Co-manufacturing: your formulation, their equipment, their calendar

Making and filling are separate jobs with separate contracts

The distinction that catches new buyers is between the party producing the product and the party putting it into its final pack. A co-manufacturer does the first: mixing, cooking, reacting, forming, whatever converts ingredients into product. Filling and packing may happen on the same site or somewhere else entirely, occasionally at a business that never touches a formulation. Splitting the two widens your choice of specialists while adding a transfer, a bulk storage step and a second party able to blame the first. Combining them simplifies accountability, at the price of being tied to whichever pack formats that site happens to run.

A recipe survives transfer badly

Formulations developed on small equipment behave differently at production scale, and the differences are rarely subtle. Heat transfer, shear, mixing time, holding temperature and the order in which ingredients go in all shift with vessel geometry. What results can be perfectly good and still not be the product you approved: texture moves, colour deepens, an emulsion holds differently. Expect an adjustment period and budget for the trials it consumes, because the alternative is arguing over whether the factory followed your recipe when the honest answer is that the recipe was never specified tightly enough for anyone else to reproduce.

You are buying line time, and line time is allocated

Production happens in slots on equipment shared with other people's products. Your batch sits between somebody's changeover before it and somebody's cleaning after it, and slots are allocated well ahead against forecasts. Short-notice runs are possible and expensive; missed slots are generally charged whether or not your materials arrived. This inverts the responsiveness a brand expects, since reacting to a sales spike means finding room in another company's calendar rather than adding a shift. Buyers who plan around booked repeating slots, and carry their own stock cover across the gaps, get noticeably better service than those treating the plant as available on demand.

Yield, giveaway and who absorbs the difference

Every process loses material: to the line at startup, to vessel walls, to trim, to the overfill that protects a declared weight. Someone owns that loss and the contract decides who. Where the plant buys the ingredients, loss sits inside their price and stays invisible. Where you supply them, it becomes a reconciliation that eventually turns into an argument. Establish expected yield during trials, establish how it will be measured, and establish what follows when actual yield falls short — investigation, shared cost, or a stated tolerance beyond which the plant absorbs it. Introducing that conversation after several runs is close to impossible.

Ingredient ownership decides who feels a commodity move

Free-issuing materials keeps you in control of specification and lets you buy at your own scale, but puts procurement, storage, release testing and shortage risk on your side, where a late delivery becomes your forfeited slot. Letting the plant purchase simplifies your operation and buries the ingredient market inside a unit price that gets reopened whenever commodities shift. A middle route suits many buyers: nominate the specification and approved sources for whichever ingredients define the product, and let the factory purchase the rest on its own terms and at its own scale.

Frequently asked questions

Should we buy the ingredients or let the co-manufacturer buy them?

It turns on which risk you would rather hold. Supplying materials yourself protects the specification, captures your own buying power and keeps cost visible, at the price of owning storage, release testing, shortages and any delay that forfeits a production slot. Factory purchasing removes that work and converts it into a price you cannot fully see. Plenty of buyers split it, nominating sources for the character-defining ingredients and leaving commodity items to the plant.

Why did the first production run differ from the sample we approved?

Because it was made on other equipment. Scale alters heat transfer, mixing energy, residence time and how fast a product cools, any of which can move flavour, texture or colour without anybody doing anything wrong. The useful response is a structured trial programme judged against the reference, adjustments logged as formulation revisions rather than informal tweaks, and a signed standard describing the production version. Treating the bench sample as a permanent target usually ends in stalemate.

How much of our formulation do we actually have to disclose?

Enough for the plant to make it safely and lawfully, which in most consumable categories means all of it. Ingredient declarations, allergen status and regulatory compliance each need the full picture, and a site unable to see composition cannot control it. Where one component is genuinely proprietary, some buyers supply it as a pre-blended concentrate made elsewhere, so the factory handles a single specified input without learning what sits inside it.