Inside a Korean OEM Contract: What Indie Founders Sign in 2026

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

An indie K-beauty founder in 2026 spends months finding the right Korean partner, and then, at the moment of signing, is asked to sign a document that decides who owns the formula, how many units must ship every year, what happens if a shade discontinues, and whether the brand can move to another factory if the relationship breaks. Most first-time founders sign it in a single afternoon because they are worn down by the search. The contract is where the actual economics of the brand live, not the marketing deck, and it is where most of the disputes we see later trace back to.

This post walks through what is usually in a Korean OEM contract in 2026, what the different structures mean for cash, and what specific clauses tend to matter a year or two after signing. It is written for US indie founders in the seed-to-Series-A range who are working directly with a Korean OEM or ODM for the first time.

Why OEM contract literacy suddenly matters in 2026

Two things changed in the last twelve months that made contract literacy more important than it used to be.

First, the Korean ODM and OEM sector consolidated. Cosmax, Kolmar, Cosmecca, and COSON are still the four biggest names, but a second tier of specialists (Englewood Lab for gel and serum runs, HANACOS for stick and balm, iM1NE for sun) opened US-facing sales desks in 2025 and 2026 and started competing for indie business more openly. That competition raised the sophistication of the contracts. In 2023 an indie founder could still get a two-page purchase agreement that skipped most of the IP and MOQ language. In 2026 the paper templates that even mid-tier ODMs run are ten to eighteen pages, and they read like a mid-market supplier agreement.

Second, US MoCRA came into force in stages between December 2023 and July 2024, and the FDA started issuing warning letters at a slow but steady cadence in 2025. Facility registration, product listing, and responsible-person obligations moved from a soft ask to a hard line item that either the OEM or the brand must sign for. A 2026 contract has to say who does that work and who pays for it. If the paper is silent, the risk sits with the brand.

The result: the "sign it fast and figure it out later" era is over for indie K-beauty. What was true in 2022 (paper is thin, relationships carry weight) is still culturally true in 2026, but the paper now allocates real regulatory exposure, so the paper needs to be read.

The four contract structures Korean OEMs write in 2026

There are four common structures. They are not always called by these names in Korean paperwork (KOR-language documents lean on the terms 위탁생산 for OEM contract manufacturing and 자체개발 for in-house development), but they map cleanly to how US indie founders think about the deal.

1. Fixed-cost per-unit OEM

This is the simplest structure and the most common one an indie founder sees first. The OEM quotes a fixed per-unit price at a given MOQ tier (for example, $2.10 per unit at 3,000 units, dropping to $1.75 at 10,000 units), the brand pays a deposit (usually 30 to 50 percent), and the balance is due either at shipment or 30 days after ship-out. The formula is owned by the OEM, or the OEM owns a base and the brand owns any custom modifications the OEM developed for that project.

Fixed-cost contracts are what you sign when you are buying a "white-label plus small tweak" product. The OEM pulls a base gel out of its library, swaps in your requested humectant blend, matches your color and fragrance target, and prints your label. The paper is short and the cash timing is predictable. The downside is that you cannot easily move the formula to a second factory because you do not own the master formula and the base is proprietary to the OEM.

2. Cost-plus OEM

The contract exposes the actual raw-material cost line, the labor and overhead allocation, and a fixed margin. The brand pays for what is used plus a defined markup, typically 15 to 25 percent on materials and 100 to 200 percent on direct labor and overhead. The invoice looks like a construction change order rather than a purchase price.

You see this on more technical products (spicule ampoules, high-percent actives, prescription-adjacent skincare) where raw-material prices are volatile and both sides want to avoid renegotiation every batch. It also shows up when the brand is expected to hit fast growth and the OEM wants a shared understanding of margin as scale changes the cost curve. Cost-plus takes more accounting attention from the brand: monthly reconciliation, receipts on ingredient batches, and a documented process for handling wastage claims.

3. Hybrid (fixed base plus pass-through)

The most common contract we see in 2026 for indie K-beauty. The OEM quotes a fixed per-unit base price for the formula, packaging, and standard testing, and then passes through specific line items that vary batch to batch: custom actives above a threshold percentage, imported components (glass droppers from Italy or France, decorative caps sourced by the brand), specialty stability testing, and MoCRA compliance work. The hybrid protects both parties from raw-material shocks while keeping the core pricing predictable.

Hybrid contracts require the brand to read the list of "pass-through" items carefully. A common failure mode: the base price looks competitive, then the brand receives an invoice with $0.28 per unit of "MoCRA registration surcharge" that was buried in an appendix and the founder did not model in the launch P&L. Every pass-through category should be named in the paper.

4. Full-custom (brand-owned formula)

The most expensive structure and the one that gives the founder the most control. The OEM develops a new formula for the brand from scratch, the brand pays a development fee (typically $8,000 to $35,000 in 2026 for a Korean full-custom serum, higher for actives that require chelation, encapsulation, or specialty preservation), and the brand owns the resulting formula. The OEM produces at an agreed per-unit price with an MOQ tier structure.

Two clauses matter here. First: the formula transfer clause, which decides whether the brand can move the formula to a second manufacturer (dual sourcing) or a US-based facility if trade conditions change. Second: the exclusivity clause, which decides whether the OEM can sell the same or a substantially similar formula to another brand. Full-custom is the only structure that gives you either of these levers, and even then only if the paper explicitly says so.

What is actually inside the paper

Regardless of structure, a Korean OEM contract in 2026 will typically contain nine to twelve specific clauses that decide the economics of the brand a year later. The list below reflects the paper we see across roughly fifty US indie brands ALTA MEET has helped structure between 2022 and 2026.

Formula ownership and IP

The single most important clause. In the Korean market it is normal for the OEM to own the base formula and the brand to own the label. If the brand paid a development fee to modify or create a formula, the paper needs to say explicitly whether the modification transfers to the brand, whether the underlying base is licensed, and whether the brand can commission a second manufacturer to reproduce the modification. Silence defaults to the OEM owning everything.

A useful test: if the contract does not distinguish between "base formula," "modifications," and "custom development," the brand does not own anything portable.

MOQ and reorder MOQ

The initial MOQ is easy to read. The clause most founders miss is the reorder MOQ, which is often lower than the initial MOQ (a 3,000-unit initial run may allow 1,500-unit reorders) but sometimes higher if the OEM's line schedule prices smaller runs punitively. In 2026 we see reorder MOQ set at 40 to 75 percent of initial MOQ across mid-tier Korean OEMs.

The other MOQ trap: annual minimum volume. Some contracts require the brand to purchase a stated minimum quantity per calendar year (for example, 15,000 units of the primary SKU per year) to keep the pricing and formula rights. If the brand misses the annual minimum, the pricing steps up or the formula rights lapse to the OEM. This is common on full-custom and cost-plus contracts and rare on fixed-cost.

Payment terms

Standard 2026 payment terms for a US indie brand buying from a Korean OEM: 30 to 50 percent deposit at PO, 50 to 70 percent balance either at shipment or net 30 from ship-out. Larger brands negotiate net 60 or net 90 but this is rare for a first-time indie. If the contract asks for 100 percent payment before production starts, that is a signal the OEM has treated you as a higher credit risk than normal, and it is worth asking directly why.

Wire routing usually goes to a Korean corporate account. FX exposure sits with the brand unless the contract prices in USD (some Korean OEMs will do this at a small premium of 2 to 4 percent to absorb the risk).

Term and termination

The contract term is typically one to three years for indie contracts and three to five years for larger brands. Termination clauses are the second most-overlooked part of the paper (after IP). Look for:

• Notice period for either party to terminate (60 to 180 days is standard)
• Whether the OEM can terminate for missing annual minimums
• Whether the brand can terminate if quality events (batch failures, stability failures, contamination) exceed a threshold
• What happens to inventory and work-in-progress on termination (usually the brand pays for finished goods and raw materials at cost)

A well-drafted termination clause protects the brand's downside if the OEM has capacity problems, a quality event, or a change in ownership.

Exclusivity

Real exclusivity is expensive. If the brand wants the OEM to promise it will not sell the same or substantially similar formula to another brand, the OEM will normally require either a higher per-unit price (10 to 30 percent premium) or an annual volume commitment that makes the exclusivity economically worth the OEM's while. Silent exclusivity (the OEM says "we won't do that but we won't put it in the contract") is not exclusivity. If it matters, it needs to be in writing with a definition of "substantially similar."

Region-limited exclusivity (for example, US-only) is a middle ground and is often the practical compromise for indie brands. Category-limited exclusivity (the OEM won't sell the formula for the same product category) is another middle ground.

Quality and stability specifications

The paper should reference specific standards: ISO 22716 (cosmetic GMP), ISO 17516 (microbiological limits: total aerobic <100 to <1,000 CFU/g depending on product category), and ICH Q1A(R2) for stability testing methodology. If the paper only says "manufactured to industry standards" without naming a standard, the brand has no lever if a batch fails outside the OEM's private definition of the specification.

The stability specification also names how long the OEM guarantees the product, typically 24 or 30 months from manufacturing date for standard Korean skincare, and 18 months for products with fragile actives like vitamin C or retinal.

Change orders and reformulation

Any post-signing change to the formula, packaging, or specification is a change order. The paper should describe what a change order costs (typically $500 to $3,000 in 2026 for a small reformulation, higher for anything that triggers re-stability), how long the OEM needs to execute the change, and whether the brand or the OEM absorbs the cost of writing off existing inventory that becomes obsolete.

Force majeure

In 2026 force majeure clauses are longer than they used to be. Post-2020 and post-2023 events pushed Korean OEMs to add specific language about port closures, US-Korea trade action (tariffs and Section 301-adjacent duties), and raw-material supply shocks (the Ukrainian sunflower oil disruption of 2022 to 2024 is a still-live reference case in the paper we see). The clause should say who eats the cost of a force majeure event, how long the OEM has to resume production, and whether the brand can source from an alternate facility during the event.

Regulatory responsibility

This is where MoCRA sits in 2026. The paper needs to name explicitly who is registered as the responsible person under FDA MoCRA rules, who pays for facility registration and product listing, and who owns the adverse event reporting workflow. In our experience across US indie brands using Korean OEMs, roughly 60 percent of the paper we see in 2026 puts the responsible-person duty on the US brand and 40 percent puts it on the OEM's US-facing entity, and the price difference is embedded in the per-unit cost.

If the brand is the responsible person, the brand needs to set up: a listed facility registration with the FDA, a product listing filing for each SKU, an adverse event reporting protocol (a mailbox that a real person monitors), and a records retention system compliant with 21 CFR 720 and MoCRA-linked FDA guidance. This work is not trivial and it should be scoped and priced before the paper is signed.

Confidentiality

Standard NDA language. Two carve-outs to look for: whether the brand can share the OEM's identity in marketing (many Korean OEMs prefer to remain behind the brand for competitive reasons, but some are comfortable with attribution), and whether the OEM can list the brand as a case study on its Korean website (this becomes a real question if the brand grows quickly and does not want its OEM identified for competitive reasons of its own).

Warranty and remedies

The paper should describe what the OEM guarantees on delivery (typically: correct SKU, correct fill volume within tolerance, meets microbiological and stability specifications), how the brand claims a defect, and what the remedy is. Common remedies: replacement lot, credit against next PO, refund of unit cost. Punitive damages are rare in Korean OEM contracts and if the paper is silent on remedies the brand's practical recourse in a dispute is limited.

What Korean OEMs quietly bake into the base price

Every mid-tier Korean OEM builds a set of standard tasks into the base per-unit price. Understanding what those are will help the founder read the invoice line-by-line and identify what is genuinely a pass-through cost versus what is being double-billed.

Typical inclusions in a 2026 fixed or hybrid contract at a US-facing indie ODM:

• Formula qualification (matching the target ingredient list to an existing base or minor modification)
• One round of packaging integration (bottle, cap, and label matched to the fill line)
• Base microbiological testing to ISO 17516 on the first production lot
• Certificate of analysis (CoA) delivery per batch
• One round of accelerated stability testing (40 degrees Celsius, 75 percent relative humidity, three months) for the qualifying lot
• Standard MSDS delivery for shipment
• Korean government affiliate documentation (KFDA-linked paperwork if the OEM is registered with a Korean regulatory adjunct)

Typical exclusions (billed as pass-through or add-on):

• Full 24-month real-time stability
• Challenge testing (USP 51 / EU CTFA preservation efficacy testing)
• FDA MoCRA facility and product registration
• UK SCPN filing for UK market entry
• Third-party clinical testing (patch, in-vivo efficacy studies)
• Custom active ingredient sourcing (bakuchiol, PDRN, ectoin, tranexamic acid at higher percent inclusion levels)
• Bespoke primary packaging (unique bottle molds, custom decorative components)
• Rush production (any timeline faster than the OEM's standard 8 to 12 week production window from PO)

If any of these items are missing from the paper's "included" list and the founder assumed they were included based on a verbal, the invoice will look different from the launch P&L.

I'm Liz, and I run altameet from Manhattan, NYC. Most of the contracts we help review sit in the hybrid category, which is the structure that catches the most indie founders on pass-through line items. If you want a quick gut-check on whether a Korean OEM contract on your desk is standard or overreaching, grab 15 minutes free with Liz.

Red flags to watch for

Certain contract patterns should slow the signing process down. None of them mean the OEM is bad. All of them mean the paper needs one more revision before signing.

1. Silence on formula ownership after a development fee

If the brand paid a development fee (any amount, even $2,000) and the paper does not explicitly say who owns the resulting formula, the default under Korean commercial law leans toward the OEM. This is the single most common IP loss we see for indie founders in year two, when they try to move to a second manufacturer and discover they cannot.

2. MOQ dropping to reorder MOQ silently

If the initial MOQ is high and the reorder MOQ is not named, the reorder MOQ is at the OEM's discretion. This can be inconvenient (the OEM sets reorder at 80 percent of initial and the brand can only reorder in large runs) or expensive (the OEM sets reorder pricing at a higher per-unit than initial pricing).

3. Regulatory language that says "compliant with applicable laws"

That phrase, without naming specific regulations, allocates 100 percent of MoCRA and FDA compliance work to the party that is legally responsible under those regulations. For a US indie brand selling in the US, that is the brand, not the OEM. If the founder assumes the OEM will handle it because "the OEM is registered," the brand is now unregistered under MoCRA and exposed to enforcement action.

4. No named termination trigger for quality events

If the paper does not describe what qualifies as a quality event that lets the brand terminate (contamination, stability failure, missed delivery windows), the brand's practical exit from a failing OEM relationship is limited to non-renewal at term-end. That is a slow exit, sometimes two years, in which the brand may be shipping product it does not want to ship.

5. Pass-through categories that are not enumerated

"Additional costs may apply for specialty testing, custom ingredients, and rush production" is not enumeration. Enumeration is a list of every pass-through line item with a unit price or a formula. If the paper uses the "may apply" language, the founder should ask for the list.

6. Exclusivity language without a "substantially similar" definition

"OEM will not sell the same formula to another brand" is meaningless without a definition of "same." A formula with a 3 percent ingredient swap is a different formula on paper but the same formula in the brand's real market position. The definition of "substantially similar" should reference specific criteria: percent overlap in the top ten ingredients by concentration, matching claim structure, or overlapping target demographic.

7. Payment terms that shift 100 percent risk to the brand

A 100 percent deposit before production is unusual. So is a clause requiring the brand to pay for raw materials at PO before the OEM procures them. Both are structures the OEM uses when it does not want to carry the working capital risk of the brand. If either shows up, ask why. Sometimes the answer is fair (Korean OEMs occasionally require full pre-payment for very small orders under 1,500 units) and sometimes it is a signal the OEM is under-capitalized.

8. Force majeure with no time cap

Force majeure is a legitimate protection for both sides. Force majeure with no time cap means the OEM can invoke it indefinitely and the brand has no reset. A well-drafted force majeure clause names a time cap (typically 90 to 180 days) after which either party can terminate the contract without penalty.

What the paper does not decide

Contracts are not the whole relationship. Every experienced sourcing partner will tell an indie founder that a well-drafted contract cannot save a bad OEM relationship, and a mediocre contract with a strong relationship almost always outperforms a strong contract with a weak relationship. The paper handles the exits and the disputes. The day-to-day work handles the launches, the reformulations, the crisis calls when a batch fails stability at month 4 and the founder needs to ship in 6 weeks.

What the paper cannot decide:

• Whether the OEM's plant manager will pick up the phone at 9pm Seoul time when a batch is running hot in accelerated stability
• Whether the OEM will proactively flag a raw material shortage before it hits the brand's inventory
• Whether the OEM will remember the brand's brand voice when the packaging vendor is late and a substitution decision has to be made
• Whether the OEM's US-facing sales lead will stay for three years or leave in year two
• Whether the OEM will treat the brand as strategic when the brand is small and treat it the same way when the brand is at scale

These are relationship questions, not contract questions. The Korean K-beauty industry is culturally more relationship-heavy than the US supply chain norm, and the difference between a "signed relationship" and a "living relationship" is the difference between year-one success and year-three growth.

How ALTA MEET reads OEM contracts

Reviewed for accuracy by ALTA MEET's formulation consulting team. When we review a Korean OEM contract for a US indie brand, we work through it in a specific sequence to catch the twenty-percent-of-clauses that decide eighty-percent-of-outcomes.

First pass: formula ownership, MOQ (initial and reorder), payment terms, term length, and termination. These five clauses decide whether the brand can exit the relationship if it needs to. If any of them are missing or vague, the review pauses until they are cleared up.

Second pass: regulatory allocation (who is the MoCRA responsible person, who pays for registration and listing, who owns adverse event reporting) and quality specifications (which ISO standards, what stability window, what testing frequency). These decide the FDA exposure and the ongoing cost of compliance.

Third pass: change orders, pass-through categories, force majeure, and confidentiality. These decide the friction of the relationship over its life. A relationship with clean change-order language and enumerated pass-through categories is a relationship that will not have surprise invoices.

Fourth pass: exclusivity, warranty, and remedies. These are the negotiation levers. The brand rarely gets everything it wants on these, but the negotiation itself surfaces how the OEM thinks about the relationship. An OEM that is unwilling to define "substantially similar" is telling the brand something about how it plans to sell the formula in the future.

Total time to review a mid-tier Korean OEM contract for a US indie founder: usually 4 to 8 hours of consulting time, plus a Korean-language cross-check of any Korean-only appendices that describe production specifications, plus a follow-up call with the brand's counsel if the brand has US legal representation.

Key takeaways

The single most important clause in a 2026 Korean OEM contract is formula ownership, and it is often silent by default. If the brand paid a development fee and the paper does not distinguish base, modification, and custom development, the brand does not own portable IP.

The four common contract structures (fixed-cost, cost-plus, hybrid, full-custom) map to different indie founder situations. Fixed-cost is right for white-label plus small tweaks. Hybrid is the 2026 default for indie K-beauty. Full-custom is the only structure that gives the brand real portability and real exclusivity.

MoCRA responsibility should be named explicitly. "Compliant with applicable laws" language allocates the responsible-person burden to whoever is legally on the hook, which for a US brand is the brand.

Pass-through categories should be enumerated with unit prices. If the paper uses "may apply" language, the founder is signing an open-ended obligation.

The paper handles the exits and the disputes; the relationship handles the day-to-day. Both matter and neither substitutes for the other.

FAQ

Do I need a Korean lawyer to review a Korean OEM contract?

Not always, but Korean-language cross-check of any Korean-only appendices is worth doing. The English-language main contract is usually clean enough for US legal review, but Korean-language production specifications, MOQ appendices, and stability testing schedules often have detail that does not survive translation. In practice most US indie brands use their US counsel for the main body and a Korean sourcing consultant (or bilingual counsel in Seoul) for the appendices.

What does a full-custom development fee actually cover in 2026?

For a mid-tier Korean OEM in 2026, a full-custom serum development fee ($8,000 to $35,000) usually covers: 2 to 4 rounds of formula iteration with the brand's target profile, base stability at 3 months accelerated, one round of compatibility testing with the brand's chosen primary packaging, and delivery of an approved production formula. It usually does not cover: full 24-month real-time stability, third-party clinical testing, or MoCRA registration work. Higher fees (above $25,000) typically include one or two of the excluded items.

How much MOQ flexibility do Korean OEMs actually give in 2026?

Reorder MOQ is more flexible than initial MOQ across most mid-tier Korean OEMs in 2026. Typical range: initial 3,000 units, reorder 1,500 to 2,000. Some indie-facing ODMs will offer initial MOQ as low as 1,000 to 1,500 on a hybrid contract with a higher per-unit price. Below 1,000 units is possible but usually requires the OEM to run the batch as part of a shared production run with another brand, and the paper should say so explicitly if that is the arrangement.

If I sign with one Korean OEM, can I move to another manufacturer later?

Only if the contract explicitly permits it and only if the brand owns the formula IP. For fixed-cost OEM contracts, the answer is almost always no, because the brand does not own the formula. For full-custom contracts with a clear formula-ownership clause, the answer is yes but requires the second manufacturer to reproduce the formula, which is a re-qualification project (usually 3 to 6 months, plus stability re-testing) rather than a simple transfer. Dual-sourcing (running the same formula at two OEMs simultaneously) requires the primary OEM's cooperation on formula documentation and is rare in indie K-beauty.

What happens under MoCRA if my Korean OEM is registered as the responsible person and later loses its US registration?

The brand's product listing is tied to a registered responsible person. If the OEM's US-facing entity loses its FDA registration, the brand's product listing is affected. In practice this creates a compliance gap that needs to be closed within a defined window (usually 30 to 60 days depending on the specific violation). The brand should have a fallback responsible-person arrangement documented in the contract, either the brand's own registered entity or an approved third-party responsible person service.

How long should a Korean OEM contract term be for an indie brand?

For a first-time indie founder in 2026, 12 to 18 months is a reasonable initial term, with a renewal option. Longer terms (3 to 5 years) typically come with better pricing and stronger exclusivity language but reduce the brand's ability to pivot if the OEM's capacity or quality changes. The middle ground: 24 months with an annual review clause that lets either party renegotiate pricing at the anniversary.

Do Korean OEMs usually accept US legal jurisdiction?

Not usually. Standard Korean OEM contracts specify Korean jurisdiction and Seoul arbitration under KCAB (Korean Commercial Arbitration Board) rules. Some larger OEMs will accept US jurisdiction for very large accounts, but for indie brands the norm is Korean jurisdiction. This matters because a dispute has to be resolved in Seoul, in Korean, and under Korean commercial law, which is why the relationship layer matters as much as the paper.

Working With ALTA MEET

ALTA MEET is a New York based cross-border sourcing partner for US indie K-beauty brands. We handle the intensive work of matching a brand to the right Korean ODM or OEM, structuring the contract terms that decide the economics a year later, and running the Seoul-side oversight that makes the paper actually work.

If you have a Korean OEM contract on your desk right now and you want a second opinion on formula ownership, MOQ structure, or MoCRA allocation, book a free 15-min gut-check with Liz. Fifteen minutes is usually enough to say whether the paper is standard, overreaching, or worth a longer conversation.

For a broader picture of Korean ODM sourcing, our companion pieces are useful reading. The Korean ODM direct vs. trading house vs. US broker decision covers sourcing channel structure. How to read a Korean ODM quote line by line covers the quote that usually precedes the contract. And the Korean ODM launch timeline gotchas piece covers what happens after the paper is signed.

Reviewed for accuracy by ALTA MEET's formulation consulting team.