TL;DR: A brand will come in with a specific on-pack claim locked (“visibly brighter in 7 days”), but no decision yet on texture, delivery system, or packaging
TL;DR: In one 2023 project, a brand flagged PEG-exclusion only after pilot formulation was complete
Key Technical Parameters #
Brand founders coming to us for the first time almost always have the same problem: they know what they want the product to do, but they don’t know how to communicate it in a way a formulation lab can actually act on. The gap between “I want a brightening serum that feels lightweight and works fast” and a workable formulation brief is wider than most people expect. This guide walks through exactly what we need from you, when we need it, and what happens at each stage from first conversation to production-ready formula. The audience is a brand owner who has a concept, maybe a benchmark product or two, and is now trying to figure out how to actually get it made.
What Goes Wrong Before the Brief Is Even Written #
The projects that run smoothly almost always start with a well-structured brief. The ones that drag on for four to six months — and some do — usually trace back to a vague or incomplete intake document.
The most common gap we see is claim-first, format-last thinking. A brand will come in with a specific on-pack claim locked (“visibly brighter in 7 days”), but no decision yet on texture, delivery system, or packaging. Those downstream decisions directly affect which actives we can use and at what concentration. You can’t formulate a niacinamide-forward brightening serum for a pump bottle and a jar with the same formula and expect the same stability performance. The packaging choice changes the oxygen exposure, which changes the oxidation risk, which changes whether you need an antioxidant booster and what that adds to the BOM cost.
A second issue: brands often send us competitor benchmarks without telling us what specifically they’re benchmarking against. “Similar to [Brand X] serum” gives us texture direction, roughly. But are you benchmarking the skin feel, the actives, the finish, the price point? When you send us a benchmark, tell us which dimension matters. We use the benchmark to calibrate — but we need to know what part of it we’re calibrating to.
The third failure mode is ingredient exclusions. Clean beauty brands in particular tend to have lengthy exclusion lists, and we fully respect that. But those lists need to be in the brief on day one, not surfaced after we’ve already run three batches. In one 2023 project, a brand flagged PEG-exclusion only after pilot formulation was complete. That required a full re-emulsification approach and added five weeks to the timeline. We now include a Category B Exclusion Checklist as part of our standard brief intake form, which asks about preservative class preferences, silicone policy, allergen fragrance restrictions, and EU-specific prohibitions under EU Cosmetics Regulation 1223/2009.
The Brief Parameters That Actually Drive Formulation Decisions #
Here is what we need in writing before any lab work begins. Some of these seem obvious. Others, brands consistently underestimate.
Target market. This is not a marketing question — it’s a compliance question. A niacinamide brightening formula destined for NMPA registration in China has different concentration thresholds, documentation requirements, and labeling rules than one going to EU or the US. Under NMPA Cosmetic Regulation, whitening claims (美白) trigger a special cosmetic classification requiring clinical substantiation and separate product registration. That changes the timeline by months and the development budget by thousands of dollars. We need to know market destination at brief intake, not after the formula is locked.
Texture and format. Serum, cream, essence, toner, mask — each category carries different active delivery constraints. In our brightening & whitening formulation work, we see the most brief mismatches in the essence/toner category, where brands want high active load but the format’s water-heavy base creates stability pressure at any alpha-arbutin concentration above 2%.
Actives: inclusions and concentrations. If you have a hero ingredient in mind, tell us the concentration range you want on-pack. If you want “tranexamic acid at 3%,” say that. We’ll tell you whether that’s achievable in your chosen format, whether it’s stable at that level in a water-based system, and whether it triggers any regulatory flag in your target market. What we don’t want is to develop a formula and then discover the brand wanted 5% kojic acid — that’s outside the safe-use range in multiple jurisdictions and would have been redirected immediately.
Benchmark products. Send physical samples if you can, and annotate them. “I like the skin feel of this one, I don’t want the white cast of that one, I want the actives profile somewhere between both.” That level of direction cuts iteration cycles significantly.
Claims and substantiation appetite. Cosmetic claims don’t need clinical trials behind them in most markets, but if you want clinically validated efficacy copy (“reduces dark spots by X% in Y weeks”), you need to budget for an in-vivo study. If your claim is softer (“supports an even-looking skin tone”), we can design a formula to meet that standard without a full RCT. We align this decision with the brief, not after the formula is finalized.
The parameter brands most often underestimate is preservation system selection relative to formula pH. Brightening formulas tend to run acidic — vitamin C systems often sit at pH 2.8–3.5, alpha-arbutin performs well around pH 4.0–5.5. Several traditional preservation systems lose efficacy below pH 4.5. If no one addresses this at brief stage, you can end up with a formula that passes initial micro challenge testing and fails re-test at month four. We’ve logged this failure type in our QC-09 stability incident database across multiple projects. It’s fixable, but it costs time.
| Parameter | What Brands Often Provide | What We Actually Need |
|---|---|---|
| Target active | “Something with vitamin C” | Specific compound (L-AA, SAP, MAP) + desired on-pack % |
| Market | “US and EU to start” | Confirmed registration pathway, any NMPA plan? |
| Texture | “Light serum” | Benchmark product(s) + annotated what you do/don’t want |
| Claims | “Brightening” | Specific claim copy + whether clinical data is needed |
| Exclusions | None stated | Full exclusion list: PEG policy, silicone, fragrance, preservative class |
| Timeline | “ASAP” | Hard deadlines: trade launch, retailer window, IP date |
| Budget | Not mentioned | Development fee tolerance + target FOB unit cost range |
From Brief to Batch: The Actual Timeline #
Once we have a complete brief, lab samples typically take two to three weeks for an initial set. That assumes actives are in-stock at our facility and no novel delivery system is required. Encapsulated actives or encapsulation technology components can add one to two weeks to the first-sample timeline because encapsulated grades need additional compatibility testing before we incorporate them into a base system.
After you receive lab samples, the evaluation phase is on your side. We generally see one to three rounds of sensory feedback before texture is locked. Each round is one to two weeks of re-formulation plus shipping. Most projects land at formula lock within six to ten weeks of the first sample shipment — though we’ve had straightforward briefs complete in four weeks and complex multi-active systems take fourteen.
Accelerated stability runs concurrently from the moment formula is provisionally locked. We run ICH-aligned conditions: 40°C/75% RH for six months, plus cycling and freeze-thaw, per ICH Stability Guidelines. A 3-month accelerated read is our internal gate for production green-light on most formulas. Real-time 24-month stability is initiated at the same time. You don’t wait for 24-month data to go to production — but your technical file needs to document that real-time data collection is ongoing.
Sample MOQ and cost. For lab samples, we typically produce 100–300g per variant, at no additional charge for the first two rounds under our standard development agreement. If iteration extends beyond three rounds due to brand-side direction changes, we discuss project management terms. For stability and safety testing batches, we produce at 1–5kg scale. These are billed at cost.
Production MOQ varies by product category and complexity. For brightening serums and essences, our standard production MOQ sits at 3,000 units. Creams and masks with standard formulas can go lower, at 2,000 units, because equipment changeover is simpler. Custom-developed formulas with novel delivery systems typically require 5,000 units minimum to cover the overhead of specialized manufacturing steps.
Development fee structure. We charge a one-time development fee for custom formulation work, separate from per-unit production cost. For a single SKU with standard actives, this is typically in the range of $800–$1,500 depending on complexity. Multi-SKU projects or those requiring in-vivo efficacy testing are scoped separately. The development fee covers lab labor, raw material prototyping cost, and documentation for your technical file. It does not cover third-party clinical study costs.
One thing worth stating plainly: the per-unit cost at sample and pilot scale is not your production FOB cost. Pilot batches at 5–10kg carry a premium of 20–35% over production-scale unit economics because of batch size overhead and equipment utilization. Don’t lock in a retail price based on pilot batch costing.
Progressing to first production. The gate from formula approval to production requires: signed formula approval form, confirmed packaging specs with compatibility test data, approved safety assessment (under EU Cosmetics Regulation 1223/2009 for EU-bound products, or FDA Cosmetics Guidelines requirements for the US), and a purchase order with confirmed delivery window. We target a 6–8 week lead time for first production once all pre-production gates are cleared.
Regarding clinical evidence for brightening claims: if your brand requires substantiated efficacy copy, the typical study design we recommend to third-party CRO partners is a single-blind, split-face protocol (n=30 minimum for cosmetic claims, n=40+ if you want publication-ready data). A 2022 split-face study we supported for a partner brand (n=36, 8 weeks, twice-daily application) showed a 22% improvement in ITA° angle versus untreated control, measured by Mexameter. That data underpinned a “visibly more even skin tone” claim with quantified backing. Studies of this design typically take 10–12 weeks including recruitment and reporting, and cost in the range of $8,000–$15,000 depending on the CRO and measurement panel.
Formulation Notes for Brand Partners #
When you brief us, the first three questions we ask are: what market, what format, and what’s the on-pack claim? Those three answers shape everything — the active selection, the pH strategy, the preservation approach, the regulatory documentation scope.
The most common brief mistake we see is treating the formula as separable from the regulatory filing. They’re not separable. If you want NMPA whitening registration for China, we need to know that before we select actives, because the approved whitening ingredient list for Chinese registration is specific and not all of your preferred actives will be on it. Discovering this after formula lock means starting over — or accepting a China-specific variant, which adds cost.
On timeline: first lab samples in two to three weeks from brief sign-off. Texture iterations add one to two weeks per round. Accelerated stability runs from formula lock and takes four to eight weeks for an initial read; we initiate 24-month real-time data concurrently. First production runs six to eight weeks after pre-production gates are cleared. For a clean, well-briefed project targeting a single market, formula-to-production can happen in fourteen to eighteen weeks. Add clinical study time if efficacy copy is required.
Tell us your packaging choice early. It changes more than most people expect.
Frequently Asked Questions #
We want to launch in both the US and EU — do we need two different formulas?
A: Not always, but sometimes. The formula itself can often be identical — the difference lands in documentation, labeling, and in some cases preservative choice. Where you’re most likely to need a variant is if your EU formula requires a SCCS-reviewed ingredient at a concentration that creates a US labeling conflict, or vice versa. We flag this at brief review, not at launch.
What’s the minimum order we can start with?
A: For production, our standard minimum is 3,000 units for serums and essences, 2,000 for creams. Lab samples are produced at 100–300g and aren’t subject to MOQ. If your launch volume is below production MOQ, we can discuss a pilot run at higher unit cost — but that’s a project-by-project conversation.
We sent a benchmark product — why did the first sample feel so different?
A: Benchmarks tell us direction, not destination. If the benchmark contains silicones and your exclusion list bars them, we’re already working with a different tactile toolkit. We’ve had projects where the benchmark and the exclusion list were essentially incompatible, and no one had noticed until we mapped them against each other. Send the benchmark and the exclusion list at the same time and annotate which properties you’re actually trying to match.
How do we know the formula is stable before we go to production?
A: You’ll have 3-month accelerated stability data (40°C/75% RH) before we recommend production green-light. That data covers appearance, pH drift, viscosity change, and micro challenge. It’s not the same as 24-month real-time data, which runs in parallel. For brightening formulas specifically, we also run oxidation marker testing for vitamin C systems and arbutin hydrolysis monitoring — both are early predictors of shelf-life risk that standard accelerated conditions can miss.
Should we register the formula under our brand name or yours?
A: This is something brands often don’t think to ask until it becomes a problem. For most markets, the cosmetic product is registered or notified under the responsible person or importer — not the manufacturer. But for NMPA special cosmetic (whitening) registration in China, the registration holder structure is different and has implications for IP ownership, re-export rights, and label claims. We walk through this in every China-bound brief. It’s worth a 30-minute call before you commit to a filing strategy.
Have a product concept in mind? Contact our formulation team to request a complimentary brief review.
The “visibly brighter in 7 days” claim example hits close to home — that’s a consumer perception claim, which means you’re looking at a minimum 30-subject HRIPT-adjacent consumer use study if you want it defensible for EU Regulation 655/2013 Criterion 6. We’ve had brands come in assuming a 7-day timeline on pack means 7 days of testing. The actual study timeline to substantiate it is closer to 8-10 weeks from formula lock, and that’s before you factor in the stability data the lab still needs to generate on the finished packaging configuration.
The packaging-before-formula sequencing issue is real — we spent three weeks reworking a SAP formula after the brand switched from airless pump to a standard disc-top mid-pilot, and the oxidation profile was completely different.
For MAP specifically, are there OEMs that will guarantee a minimum 3% on-label claim with stability data to back it at 40°C/75% RH through 12 months, or is that still a “we’ll see after accelerated” situation in most contracts?
We had a Shenzhen OEM that kept defaulting to niacinamide as the primary brightening active regardless of the brief — took three rounds of samples before we realized their “standard brightening base” was baked into their internal workflow and they were just swapping fragrance and texture around it. Once we got an actual formulator on a call instead of the BD rep, the whole dynamic shifted and we had a workable MAP prototype within two weeks.
Challenge testing tripped us up on a brightening serum launch — passed all stability screens at 40°C/75% RH but failed preservative efficacy (ISO 11930 Category A) at month 3 because the low pH optimized for ascorbic acid was suppressing our preservative system just enough to cause issues under microbial challenge. Didn’t catch it until we’d already submitted artwork.
MOQ reality nobody talks about upfront: most OEMs doing custom brightening work (not stock bases) won’t move below 500kg per SKU, which at a $18-22/kg fill cost for an actives-heavy serum puts your first production run at $9,000-11,000 before you’ve touched bottles, cartons, or fulfillment.
One angle that catches brands off guard on the EU side: if you’re combining a brightening claim with SPF in the same product, you’ve just moved from cosmetic into a borderline category under the CPSR framework, and your Responsible Person has to explicitly address the dual-function positioning in the safety assessment narrative — some EU notified bodies we’ve worked with will flag this even when the SPF is below 6.