costs more.

One specific recommendation that’s non-obvious: for any eye care product going to both EU and China, run your clinical or consumer perception study to ICH Stability Guidelines accelerated stability conditions (40°C/75% RH, 6 months) and use that data in both dossiers. The EU PIF and NMPA technical package both accept ICH-format stability data, and generating one dataset that serves both markets avoids a redundant testing cycle.

A 2022 open-label, observer-blinded study (n=54, 8 weeks) on a peptide-containing eye serum formulation found a 28% reduction in measured crow’s foot wrinkle depth by profilometry, with 91% of subjects reporting reduced puffiness by self-assessment. We use this study type structure as a model for efficacy substantiation that supports both EU claim documentation and NMPA technical dossier requirements simultaneously. Single-country clinical data rarely transfers as efficiently.

For brands building a brightening-whitening eye contour range, the additional complication is that skin-lightening actives carry their own restricted lists. Alpha-arbutin, for example, is permitted in face products in the EU at up to 2% but requires SCCS opinion review for eye-area application specifically. We are still waiting on clearer SCCS Scientific Opinion guidance on periorbital application limits for several brightening actives. Our current conservative position: treat the eye area as a higher-sensitivity zone for any pigmentation-targeting active and apply a 50% safety margin relative to face-product permitted concentrations until formal opinion is published. Not everyone does this. We’ve decided it’s the right call for now, even if it limits claim strength.

Formulation Notes for Brand Partners #

When you brief us on an eye care product, the first thing we need to know is your lead market and your secondary expansion markets — not your ingredient wishlist. Market sequencing changes the entire formulation architecture.

The most consistent mistake we see: brands arrive with a claims brief that was written by marketing before any regulatory category review. “Reduces dark circles by 40%” sounds like a consumer study result. In China, depending on how the mechanism is framed, it can read as a drug claim. We spend time in the first briefing call rewriting claim language into market-appropriate framing. It’s a better use of time than reformulating later.

We also need your packaging format confirmed early. Eye care products in contact with mucosa — gel patches left on the orbital area, for example — can require additional safety substantiation in some markets. The container material matters too; certain preservative systems are incompatible with specific tube liners, and we track this internally under our M-07 packaging compatibility protocol.

Lab samples typically take 2 to 3 weeks from confirmed brief. Accelerated stability runs 4 to 8 weeks at 40°C/75% RH per ICH conditions. Real-time 24-month stability is initiated concurrently. For NMPA registration, plan a separate 12 to 18 weeks for dossier preparation on top of that, longer if any ingredient requires new substance registration.

Frequently Asked Questions #

We’re launching in the EU first. Do we need to do anything differently if China is on the roadmap in 18 months?

A: Flag the China expansion now, before we finalise the formula. Several actives common in EU eye care — certain peptides and some botanical extracts — require new ingredient registration in China that takes 12 to 18 months. If we know China is coming, we can make ingredient substitutions early that cost nothing. Finding out at month 15 means a reformulation and a second stability package.

Does phenoxyethanol have a different limit for eye products versus face products in the EU?

A: Under EU Cosmetics Regulation 1223/2009, the cap is 1.0% across all product types. There’s no separate eye-area limit in Annex V. What changes is the safety substantiation expectation — the SCCS has flagged periorbital application as a higher-sensitivity route, so the safety assessor reviewing your PIF will scrutinise a phenoxyethanol-only system more carefully for an eye cream than for a body lotion.

We had an ISO 11930 test done and the product passed. Is that enough for all three markets?

A: For EU, a passing ISO 11930 result is required documentation in the PIF and largely sufficient. For the US, there’s no formal mandate under FDA Cosmetics Guidelines, but any retailer or liability review will expect it. For China, the NMPA technical dossier requires PET data, and they want the full raw report, not just a pass/fail summary. One thing we see consistently: labs that report category 3 passes — no increase for yeast/mould, but less than 2 log reduction for bacteria at 14 days — get challenged on China submissions. If your system is at the borderline, we’d re-examine the formula before submitting.

What’s your minimum order quantity and how long does a typical project take end to end?

A: MOQ for eye care products is typically 3,000 units for tubes and 5,000 units for jar formats at standard fill weights. From confirmed brief to first production batch, a straightforward single-market project runs 16 to 20 weeks including accelerated stability. Multi-market projects with NMPA registration should be scoped at 12 to 18 months for the China pathway running in parallel.

We want to call out ISO 16128 natural origin on the pack. Can we calculate that ourselves?

A: You can, and some brands do, but we’d want to run our own calculation before it goes on pack. The methodology under ISO 16128 has specific rules around water, minerals, and processed derivatives that produce surprising results for formulators who apply it intuitively. We’ve seen brands claim 95% natural origin on a formula that calculated to 78% under the correct methodology. That’s a claim substantiation problem waiting to be found by a retailer audit or a competitor challenge.

---

Have a product concept in mind? Contact our formulation team to request a complimentary brief review.