Centella Asiatica in Korean Skincare: Madecassoside Sourcing for Indie Brands (2026)

By the ALTA MEET editorial team | K-beauty ODM consulting

If you are about to brief a Korean ODM on a sensitive-skin hero product, the ingredient on every short list is Centella Asiatica. K-beauty calls it "cica" and indie founders pitch it on PDPs as the ingredient that "calms" redness, "rebuilds" the barrier, and "rescues" reactive skin. The biology behind those words is more specific than the marketing suggests, and the cost difference between a real centella formula and a window-dressing centella formula is the kind of number indie founders should know before they sign a PO.

This is the working guide we hand to ALTA MEET clients before their first centella product brief: what is actually inside a centella extract, what concentration ranges have clinical support, what Korean ODMs typically supply versus charge a premium for, what claim line MoCRA and KFDA permit, and what 1,000-unit indie MOQ economics actually look like in 2026.

The Centella Plant And Its Four Active Triterpenes

Centella Asiatica is a small perennial plant in the Apiaceae family, native to South and Southeast Asia, the Pacific Islands, and Madagascar. Korean ODMs source most cosmetic-grade material from Madagascar, Indonesia, and increasingly contract farms in Jeju and Jeollanam-do province. The plant itself is not the active. The actives are four pentacyclic triterpenes concentrated in the leaves: madecassoside, asiaticoside, madecassic acid, and asiatic acid.

The two glycosides, madecassoside and asiaticoside, are the polar, water-loving forms and account for the bulk of the leaf's bioactive content. They are heat-stable, water-soluble, and easier to formulate into serums and essences. The two aglycones, madecassic acid and asiatic acid, are the hydrophobic forms produced when the glycosides are hydrolyzed; they sit better in oil phases and emulsion creams and have somewhat different biological behavior.

When a Korean ODM line sheet lists "Centella Asiatica Extract", it can mean any of the following: a glycerin-water leaf macerate (1–10% extract, but only fractional triterpene content), a butylene glycol or 1,3-propanediol-extracted concentrate (higher actives, typically 0.5–5%), a standardized titrated extract with declared triterpene content, or TECA (Titrated Extract of Centella Asiatica). TECA is a pharmaceutical-grade reconstituted blend of three or four triterpenes at defined ratios; the French dermatology brand that popularized it in Europe used a 40:40:20 ratio of asiaticoside : asiatic acid : madecassic acid, and that ratio still anchors many cosmetic claims in Korean ODM datasheets.

A founder who briefs a hero product as "centella serum" without specifying the triterpene fraction is asking the ODM to choose for them. The choice is rarely random; it follows what the ODM has in inventory and what raises margin.

Clinical Evidence: What The Molecules Actually Do

The peer-reviewed literature on centella and its triterpenes is unusually rich for a cosmetic ingredient because the molecules have a pharmaceutical history. TECA has been used in Europe for decades as an adjunct wound-healing therapy and as a treatment for venous insufficiency. That regulatory pedigree is the reason ODMs feel confident leaning on barrier-repair language for centella formulas in a way they cannot for, say, snail mucin.

The mechanisms are reasonably well characterized. Asiaticoside and madecassoside stimulate type I and type III collagen synthesis in dermal fibroblasts in vitro, with measurable effects on extracellular matrix deposition (Maquart et al., Connective Tissue Research, 1990; replicated multiple times since). They suppress TNF-α, IL-6, and IL-1β production in keratinocytes exposed to UVB and other inflammatory triggers, which is the bench evidence behind the "calming" claim. A 2025 paper in Frontiers in Pharmacology documented that centella extract reduces macrophage-driven inflammation in diabetic wound healing through the AKT/MAPK/NF-κB pathway, which is the same pathway implicated in chronic facial redness.

On the clinical side, the most cited cosmetic study is a six-month randomized double-blind trial that combined 5% vitamin C with 0.1% madecassoside on photo-aged facial skin. The trial reported measurable improvement in deep wrinkles, fine lines, suppleness, firmness, roughness, and hydration versus a vitamin C only arm (Haftek et al., Experimental Dermatology, 2008; subsequently cited in the Journal of the American Academy of Dermatology). The takeaway that matters for indie founders: at 0.1% madecassoside, paired with another active, with six months of compliant use, the photo-aging effect is real but modest.

A 2024 systematic review in PMC on topical centella in wound healing concluded that centella-containing formulations accelerate epithelialization and reduce scar pigmentation, with the strongest evidence for TECA and standardized madecassoside extracts at 1–2% of total formula weight. Below 0.1% the clinical signal disappears in most studies. Above roughly 3%, the marginal benefit plateaus while sensitization risk in reactive skin rises, particularly in formulas lacking pH control or antioxidant co-actives.

This is the working range Korean ODMs quote against: a "low" centella formula at 0.05–0.2% total triterpenes (mostly marketing), a "real" centella formula at 0.5–2% total triterpenes (mid-tier indie hero), and a "premium" centella formula at 2–3% total triterpenes with declared standardization (the kind of formula Skin1004 and Dr. Jart+ Cicapair use as anchors). The cost gap between low and premium is roughly 4–7x on the active line alone.

The Korean ODM Formulation Reality

Korean ODMs have an outsized library of centella materials because the ingredient drove the first wave of K-beauty barrier-repair products in 2014–2017. The CICA category in Korea today is roughly 18% of the sensitive-skin segment by retail value, and major ODMs maintain three to five centella concentrates in standing inventory for fast quoting.

When a Korean ODM sends a quote referencing "Centella Asiatica Leaf Extract" without a triterpene spec, ask for three things: the extraction solvent (water, BG, 1,3-PD, ethanol, supercritical CO₂), the triterpene assay by HPLC (madecassoside + asiaticoside + madecassic acid + asiatic acid as % of dry weight or as % of finished raw material), and the certificate of analysis from the most recent batch. The legitimate Korean suppliers, including the major centella raw-material houses in Hwaseong, Cheongju, and Jeju, will send all three on request. If they cannot, the material is most likely a low-actives macerate dressed up as a hero ingredient.

The pH window matters more for centella than for most actives. Madecassoside and asiaticoside are stable across pH 4.5–7.0 but the glycoside bond can slowly hydrolyze below pH 4. Above pH 7 the molecules are stable but the typical fragrance and preservative system shifts in ways that affect sensory feel. Most Korean ODM centella serums sit at pH 5.5–6.5, which is also the typical resting pH of the stratum corneum.

Stability programming under ICH Q1A(R2) is non-optional for any indie founder shipping to the US under MoCRA. Accelerated stability at 40°C and 75% relative humidity for three months, plus real-time stability at 25°C and 60% relative humidity for at least six months, are the working minimums. A reputable Korean ODM will run pH, viscosity, color, odor, microbiology, and HPLC triterpene assay at T0, T1 month, T2, T3, and T6. Below 90% triterpene retention at T3 accelerated is a yellow flag; below 80% means the system is not shelf-stable in a US e-commerce supply chain that may sit in a warehouse for months.

Microbiology limits follow ISO 17516:2014: 1,000 CFU/g aerobic count for general leave-on products, 100 CFU/g for products intended for the eye area, pediatric use, or mucosal contact. Preservative efficacy testing (ISO 11930) is the supporting test that confirms the system survives a six-week microbial challenge.

The most underweighted formulation variable is texture pairing. Centella serums anchored on madecassoside alone often feel watery and slip off the skin. Korean ODMs typically pair the triterpenes with a humectant ladder (panthenol 2–5%, glycerin 3–7%, multi-molecular-weight hyaluronic acid 0.5–1%, sometimes polyglutamic acid 0.1–0.3%), a barrier-friendly emollient (squalane, jojoba, or fermented oils at 1–3%), and a calming co-active (allantoin, bisabolol, niacinamide 2–4%, or madecassic acid at 0.1%). The texture is what makes the product feel like a hero rather than a watery toner with marketing on the label.

Founder Note

I'm Liz, I run ALTA MEET out of Manhattan, NYC, with a Seoul-side bench that includes the same Hwaseong and Jeju centella suppliers most major K-beauty brands work with. The single biggest centella mistake I see indie founders make is briefing the product on Pinterest aesthetics rather than triterpene economics: a beautiful green-bottle "Cica Serum" with 0.05% leaf extract is the formula your ODM will quote you cheapest and your repeat-purchase rate will tell you the truth six months later. If you want a quick gut-check on a centella brief before you send it to Seoul, grab 15 minutes free with me. No deck required, just bring the line sheet.

The MoCRA And KFDA Claim Lines

Centella sits comfortably on the cosmetic side of the MoCRA cosmetic-versus-drug line in the US, but the claim language has to follow. Permitted cosmetic claims under MoCRA 2022 include hydration, softness, smoothness, soothing, comforting, calming the look of redness, and improvement in the appearance of fine lines, texture, or radiance. Off-limits cosmetic claims (which would push the product into OTC drug territory and out of MoCRA's framework) include treating eczema, treating rosacea, healing wounds, anti-inflammatory action, repairing damaged skin, or any disease claim.

The phrase "barrier-repair" sits in a gray zone. The FDA has not formally challenged the cosmetic use of "supports the skin's natural barrier function" or "helps reinforce the appearance of the skin barrier", and most major US-shipping K-beauty brands use language in that family. The phrase "barrier repair" without qualification, especially when paired with a wound-healing visual or before-and-after photo of clinical skin damage, is the kind of language that triggers FDA warning letter risk. A reputable Korean ODM legal review (which most Tier 1 ODMs offer as part of the package above 5,000 MOQ) will flag this before the artwork goes to print.

On the Korean side, KFDA does not currently list centella as a "functional cosmetic" ingredient in the same way it lists adenosine, niacinamide, or arbutin. This matters for two reasons. First, a Korean ODM cannot claim a specific anti-aging or whitening function on the Korean label without an additional dossier, which most cosmetic-tier centella products do not bother to pursue. Second, the Korean export label and the US label can diverge legitimately on claims, and an indie founder selling on Amazon US should know that the Korean inner-domestic version of a product they admire may carry different (and sometimes weaker) language for KFDA reasons rather than formulation reasons.

For EU exports, the relevant framework is Regulation (EC) 1223/2009 and the EU CPNP notification system. Centella triterpenes are permitted as cosmetic ingredients without concentration restriction, but the Cosmetic Product Safety Report (CPSR, Annex I) must address sensitization risk, especially for products marketed to sensitive skin. The Cosmetics Europe technical guidance recommends a patch test on at least 50 volunteers before mass shipment for any centella formula above 1% total triterpenes claiming "sensitive skin" suitability.

Cost Structure At 1,000-3,000 MOQ

This is the part indie founders most need numbers on, and where vague ODM quotes can quietly cost a brand 20–30% on landed cost without anyone noticing.

At 1,000 unit MOQ for a 30 mL glass-dropper serum, expect roughly the following cost ranges from a Tier 2 Korean ODM (good quality, mid-tier service, no celebrity-tier markup) in 2026 US dollars:

A baseline centella serum at 0.05–0.2% total triterpenes (water-glycerin macerate as the active) runs about USD 1.40–2.20 per finished unit before tariffs. Most of that is packaging and labeling; the active is well under USD 0.10 per unit.

A mid-tier centella serum at 0.5–1.5% total triterpenes with declared HPLC assay runs about USD 2.40–3.80 per finished unit. The active jumps to roughly USD 0.30–0.60 per unit. The full humectant and emollient ladder pushes formula cost up another USD 0.30–0.50.

A premium centella serum at 1.5–3% total triterpenes with TECA-style standardization and pharmaceutical-grade raw material runs about USD 3.90–5.80 per finished unit. The active alone is USD 0.80–1.40, and the stability-testing program (full ICH Q1A(R2) plus PET plus T3 and T6 HPLC retest) adds roughly USD 0.20–0.30 amortized across the batch.

At 3,000 units, expect about 12–18% reduction on per-unit cost; at 5,000 units, about 18–25%. Below 1,000 units a Tier 2 ODM is unlikely to engage unless the founder is buying through a NYC-based broker or consultancy that aggregates volume across clients.

Packaging is the silent cost. A 30 mL frosted glass dropper bottle with custom Pantone, a custom secondary box with FSC certification, and an insert card typically runs USD 1.10–1.80 per unit at 1,000 MOQ, dropping to USD 0.70–1.10 at 5,000 MOQ. Founders briefing for a "high-end" feel often spend more on packaging than on the active, which is fine as a business decision but should be intentional.

US landed cost should also account for the 15% reciprocal tariff on Korean cosmetic imports introduced under the US-Korea Strategic Trade and Investment Deal (November 2025), end-of-de-minimis duty for shipments under USD 800 (effective August 29, 2025), MoCRA Responsible Person fees if the founder has not registered themselves, and FDA-compliant labeling cost. Our internal benchmark: add roughly 22–28% to the FOB Incheon cost to estimate the landed cost into a US 3PL warehouse on the East Coast.

Five Formulation Mistakes Indie Founders Make With Centella

The first is briefing the product as "centella serum" without specifying triterpene assay. The ODM will quote whatever is on the shelf, and the founder will not realize until repeat-purchase data comes in that the product underperforms its visual story.

The second is stacking centella with low-pH actives in the same product. Centella triterpenes are stable down to pH 4.5 but degrade slowly below that. Pairing a centella serum with a 10% glycolic acid toner is fine; pairing centella, niacinamide, and a 10% ascorbic acid (which drops pH to 3.0–3.5) in a single bottle is asking for batch-to-batch variability and an unstable product on shelf.

The third is over-fragrancing. The centella story is "sensitive skin", and the typical fragrance load Korean ODMs add by default (0.3–0.8% fragrance for sensory feel) is meaningful sensitization risk in reactive skin. The compliant move is fragrance-free or trace fragrance under 0.1% with full IFRA disclosure.

The fourth is claiming "wound healing" on the label or in marketing copy. Even if the underlying TECA data supports the claim mechanistically, MoCRA classifies that language as a drug claim. Brands that have made this mistake have received FDA warning letters within 90 days of US e-commerce launch.

The fifth is treating centella as a single-active hero. The clinical effect at 0.1% madecassoside takes six months in compliant users to show up; consumers stop using a product on roughly day 18 if they do not feel something. The Korean ODM playbook pairs centella with at least one immediate-feel co-active (panthenol or polyglutamic acid for hydration, niacinamide for radiance, allantoin for instant comfort) so the consumer feels something on day one and stays on the product long enough for the centella effect to compound.

How To Brief Your Korean ODM On A Centella Product

The brief that gets a clean quote from a Tier 1 or Tier 2 Korean ODM has six lines: product format and volume (e.g., 30 mL glass-dropper serum), target triterpene assay (e.g., 1.5% total triterpenes minimum by HPLC, with madecassoside ≥0.5%), pH target (5.5–6.5), preservative system (paraben-free, EU-compliant, ISO 17516:2014 leave-on limits), claim line (cosmetic, MoCRA-friendly, "calms the look of redness" rather than "treats redness"), and MOQ with stability program (1,000 MOQ with full ICH Q1A(R2) accelerated + real-time + ISO 11930 PET).

A Korean ODM that pushes back on the HPLC assay specification or refuses to put the triterpene number on the spec sheet is telling you they want to use whatever leaf extract is cheapest that week. A Korean ODM that asks clarifying questions about claim line and US distribution is the kind of partner who will keep your brand out of MoCRA trouble.

For a full step-by-step ODM brief template, see our Korean ODM brief template guide. For more on what a real Korean ODM quote should contain and how to read it line by line, see our Korean ODM quote breakdown. For landed-cost math after the 2025 tariff reset, see our US tariff guide. For MoCRA registration and FDA labeling, see our FDA Korean skincare import guide.

Key Takeaways

Centella Asiatica is not one ingredient; it is a family of four triterpenes with different solubility, stability, and bench evidence. The cost difference between a low-actives centella formula and a clinically meaningful one is 4–7x on the active line and 1.5–2x on finished unit cost at 1,000 MOQ.

Briefing your Korean ODM with a triterpene assay specification (total triterpenes ≥1.5%, madecassoside ≥0.5%) and demanding HPLC confirmation on every batch is the single most important quality-control move for a centella-hero brand.

MoCRA permits a wide cosmetic claim line for centella but disallows wound-healing, anti-inflammatory, and disease-treatment language. A reputable Korean ODM legal review will catch this before artwork goes to print.

Landed cost into a US 3PL warehouse after November 2025's 15% reciprocal tariff is roughly FOB Incheon plus 22–28%.

The Korean ODM playbook pairs centella with at least one immediate-feel co-active (panthenol, polyglutamic acid, niacinamide, or allantoin) so the consumer feels something on day one and stays on the product long enough for the centella effect to compound.

FAQ

Is "Cica" the same as Centella Asiatica?

Yes. "Cica" is a marketing shorthand that became popular in Korean skincare in 2014–2017 and now refers to any product anchored on Centella Asiatica extract or its isolated triterpenes. The word "cica" comes from "cicatrization", the medical term for wound-healing scar formation, which references TECA's pharmaceutical history in Europe.

What is the difference between madecassoside and asiaticoside?

Both are glycoside triterpenes from Centella Asiatica. Madecassoside is more abundant in most Centella varieties and has stronger bench evidence for anti-inflammatory and UV-protective effects. Asiaticoside has stronger evidence for collagen stimulation and wound-healing matrix deposition. Most Korean ODM premium centella formulas contain both, with madecassoside typically dominant by mass.

What concentration of madecassoside should I ask for in a serum?

For a clinically meaningful effect within a reasonable use window, 0.1% madecassoside is the floor cited in the most-referenced six-month vitamin C plus madecassoside trial. For a "real" Korean ODM hero formula, 0.5–1% madecassoside is the working range, paired with asiaticoside and at least one immediate-feel co-active. Above 2.5–3% total triterpenes the marginal benefit plateaus and sensitization risk in reactive skin rises.

Can I claim "barrier repair" on my centella product in the US?

The phrase "supports the skin's natural barrier function" or "helps reinforce the appearance of the skin barrier" is widely used by major brands and has not been formally challenged by the FDA. The phrase "barrier repair" without qualification, especially when paired with clinical-damage imagery, is FDA warning-letter risk. Get your Korean ODM legal review or a US cosmetic regulatory attorney to confirm the exact wording for your artwork before mass production.

Is Korean centella different from Indian or Madagascar centella?

Botanically the same species (Centella Asiatica L. Urban), but the triterpene profile varies by climate, soil, harvest age, and post-harvest processing. Madagascar-sourced material has historically been the reference for TECA-grade extract because of consistent triterpene ratios. Korean contract-farmed centella from Jeju, particularly newer cultivars like the BT-Care variety documented in a 2025 paper, is now meeting or exceeding Madagascar standards for premium cosmetic applications.

What's the minimum MOQ for a Korean ODM centella product?

A Tier 2 Korean ODM will typically quote from 1,000 units for a stock formula with custom packaging, or 3,000 units for a custom formula. Below 1,000 units indie founders usually have to go through a NYC-based broker or consultancy that aggregates volume across clients, or use a private-label catalog product with minimal customization.

How long does a centella product brief-to-launch timeline look like?

For a stock-formula private label with custom packaging, plan on 10–14 weeks from signed PO to FOB Incheon. For a custom formula with full ICH Q1A(R2) stability program and ISO 11930 PET, plan on 5–7 months. Add 3–4 weeks for ocean freight to East Coast 3PL and 2 weeks for US MoCRA compliance review and FDA labeling QA before launch.

Working With ALTA MEET

ALTA MEET is a boutique NYC × Seoul consultancy that partners indie K-beauty founders with vetted Korean ODMs across the centella, peptide, retinoid, exosome, and broader sensitive-skin categories. We sit between the founder and the factory: we write the brief, read the quote line by line, validate the lab sample against the spec, and shepherd the launch through MoCRA, KFDA, and EU CPNP compliance.

If you are about to brief your first centella product or you suspect your current centella formula is under-spec, book a free 15-min K-Beauty manufacturing gut-check with Liz. Two questions before the call route you to the right conversation; no deck required.

Reviewed for accuracy by ALTA MEET's formulation consulting team. Sources: Maquart F-X et al., Connective Tissue Research, 1990; Haftek M et al., Experimental Dermatology, 2008; PMC systematic review on topical centella in wound healing (PMC11510310, 2024); Frontiers in Pharmacology, 2025 diabetic wound healing pathway study; EMA Assessment Report on Centella Asiatica L. Urban; Cosmetics Europe technical guidance on sensitive-skin cosmetics; ISO 17516:2014; ISO 11930; ICH Q1A(R2) ICH Q1B; MoCRA 2022; EU Regulation (EC) 1223/2009; KFDA Functional Cosmetic framework; US-Korea Strategic Trade and Investment Deal (November 2025); BT-Care Centella variety paper, PMC12283260, 2025.