← All posts
compliancetechnologycosmetics

Why ingredient screening takes longer than it should

The NZ CPG schedule lists 2,166 prohibited substances and 23,334 restriction rules. IFRA adds 214 more. A manual check takes 30–45 minutes. Here's the real cost.

Any cosmetics brand selling into New Zealand navigates two standards: NZ CPG (Cosmetic Products Group Standard) and IFRA (International Fragrance Association). Both matter. Both have to be checked. And for most teams, both are checked the same way — manually, against PDFs and spreadsheets, one ingredient at a time.

The numbers

NZ CPG Schedule 4 lists 2,166 prohibited substances. These are outright bans — if a substance appears on Schedule 4, it cannot be in the product.

Schedules 5–8 add 23,334+ restriction rules. These aren’t bans. They’re concentration limits that vary by product category: the permitted level in a rinse-off product differs from a leave-on, which differs again from products intended for use around eyes or on mucous membranes. The same ingredient that passes at one concentration in one product category might fail at that same level in another.

IFRA adds 214 fragrance materials with restrictions across 12 product categories. Amendment 51, published in June 2024, is the current reference. Here too, limits are category-specific. A material restricted to a given percentage in a Category 5 body lotion has a different threshold in a rinse-off product. The IFRA category a product falls into determines which limits apply — and those categories don’t map neatly onto how marketing teams describe products.

This is the weekly reality of a regulatory affairs team doing its job.

Why it takes so long

A manual ingredient check is not just a lookup. It’s a multi-step process done for every substance in a formulation:

  1. Resolve trade names to INCI identifiers. Supplier labels don’t always use standardised nomenclature. An ingredient might be listed under a brand name, a chemical synonym, or an abbreviation that requires cross-referencing a separate trade-name database before you can even begin checking the schedule.

  2. Cross-reference each resolved substance against Schedule 4. A confirmed match means the product cannot be sold. No match means you proceed.

  3. Check Schedules 5–8 for restriction entries applicable to your specific product type. This requires knowing which CPG product category applies, finding the relevant entry, and confirming the formulation concentration is within the permitted limit.

  4. Verify the IFRA category against the product, then check each fragrance substance against Amendment 51’s restriction table for that category.

  5. Document the outcome in a form that would satisfy a regulatory inquiry.

Done carefully — accounting for the name resolution step and the category-matching — this takes 30 to 45 minutes per product. For a brand with 50 SKUs, that’s 25 to 37 hours of screening time just to establish a baseline. And that baseline has to be re-run every time a formulation changes.

The update problem

IFRA publishes one to two amendments per year. When Amendment 51 arrived in June 2024, it changed the permitted levels for a set of fragrance materials. Any formulation that had previously been screened against an earlier amendment and contained one of those substances needed re-evaluation.

For teams running manual processes, there’s no systematic way to identify which products are affected. The usual response is a manual triage: go through the list of newly restricted materials, think through which products might contain them, pull those formulations, and re-screen. That’s hours of work triggered by a standards update that affected a subset of substances.

For brands that have grown their SKU count over time, this compounds. The bigger the catalogue, the more exposure each amendment cycle creates.

The structural problem

This is not primarily a resourcing problem. Adding more regulatory affairs hours doesn’t change the underlying dynamic — it scales the cost linearly with the number of products and amendment cycles. The problem is structural: compliance work is being done with tools designed for general-purpose record-keeping rather than for regulatory cross-referencing.

Software can address the mechanical parts. Not the judgment about whether a product is acceptable — that still requires expertise. But the lookup, the cross-reference, the category matching, the documentation: these are deterministic steps that a purpose-built tool handles reliably and in seconds rather than minutes.

This is the problem we’re building toward solving with ComplianceFlow. We’ll keep writing about what we’ve learned in the process.