R&D and NPD as a Platform, Not a Favour: From Brief to Shelf-Ready SKU

Buyer’s snapshot

• In 2025, 83% of food and beverage companies planned to raise NPD budgets, up from 76% in 2024, yet 82% still ran development on spreadsheets, email, or paper (TraceGains 2025 NPD Report, n=190).
• Money is rarely what stalls a new SKU. Sequencing is: decisions made in the wrong order, by the wrong side of the table.
• Silk Foods Ceylon (SFC) runs its R&D bench and its production lines on one 10,000 sq ft floor in Matale, under BRCGS and FSSC 22000 V6. The R&D stage, from signed brief to approved formulation, runs three to four weeks at most.
• The two questions a brand owner should bring to the first call are near the end. They are the part to forward.
• This piece is for brand owners who intend to own a specification. It is not for anyone shopping for free samples.

Most contract manufacturers treat custom development as a sales cost. A buyer asks for a blend, the plant makes a few samples for free, and everyone hopes an order follows. Sometimes it does. More often the formulation goes into a folder, the buyer goes quiet, and a week of bench time goes with them. The problem is rarely the buyer. It is the way the work is framed: R&D treated as a favour done to win orders, when it should be the platform the orders are built on.

What does it mean to treat R&D and NPD as a platform?

Treating R&D and NPD as a platform means the manufacturer runs product development as a scheduled, gated service on the same floor that will make the product, and the brand owner makes the commercial calls at each gate. A favour has no schedule, no owner, and no stopping rule. A platform has all three, which is why it produces SKUs rather than samples.

The favour model fails in a predictable way. Because nobody is paying for the development, nobody is accountable for its pace. The brand treats the manufacturer’s bench as free capacity and sends four directions instead of one. The manufacturer treats the brief as a sales cost and quietly deprioritises it when a paying order lands. Both sides end up with a sample neither of them trusts.

The platform model changes the incentives. Development runs against a written brief, a stage plan, and a decision at each gate. Silk Route Ventures (SRV) publishes the gate sequence in detail in NPD as a service, from bench sample to first certified export run, and the capability map in R&D and NPD at Silk Foods Ceylon. This post is the argument for why the structure matters, and what a brand owner should ask before any of it starts.

Why “yes, we can” only works when it has a schedule attached

Saying yes still matters. The low-MOQ policy at SFC (50 kg per SKU for powders, 180 bottles for a capsule run) exists so that small and early-stage brands can get a quote at all. What matters just as much is what that “yes” commits both sides to.

A yes with no brief, no gate, and no ownership terms feels generous. In practice it is the most expensive answer a manufacturer can give, because it hides the cost of development inside the price of a future order that may never be placed. The brand pays for that eventually, through a higher unit price or a slower response on the next project.

The trade press has started describing the same shift from the other direction. In August 2026, FoodNavigator reported that large CPG companies are expanding their use of contract manufacturers partly to validate demand before committing capital, with co-manufacturing moving a concept to shelf in three to six months against 12 to 36 months for a brand building its own capacity. That speed only exists when development runs on a schedule. A favour does not have one.

Where the platform model doesn’t fit Briefs that expect a formulation to be developed at no cost against an unquantified volume promise. Launches that need finished stock on shelf inside a month from a standing start. Projects where the target landed cost sits below what a certified origin can deliver. For those, a regional uncertified co-packer or an off-the-shelf private-label SKU is the more honest answer.

How does a brief become a shelf-ready SKU?

A brief becomes a shelf-ready SKU in five steps: scoping, bench samples, a pilot batch on production equipment, shelf-life confirmation, and the first certified export run. Each step ends with a decision the brand owner signs. At SFC the R&D stage, from signed brief to an approved formulation, runs three to four weeks at most. Production then takes two to three weeks from purchase order to dispatch.

Stage	What the brand owner decides	What the plant proves	What goes wrong in the favour model
Scoping	Target market, claim platform, pack format, landed-cost target	Feasibility, ingredient sourcing, indicative cost band	Skipped entirely; the bench starts on a vague idea
Bench samples	Which single direction goes forward	Taste, colour, texture, rough ingredient cost	Four directions run in parallel, none finished
Pilot batch	Whether the real yield supports the target cost	Yield, cycle time, fill accuracy, process stability	Never run; the first production batch becomes the pilot
Shelf-life confirmation	Whether the date code is defensible to the retailer	Accelerated and parallel real-time stability data	Date code copied from a competitor’s pack
First certified run	Ship, or hold for a regulatory dependency	Batch COA, full export document pack	Documentation assembled after the goods are packed

The column on the right is the one that matters. Every failure in it is a decision that got made late, or made by the wrong side. None of them are technical problems. They are sequencing problems.

Shelf life is the step brands most often try to compress. The Institute of Food Technologists notes that accelerated testing can shorten a long question into weeks, but that the more extreme the acceleration, the less the result reflects real storage, which is why a parallel real-time study is the recommended practice. An accelerated number opens a launch. A real-time number defends it when a retailer audits the date code.

What changes when the bench and the line share one floor?

When the R&D bench and the production line share one certified floor, the constraints that normally surface at scale-up surface at the bench instead, where they cost a week rather than a quarter. At SFC the spread line runs 3,000 jars per day, the capsule line 100,000 capsules per shift, and the patty line 15,000 units per day, all inside one audit scope.

The most common example is a small one, and the SRV development desk has seen it repeat across pilots: the failure has nothing to do with the formulation. The pack was chosen at concept stage, before the process was fixed, and the jar geometry or the film would not survive the fill or the seal test. The formulation was fine. The artwork had already been paid for.

The fix was not clever. The SRV desk now asks for the target pack format on the first call, not the third. A pack decision made at scoping costs a conversation. The same decision made after a pilot costs an artwork revision, a new tooling lead time, and usually four to six weeks. That is what a platform does that a favour cannot: it learns from its own failures and moves the question earlier.

The certification position matters here too. Because SFC runs a cellular manufacturing layout across 400 plus products under BRCGS, FSSC 22000 V6, USDA Organic (per SKU), and EU Organic (per SKU), a new SKU developed on the bench lands inside the existing audit scope. For a brand whose retail listing depends on a GFSI-recognised standard, that removes a certification exercise from the launch plan. Buyers who want the supplier-side checks can use qualifying a coconut and plant-based contract manufacturer.

The two questions to bring to the first call

The first NPD call goes well when the brand owner can answer two questions: what this product has to prove, to whom, by when; and what number it has to land at, at what volume. Everything else in a brief, from flavour direction to pack colour, can be worked out on the bench. These two cannot.

Question one: what does this product have to prove, to whom, and by when?

This is the claim, the market, and the clock in one sentence. “A capsule that can carry a structure-function claim for US DTC by the second quarter” is an answer. “Something in the adaptogen space, maybe for Europe later” is not. The answer sets the ingredient list, the testing panel, and the regulatory path before a single bench sample is made.

The clock is the part brand owners underestimate. An ingredient without a significant history of consumption in the European Union before 15 May 1997 falls under the EU novel food regime, where the European Food Safety Authority alone has nine months to adopt an opinion on a valid application. A novel ingredient found at scoping is a formulation choice. The same ingredient found after a pilot is a delay measured in years.

Question two: what number does this SKU have to land at, and at what volume?

This is the landed-cost target at the launch volume and at the volume the brand needs to reach by year two. Without it, the bench optimises for taste and the first quote arrives as a surprise. With it, the development team can trade an ingredient grade, a pack format, or a fill weight against the target before anyone falls in love with a sample.

Volume matters as much as price, because the MOQ ladder changes the unit economics. A brand launching six powdered SKUs needs at least 50 kg of each, 300 kg in total, not 50 kg spread across the range. For capsules, 180 bottles is the single-shift floor. The trade-off between a short test run and a real price break is covered in working with low-MOQ contract manufacturers and, for capsules, in when 180 bottles makes sense.

Buyer’s checklist: before the first NPD call

1. One sentence covering the claim, the destination market, and the launch date
2. The regulatory framework for each market, in priority order
3. Target pack format, fill weight, and case configuration
4. Landed-cost target at launch volume and at year-two volume
5. MOQ per SKU at launch, and the price-break tier you need to defend
6. Formulation-ownership expectation, written down before the pilot
7. The person on your side who signs off each gate

Who owns the formulation when development is a platform?

When development runs as a platform, ownership is written down before the pilot, not argued about when a relationship ends. SRV’s standard position is that a formulation developed to a buyer’s brief and funded through the project belongs to that buyer, while the underlying process know-how stays with the manufacturer.

Ownership belongs in the platform argument because it is where the favour model does the most quiet damage. When development is free, ownership is ambiguous, and ambiguity always gets resolved in someone’s favour at the worst moment. Paying for development, even modestly, buys clarity. The payment terms are the same everywhere in the customer book: orders under $10,000 are 100% in advance by bank transfer, and orders of $10,000 or above are 50% in advance with the balance against scanned shipping documents.

The broader market is moving the same way. Circana’s 2025 New Product Pacesetters, published in May 2026, put the year-one sales of the top 200 US launches at $6.2 billion, and framed the larger risk for CPG companies as not launching at all. Brands that treat development as a platform launch more often, because each project leaves behind a specification, a process, and a decision record they can build on.

The open question is how small a brand can be and still use the platform well. The model assumes the brand owner can answer both questions on the first call, and some of the most interesting briefs come from founders who cannot answer either yet. Helping them find those answers is part of the R&D stage, and it is why the first call is spent on the two questions rather than on flavour.

Frequently asked questions

What is the difference between R&D as a favour and R&D as a platform?

R&D as a favour is unscoped, unpaid development done to win a future order, with no schedule or ownership terms. R&D as a platform is gated development against a signed brief, run on the production floor, with written ownership. TraceGains’ 2025 NPD Report found 82% of teams still manage development manually, which is where favour-model drift usually starts.

How long does it take to turn an NPD brief into a shelf-ready SKU?

At Silk Foods Ceylon the R&D stage, from signed brief to an approved formulation, runs three to four weeks at most. Production takes two to three weeks from purchase order to dispatch, and sea freight adds three to four weeks to the EU and Australia or four to five weeks to the US. Shelf-life studies run in parallel.

Can Silk Route Ventures develop a custom formulation for my brand?

Yes. The SRV R&D and NPD team develops custom formulations at the Matale facility and pilots them on the same lines that will run commercially, under BRCGS and FSSC 22000 V6. First-run MOQs depend on format: 50 kg per SKU for powders, 180 bottles for capsules, 1,500 jars for spreads, and 1,250 bottles for beverages.

Who owns a formulation developed through an NPD project?

SRV’s standard position is that a formulation developed to a buyer’s brief and funded through the project belongs to the buyer, while process know-how stays with the manufacturer. Industry practice varies, so the ownership terms should be written into the development agreement before the pilot batch, when either side can still walk away.

How Silk Route Ventures can help

At Silk Route Ventures, R&D and NPD is the front end of a manufacturing relationship, scoped and scheduled like any production order. The Silk Foods Ceylon facility in Matale holds BRCGS and FSSC 22000 V6, with USDA Organic and EU Organic on the relevant SKUs, and its cellular manufacturing layout lets a new formulation move from bench to pilot to first commercial run inside one audit scope. Samples ship door to door by international courier in three to five business days. If the two questions above map to a product you are planning, the easiest first step is a short briefing call. Contact us and the team will respond within one business day.

Sources

1. TraceGains, “2025 NPD Report” (survey of 190 food and beverage quality, R&D, and regulatory professionals), reported by Food Industry Executive, August 2025. Retrieved 2026-09-26. https://foodindustryexecutive.com/2025/08/fb-brands-double-down-on-new-product-innovation-but-many-lack-digital-infrastructure/
2. Donna Eastlake, “From Nestle to Mars: Why Big Food is embracing contract manufacturing,” FoodNavigator, 4 August 2026. Retrieved 2026-09-26. https://www.foodnavigator.com/Article/2026/08/04/big-food-rethinks-production-the-rise-of-contract-manufacturing/
3. Circana, “Circana Unveils 2025 New Product Pacesetters as Top Innovations Drive $6.2 Billion in Year-One Sales,” 21 May 2026. Retrieved 2026-09-26. https://www.circana.com/post/circana-unveils-2025-new-product-pacesetters-as-top-innovations-drive-6-2-billion-in-year-one-sales
4. Institute of Food Technologists, “How to Fast Track Your Shelf Life Testing,” Food Technology Magazine. Retrieved 2026-08-20. https://www.ift.org/food-technology-magazine/safety-and-quality-how-to-fast-track-your-shelf-life-testing
5. European Commission, “Novel food authorisations,” Regulation (EU) 2015/2283. Retrieved 2026-08-20. https://food.ec.europa.eu/food-safety/novel-food/authorisations_en

Further reading

• Food Safety Magazine, “From Benchtop to Scale-Up: Food Safety Considerations for New Product Development.” https://www.food-safety.com/articles/9243-from-benchtop-to-scale-up-food-safety-considerations-for-new-product-development
• Private-label spice blend ranges: formulation, labels, and MOQ

Written by the Silk Route Ventures Trade Team. Silk Route Ventures (E-Silk Route Ventures Ltd) is a Sri Lankan B2B supply-chain operator for the Food, Beverage, Wellness, and Nutraceuticals sectors. The Silk Foods Ceylon manufacturing arm holds BRCGS and FSSC 22000 V6 certifications. Questions or to request a sample: Contact us or email info@esilkroute.com.lk.