The margin of safety in cosmetics is the single number that decides whether your product passes its safety assessment. Everything else in a CPSR — the toxicological profiles, the exposure estimates, the impurity data — exists to feed one calculation, and the answer needs to come out at 100 or above. If it does not, the assessor cannot conclude the product is safe, and no amount of good intentions changes that.
Brand owners rarely see the arithmetic, which is a shame, because understanding it explains almost every question a safety assessor asks: why they need exact percentages rather than ranges, why a leave-on product is harder than a rinse-off, why a body lotion is harder than a face cream, and why “it is only 0.5%” is not an argument. This guide walks through how the MoS is calculated, what drives it up and down, and what to do when it fails.
Key takeaways
- The MoS must be ≥ 100 for a substance to be considered safe in the intended use — the standard threshold in the SCCS Notes of Guidance.
- MoS = PoD ÷ SED, where the point of departure is usually a NOAEL from a repeated-dose study and SED is systemic exposure from your product.
- The factor of 100 is not arbitrary: it is 10× interspecies (animal to human) × 10× intraspecies (human variability).
- SED is driven by concentration, amount applied, dermal absorption and retention factor — a rinse-off retention factor of about 0.01 is why shampoos pass where leave-ons fail.
- MoS is calculated per substance, not per product, and only for substances that can be systemically absorbed.
- The current reference is the SCCS Notes of Guidance, 12th revision (SCCS/1647/22).
- A failing MoS is usually fixed by reducing concentration, changing product type or finding better toxicological data — not by arguing.
What the MoS actually is
The margin of safety compares the dose known to cause no adverse effect in toxicological studies with the dose a consumer actually receives from using your product. It answers a simple question: how much headroom is there between real-world exposure and the level where something starts to happen?
The calculation is:
MoS = PoD / SED
Where PoD is the point of departure — typically a NOAEL, expressed in mg per kg of body weight per day — and SED is the systemic exposure dosage from your product, in the same units. Because both sides are in the same units, the MoS is a dimensionless ratio. An MoS of 300 means the consumer is exposed to one three-hundredth of the no-effect dose.
Why 100?
The threshold is a default uncertainty factor built from two components:
- 10× for interspecies variation — the NOAEL usually comes from rats, and humans may be more sensitive.
- 10× for intraspecies variation — humans differ from each other in metabolism, skin condition, age and health.
Multiply them and you get 100. Where data are weaker or the substance is of particular concern, an assessor may apply additional factors and require a higher MoS. Where good human data exist, the default factors can sometimes be refined downward — but that requires substantial justification.
MoS is calculated per ingredient, not for the finished product. A cream with twenty ingredients might need MoS calculations for four or five of them — those with a systemic toxicity concern and meaningful absorption. Water, most emollients and many polymers do not require one. Colorants, preservatives, UV filters and actives usually do.
The point of departure
The PoD is derived from toxicological data on the substance, and its quality determines how much room you have.
The preferred source is a benchmark dose (BMD), which uses the full dose-response curve. In practice, most cosmetic ingredient dossiers rely on a NOAEL — the highest tested dose showing no adverse effect — typically from a 90-day repeated-dose oral toxicity study in rodents. Where no NOAEL is available, a LOAEL may be used with an additional uncertainty factor applied.
Where does it come from? For restricted substances, an SCCS opinion will often state a usable PoD directly. Otherwise the assessor draws on the ECHA registration dossier, published literature, CIR reviews or supplier-provided toxicological data.
The most frequent cause of a stalled CPSR is not a bad MoS — it is no usable PoD at all. If an ingredient has no repeated-dose toxicity data anywhere, the assessor cannot calculate a margin of safety and cannot conclude it is safe. This is a common problem with novel botanicals, niche peptides and small-supplier actives. Ask your supplier for the toxicological dossier before you formulate around an ingredient, not after.
Calculating the SED
The systemic exposure dosage is where your product’s specifics enter the calculation. The standard form is:
SED = (A × C/100 × DAp/100 × F) / BW
Expressed in mg/kg body weight per day, where:
| Term | Meaning | Typical value |
|---|---|---|
| A | Estimated daily amount of product applied | Product-type specific (see below) |
| C | Concentration of the substance in the product (%) | Your formula |
| DAp | Dermal absorption (%) | Measured, or 50% / 100% default |
| F | Retention factor | 1 leave-on; ~0.01 rinse-off |
| BW | Default body weight | 60 kg (adult) |
The SCCS Notes of Guidance provide standard daily application amounts so that assessments are comparable across the industry. The commonly used figures include:
| Product type | Daily amount (g) | Retention factor |
|---|---|---|
| Body lotion | 7.82 | 1 |
| Face cream | 1.54 | 1 |
| Hand cream | 2.16 | 1 |
| Deodorant (non-spray) | 1.50 | 1 |
| Lipstick | 0.057 | 1 |
| Shampoo | 10.46 | 0.01 |
| Shower gel | 18.67 | 0.01 |
| Toothpaste | 2.75 | Ingestion-based |
Notice what this table does to your formulation freedom. A body lotion applies roughly five times as much product as a face cream, over a far larger surface area, and none of it is rinsed off. That is why the same active at the same percentage can pass comfortably in a face serum and fail outright in a body lotion — a point we cover in more depth in our guide to CPSR requirements for face creams.
The retention factor is the big lever
For rinse-off products the retention factor of approximately 0.01 reflects that only about 1% of what you apply stays on the skin. That two-order-of-magnitude reduction is why shampoos and shower gels tolerate ingredients that a leave-on product cannot — see our guide to CPSR for shampoos and conditioners.
Dermal absorption matters enormously
If you have no measured dermal absorption data, the SCCS default is conservative — 50% for substances with molecular weight above 500 Da and log Pow outside the −1 to 4 range, and 100% otherwise. Using a default of 100% when the real figure might be 2% inflates your SED fiftyfold and can be the sole reason an MoS fails.
An in vitro dermal absorption study to OECD Test Guideline 428 is the fix. It costs money, but it is often far cheaper than reformulating a product around a conservative default.
A worked example
A preservative at 0.5% in a body lotion. NOAEL from a 90-day rat study: 50 mg/kg bw/day. Dermal absorption measured at 10%.
SED = (7.82 g × 0.5/100 × 10/100 × 1) / 60 kg
= (7.82 × 0.005 × 0.1) / 60
= 0.00391 / 60
= 0.0000652 g/kg bw/day = 0.0652 mg/kg bw/day
MoS = 50 / 0.0652 = 767
Comfortably above 100 — this passes.
Now change one thing. Suppose no dermal absorption study exists and the default of 100% applies. SED becomes 0.652 mg/kg bw/day and the MoS falls to 77. The same formula now fails, purely because of a missing study. This is exactly the situation where spending on an OECD 428 study is cheaper than reformulating.
Aggregate exposure is increasingly part of the picture. Where the same substance appears across several products a consumer uses daily — a preservative in the shower gel, the body lotion and the face cream — assessors may consider combined exposure rather than each product in isolation. The SCCS Notes of Guidance address aggregate exposure scenarios, and a substance that passes individually can require attention in the round.
When the MoS fails
Options, roughly in order of practicality:
Reduce the concentration. MoS scales linearly with concentration, so halving the level doubles the margin. Often the fastest route, and frequently the level was set higher than needed for function anyway.
Get better data. A dermal absorption study replacing a conservative default is the highest-leverage single intervention. A better PoD from a more recent or more relevant study can also help.
Change the product type. Moving a formula from body lotion to face cream, or from leave-on to rinse-off, changes the SED dramatically. Sometimes the right answer is a different format.
Substitute the ingredient. If the substance is fundamentally unsuitable at the level needed for function, replace it.
Restrict the intended use. Narrowing the population or use pattern — not for children, not for use on large body areas — can bring a marginal case into range, though it constrains your marketing.
What is not an option is accepting an MoS below 100 without a robust, documented scientific justification. An assessor who signs off on a margin of 60 because the brand needed the launch date is putting their professional standing and your compliance on the line. Article 3 requires the product to be safe, and the CPSR is the evidence it is — see our guide to who can sign a CPSR and what qualifications they need.
What this means for how you brief your assessor
Understanding the MoS explains why assessors are demanding about certain inputs.
They need exact percentages, not ranges, because C sits directly in the SED formula — a range of “1–3%” forces the assessor to assume 3%. They need the full quantitative composition of compound raw materials, because a preservative blend at 1% may contain an active at 0.3% that needs its own MoS. They need the real product type and use pattern, because A and F come from it. And they need impurity data, because impurities of toxicological concern get their own calculation regardless of how minor they are in the formula.
Our checklist of documents needed to order a CPSR maps directly onto these inputs — every item on it exists because something in this calculation needs it.
Bringing it all together
The margin of safety is where a CPSR stops being a document and becomes a decision. It converts toxicology and use pattern into one comparable number, and the 100 threshold is the industry-wide line between a product an assessor can sign and one they cannot.
For brand owners, the useful takeaway is that the MoS is highly sensitive to things you control early and cheaply: the concentration you set, the format you choose, and whether your supplier can produce a proper toxicological dossier. Ingredient selection made with the MoS in mind produces formulas that sail through assessment. Ingredient selection made purely on marketing appeal produces the phone call where your assessor explains that the hero active cannot go in at the level on your pack.
Lexora’s safety assessors calculate margins of safety as part of every CPSR, and will tell you early if a formula is heading for a problem rather than at the end. Explore our CPSR service, or take safety assessment, testing and notification together with the Full Compliance Pack.
Frequently asked questions
What is an acceptable margin of safety in cosmetics?
An MoS of 100 or greater is the accepted threshold under the SCCS Notes of Guidance. It comes from a default uncertainty factor of 10 for interspecies differences multiplied by 10 for variability between humans. Where data are limited or the substance is of particular concern, an assessor may require a higher margin.
Is the MoS calculated for the whole product or per ingredient?
Per ingredient. The assessor identifies substances with a systemic toxicity concern and meaningful absorption — typically preservatives, UV filters, colorants, actives and impurities of concern — and calculates a separate MoS for each. Water, most emollients and many polymers do not require one.
Why do rinse-off products pass more easily?
Because of the retention factor. For rinse-off products such as shampoo and shower gel the retention factor is approximately 0.01, meaning only about 1% of the applied product stays on the skin. That reduces the systemic exposure dosage by two orders of magnitude compared with a leave-on product, where the retention factor is 1.
What happens if there is no toxicological data for my ingredient?
The assessor cannot derive a point of departure, cannot calculate an MoS, and cannot conclude the ingredient is safe at the intended use level. The CPSR stalls. This is common with novel botanicals and small-supplier actives, so request the toxicological dossier from your supplier before you build a formula around an ingredient.
Can a dermal absorption study rescue a failing MoS?
Often, yes. Without measured data the SCCS default is 50% or 100% absorption, which is deliberately conservative. If the real figure is a few percent, an in vitro study to OECD Test Guideline 428 can raise the MoS by an order of magnitude or more — frequently cheaper and faster than reformulating.
Why does my assessor need exact percentages instead of ranges?
Because concentration sits directly in the SED formula. Given a range, the assessor must assume the highest value, which produces the most conservative and least favourable MoS. Exact percentages, including the composition of compound raw materials, give you the margin you actually have rather than the worst case.
Which SCCS guidance version applies?
The current reference is the SCCS Notes of Guidance for the Testing of Cosmetic Ingredients and their Safety Evaluation, 12th revision (SCCS/1647/22). It sets out the standard daily application amounts, retention factors, default absorption values and the MoS methodology that assessors across the EU apply.
