"I already have a formula" — and what a factory hears when you say it

At this factory, every client who arrives with a product sits down with an R&D specialist before anything is costed. Our chairman describes that as a rule from day one, and the reason is narrow: the first question is not what the product should cost, it is whether the formula can actually be manufactured on the equipment that exists.

Across more than 1,500 formulas and products, the answer has frequently been "not as written".

That conversation happens in-house: the plant runs its own research and development centre rather than buying development in per project.

The four things "I have a formula" can mean

A competitor's label. An ingredient list with no quantities. Useful as a brief, not as a formula. Two reasons it cannot be more: a label declares amounts per dose rather than per batch, and it omits excipients entirely — so even a label with full quantities is describing only part of the actual recipe.

A list of actives with doses. Closer, and a perfectly reasonable starting point. Still missing the excipient system, the overages and the process.

A full quantitative formula from another factory. Genuinely useful, and still not portable. That formula was developed for that plant's equipment. A direct-compression blend that ran on their press may not compress on ours; granulation route, particle size and powder flow are specific to the machines in the building.

A master formula with a batch record and a validated process. Rare, and transferable. If you have this, say so early — it changes the whole conversation.

A complete manufacturing formula specifies every input quantified per batch, the grade and specification of each, the overages, the order and method of addition, mixing times and speeds, granulation liquid, drying temperature and endpoint, compression or fill parameters, expected yield, in-process limits, and the finished-product specification with test methods. That document is the deliverable of development. The ingredient list is its first page.

The manufacturability question has teeth

Dose arithmetic comes first. Add the actives up by weight. If the total exceeds what the format can physically hold, something changes — the dose, the ingredient list, the format, or the number of units per day. A multivitamin with fifteen actives at the levels a brief usually specifies does not fit in a tablet anyone will swallow. This is the single most common reason a brief has to be revised, and it takes ten minutes to discover.

Not every powder compresses. Extracts are often sticky, low-density, electrostatic or all three. Some materials are hard enough that factories decline them: a waxy material we have manufactured for years is one other plants simply pass on, and the test is not whether you can mix it but whether the correct concentration survives into the finished unit, matching the registered figure rather than approximately resembling it.

Equipment sets limits before the formula does. Available tooling sizes, blister formats, capsule shell sizes, sachet fill envelopes, bottle and jar neck finishes. An elegant seven-gram sachet is irrelevant if the line's fill range stops below it.

Incompatibilities decide architecture. Some minerals and vitamins degrade each other in contact. Oxidising and reducing actives do not share a blend happily. Live cultures and hygroscopic salts are a bad pairing. When chemistry says no, the answer is physical separation — a bi-layer tablet, two SKUs, or a different format — not a smaller quantity and hope.

Overages: the thing nobody explains to brand owners

Your label claim is what the consumer has to receive on the last day of shelf life, not on the day of manufacture.

So a formula carries an overage: more active than the label declares, calculated against the known degradation of that material over the claimed shelf life, in the chosen pack, at the stated storage conditions. Several vitamins are standard cases. Live cultures are routinely formulated well above claim for the same reason.

Three consequences worth understanding before you compare quotes:

1. Overage costs money. A cheaper quote may simply be carrying less of it. Ask.
2. Overage has a ceiling. Regulatory limits and upper intake levels are real, and a long shelf life on a fast-degrading active can be arithmetically impossible rather than merely expensive.
3. Shelf life is a cost lever almost nobody uses. Claiming 24 months instead of 36 reduces the overage you need. If your product sells through in six months, you are paying for a claim you do not use.

And the inverse, which is useful when you are reverse-engineering a competitor: their declared label is not their formula, because you cannot see their overage from the outside.

Excipients are not filler

Each non-active has a job: bulking so a 5 mg active becomes a pressable tablet, binding, disintegration, glidants and lubricants so powder flows and the tablet releases from the die, coating, sweeteners and flavour and acid systems, anticaking, preservation in liquids, carriers for oils.

"No fillers" as a marketing line is mostly nonsense — a capsule containing nothing but active is only possible when the dose happens to fill the shell. But excipient choice is a real decision, and more often a market-access decision than a technical one:

• Gelatin or HPMC shells, which decides vegan and some religious-market acceptance
• Titanium dioxide, whose regulatory status differs by market
• Sugar or polyol in a syrup, which changes both the regulatory position and the microbiological risk
• Lactose, wheat starch, soy lecithin, shellac, bovine gelatin — each one an allergen or certification constraint travelling quietly into your label

Decide the markets first. The excipient set follows from them, and retrofitting it later means a new stability study.

Trials, scale-up and stability

Bench to pilot to production is not the same process three times. A blend behaves differently at 200 kg than at 2 kg, and mixing uniformity, segregation and heat behaviour all change with scale.

In this factory's early years, trial batches were run at its own cost and lost money — treated, in our chairman's framing, as the cost of being trusted rather than a line to be recovered. The sequence and who pays for it is covered in what actually happens between your idea and your first pallet; what matters here is what the trial produces, which is a process, not just a sample.

Then stability. Real-time studies at the claimed storage condition, plus accelerated conditions to support a provisional shelf life while real time continues to run. The pack is part of the study, not a variable you change afterwards.

The planning consequence is blunt: a 24-month shelf life cannot be proven in three months. You launch on accelerated data with real-time testing continuing behind the product. Anyone offering you a validated long shelf life before a batch exists is describing a hope.

From a brief, or from an existing formula

That last row is the one to settle in advance. Who owns a formula developed by the factory's R&D, and what you may do with it if you leave, is a contract question and an awkward conversation much later if it is skipped.

What to bring to the first R&D meeting

• The consumer and the market, named specifically
• The dosage form you want, or how the consumer should take it
• Your formula if you have one — and say which of the four kinds above it is
• Target cost per unit, honestly
• Your shelf-life and storage claim
• Volume in month twelve, not only month one
• Every market you intend to register in

The most useful thing you can arrive with is not a formula. It is a clear statement of consumer, market, format, unit cost and volume. A factory's R&D can build a formula from that in weeks. It cannot build a market from a formula.