Tech Transfer to a Contract Manufacturer: Files

A technology transfer looks inexpensive on a sourcing plan because the formula already exists. The launch calendar usually says “move production”, followed by one sample round, one pilot batch & one purchase order. The cost appears later: an ingredient grade has no supplier reference, the mixing note says “cool until ready”, the pack minimum exceeds the first filling quantity, or the approved laboratory sample cannot be traced to a signed formula revision. A factory can reproduce a document only when the document contains enough controlled detail to reproduce the product.

This piece covers the transfer boundary, formula files, raw material references, process instructions, analytical methods, packaging inputs, scale-up evidence, revision numbering, acceptance batches, retention records, timing & commercial responsibility. It is written for brand owners & procurement managers deciding whether a formula package is complete enough to quote, sample & manufacture without assumptions. The central question is not whether a file exists, but whether every critical input, instruction, result & approval points to one current version.

Define the transfer boundary before files move

The first document should be a transfer scope, not a formulation sheet. It names the legal owner of the formula, the sending party, the receiving manufacturing site, the intended markets, the dosage form, the primary pack, the target batch size range & the work included in the quotation. A transfer may cover bulk manufacture only, bulk plus filling, or the complete finished unit. Those scopes require different evidence, so a single price request without a boundary produces prices that cannot be compared.

The scope also lists what is fixed versus open. A brand may lock the INCI list, sensory profile, colour, fragrance, pack, fill weight & finished specification, while allowing equivalent grades for non-critical materials. Another brand may require exact suppliers because earlier compatibility work used those grades. Both routes can work, but the receiving factory must know which substitutions require written approval. The practical sequence from a complete brief into controlled production is set out in our formula development transfer guide.

Responsibility needs the same clarity. One named person releases the source package; one named technical contact answers process questions; one brand approver signs samples; one quality representative approves the final specification. Procurement can coordinate dates, yet it should not silently approve a formula change through a purchase order. A responsibility matrix with “prepare, review, approve, retain” beside each record prevents an email thread from becoming the only evidence of a decision.

The master formula must be executable

A useful master formula contains a unique product code, document number, revision, effective date, nominal batch size, percentage composition & scaled quantities. Every material line needs an internal or supplier reference, the declared INCI name, the function, the target amount, the weighing tolerance & the point of addition. Percentages should total to the declared basis, with any processing loss or quantity adjustment stated separately rather than hidden in a water line.

Trade names alone are not sufficient. The same commercial family can include grades with different active content, carrier, particle size, viscosity or preservation system. Replacing a 50% active dispersion with a neat material at the same percentage changes the product even when the marketing name looks similar. The transfer file should therefore pair every trade name with supplier, grade, specification reference & current technical document date. If the brand permits an alternative, the alternative should appear as an approved option, not as an operator decision.

The formula should distinguish target from tolerance. A fragrance at 0.20% does not automatically carry the same weighing tolerance as purified water at 70.00%; a colour adjustment may use a pre-dispersion with a narrow limit; a pH adjuster may be written as quantity sufficient with a maximum addition. Those instructions affect both cost & reproducibility. A receiver who has to infer them will either pause the project or create a local convention that the brand never reviewed.

Raw material references close the identity gap

The raw material pack usually includes a specification, safety data sheet, certificate format, composition or INCI statement, allergen information where relevant, origin statement where requested, storage condition & retest or expiry basis. The goal is not to collect paper for its own sake. It is to define what may enter the batch. A material name without an acceptance specification leaves incoming quality staff unable to distinguish an intended grade from a similar-looking substitute.

Critical attributes should be linked to product behaviour. For an emulsifier, active content & physical form may affect the heating profile. For a polymer, grade, neutralisation requirement & hydration method may affect viscosity. For a powder, particle size or surface finish may affect dispersion. For a fragrance, colour, relative density & flash information may affect filling or shipping. The source site should identify which attributes were significant during earlier batches instead of sending every supplier page with no priority.

Supplier changes need a route in advance. A practical rule states which materials are single-source, which have pre-approved equivalents & which can be purchased against a performance specification. The receiving site then records the actual supplier lot on the batch record. This makes later comparison possible without disclosing a real commercial formula in routine correspondence, since codes can be used in cross-functional files while controlled technical staff retain the full mapping.

Process instructions need numbers, not shorthand

A laboratory note such as “mix until uniform” is not a manufacturing instruction. The process file should state vessel requirement by function, order of addition, phase temperatures, acceptable temperature bands, mixing type, speed or tip-speed basis, minimum & maximum mixing time, hold time, vacuum condition where used, cooling rate or checkpoints, adjustment sequence, transfer temperature & filtration or screen requirement. Each parameter should show whether it is a target, a permitted range or a record-only observation.

Scale matters because rpm alone does not transfer cleanly between vessels. A larger impeller, different geometry or different liquid height changes circulation even at the same displayed speed. Where possible, the source package should include impeller type, diameter ratio, batch depth, addition location & observed endpoint. The receiving team can then set a defensible scale-up range & record why it differs from the laboratory value. Copying the number without copying the physical basis creates false precision.

Sequence matters just as much. A gum dispersed into a humectant before water addition may hydrate cleanly, while the same gum tipped directly into water may form persistent lumps. A fragrance added above its intended temperature may shift odour. An electrolyte added before a polymer has opened may depress final viscosity. These are not formula changes, yet they can create a different finished bulk. The approved process must therefore sit under revision control beside the formula, not in a separate notebook.

In-process controls connect steps to outcomes

An in-process control is useful when it can stop a batch before more value is added. Typical checkpoints include phase temperature before combination, appearance after dispersion, pH after a defined equilibration time, viscosity at a named temperature with spindle & speed, density, bulk temperature before filling, screen condition, deaeration endpoint & net content checks. Each reading needs a method reference, an acceptable range, a sampling point & an action when the result is outside that range.

Timing should be explicit. Viscosity measured immediately after cooling may differ from the value after twenty-four hours; pH can settle after full hydration; colour can look different before entrained air has left the bulk. A specification that names only the instrument reading but not the conditioning time invites disagreement. The transfer package should state whether a result releases the next step, supports a later release decision, or is collected for process history only.

The quality system around these records matters more than any single form. Controlled issuance, line clearance, reconciliation, deviation review & signed release keep the formula version connected to the batch that was actually made. A wider view of production quality-system controls helps procurement distinguish a polished sample from a repeatable manufacturing route. The transfer file should reference that route without claiming that paperwork alone removes manufacturing variation.

Analytical methods must travel with specifications

A finished specification is incomplete when it lists only limits. “Viscosity: 20,000–30,000” has no stable meaning without instrument type, spindle, speed, sample temperature, container, preparation, rest time & reading point. “Colour conforms” needs a reference standard plus lighting or instrumental method. “Appearance conforms” needs the approved description. The method turns a limit into a reproducible decision.

Method transfer can start with a paper review, followed by side-by-side testing of the same representative sample where the attribute is critical. The sending laboratory supplies its method, calculation, reporting precision & example worksheet. The receiving laboratory records any unavoidable local difference. If the two sites obtain materially different values, the project needs a method alignment exercise before the first acceptance batch, not a relaxed limit written after disagreement.

Sampling is part of the method. A top sample from a settled vessel may not represent the bottom; a hot sample may lose volatile material before measurement; a viscous sample may trap air during transfer. The file should define sample location, sample quantity, container, closure, temperature, transport condition & maximum time before testing. Without that detail, a result can vary because of handling while both laboratories follow the same instrument instruction.

Packaging data belongs inside the transfer package

Bulk manufacture cannot be separated from the intended primary pack when compatibility, filling temperature or dose delivery is relevant. The packaging file should include component drawings, material identification, neck or fit dimensions, nominal capacity, brimful capacity, closure details, liner or gasket, pump output where applicable, decoration area, approved colour standard & component revision. A sample pack used for development should carry the same component code as the purchase specification.

The filling specification should state target net content, permitted unit variation, average control, bulk temperature or viscosity at filling, line hold limit, closure torque or assembly setting, coding position & visual acceptance standard. Packaging MOQ can exceed the planned filling quantity, so the commercial file should show whether excess components remain brand-owned, factory-held, returned or scrapped after an agreed period. That decision belongs in the transfer plan because obsolete printed stock can cost more than the pilot batch.

Compatibility evidence should identify the tested formula revision, component revision, storage orientation, conditions, reading points & observations. “Pack passed” is not enough when a later supplier changes a gasket or colour masterbatch. Relevant observations include leakage, panel distortion, stress cracking, pump output, blockage, colour migration, odour shift, label lifting & decoration rub. The purpose is traceability, not a universal promise that one component works with every later formula.

Pilot scale is a comparison stage

A pilot batch should answer named scale questions. These may include whether the powder addition wets at the larger surface area, whether the heating time changes phase exposure, whether the homogenisation window gives the intended structure, whether cooling is fast enough, whether the bulk can be transferred without excess air, or whether the filler can hold net content at production speed. The protocol should list those questions before the batch starts.

The pilot record then compares actual parameters with the proposed ranges. It records start & finish times, additions, temperatures, mixing settings, vacuum, yield, samples, adjustments, deviations & operator observations. Photographs may support visual history, but they do not replace signed readings. If a parameter changes, the reason should be written while the evidence is fresh, followed by a controlled update to the draft production process.

Scale-up acceptance should use the agreed product specification plus process evidence. A pilot can meet finished limits while running outside a stable process window, especially if a late correction pulls pH or viscosity back into range. The brand should know whether a result was achieved directly or through an adjustment. That distinction affects reorder risk, batch time & cost even when the released sample looks correct.

Revision numbering keeps one truth in circulation

Every controlled file needs a document identity, revision, effective date, owner, approver & superseded reference. Formula, process, raw material specification, finished specification, packaging drawing, artwork, test method & sample approval can use different numbering families, yet a master index should show which revisions form the current product package. “Latest file” in an email subject is not version control.

A clear convention might separate major product revisions from administrative corrections. A major change could alter composition, supplier grade, process sequence, specification limit or primary pack. A minor controlled update could correct formatting without changing execution. The exact convention may differ between organisations; what matters is that the meaning is written, old versions are withdrawn from points of use & every released batch can be reconstructed from the approved set.

Change requests should state the reason, affected documents, inventory impact, validation or comparison needed, regulatory review owner, cost effect, timing effect & approval route. A component supplier substitution may affect artwork, compatibility, fill setup & residual stock. A raw material substitution may affect INCI declaration, process & analytical profile. Recording only the changed line hides the work created elsewhere.

The acceptance batch closes the transfer

The first commercial-scale or agreed acceptance batch should run against a pre-approved protocol. The protocol names formula revision, process revision, approved materials, batch size, pack revision, in-process controls, finished tests, sampling plan, release responsibility & any enhanced observations. It also states which deviations require brand consultation before the batch can proceed. This turns the batch into a defined confirmation rather than an expensive experiment.

Acceptance should cover documents as well as product. The manufacturing record must be complete, actual yield reconciled, test results reviewed, deviations closed or formally accepted, packaging quantities reconciled, coding verified & retention samples assigned. If the brand receives pre-shipment samples, the sample code should point to the batch & pack lot. A courier envelope with unlabelled units is not a release record.

Transfer closure is a signed statement that the receiving site holds the current approved package, the agreed batch met the acceptance criteria, open actions are listed with owners, superseded drafts are archived & routine change control now applies. Closure does not mean the product can never change. It means later changes start from a known baseline, with responsibility for review visible to both parties.

Timing, cost & retention are file questions

Transfer timing is usually driven by missing dependencies rather than document volume. Supplier data may take longer than formula review; a custom component may take longer than pilot manufacture; an analytical method disagreement may hold release after the goods are filled. A workable plan shows dependencies by file: no pilot before material grades are agreed, no compatibility conclusion before pack revision is fixed, no acceptance batch before methods & limits are aligned.

Commercial terms should separate one-time transfer work from recurring production cost. One-time lines may include document review, sourcing of exact grades, laboratory comparison, pilot setup, method alignment, component samples & acceptance protocol work. Recurring lines include raw materials, conversion, quality testing, filling, pack, freight preparation & storage of agreed residual stock. When those categories are mixed into one unit price, a later reorder can appear unexpectedly expensive or unexpectedly cheap.

Retention completes the evidence chain. The plan should state how many bulk samples & finished units are held, where they are stored, under what condition, for how long, who may access them & what happens at disposal. Formula files, executed records, results, approvals & change history need a defined retention period as well. The correct period depends on the market, product, shelf-life rationale & quality agreement, so it should be agreed rather than assumed.

Technology transfer records at a glance

The table below maps seven core record groups to their minimum content, the decision each supports & the failure created when the record is vague. It is a procurement view of common manufacturing documentation, not a substitute for a site-specific quality agreement.

The record groups should point to each other through codes & revisions. A perfect formula beside an obsolete pack drawing is still an incomplete package. Before a purchase order is released, procurement can ask for the master index rather than every confidential page; the index is often enough to show whether the transfer is controlled.

What to put in the brief

Send a structured index before sending attachments. It lets the receiving team identify gaps, assign owners & return a quotation based on the same boundary the brand expects. Confidential formula detail can then move through the agreed controlled channel.

• The transfer scope: bulk, filling, packing or finished goods, with intended markets.
• The current formula number, revision, owner, effective date & exact material grades.
• The approved process with sequence, numeric ranges, endpoints & scale history.
• The finished specification with every test method, sample condition & reporting unit.
• The primary pack drawings, component revisions, fill target & compatibility evidence.
• The sample approval route, named signatories, response window & supersession rule.
• The acceptance-batch protocol, release responsibility & deviation communication route.
• The retention period for documents, bulk samples & finished units, with storage conditions.

Add a gap list rather than hiding missing items. A known gap can be priced, scheduled & assigned; an unknown gap appears during pilot work, when component orders or launch dates are already exposed. The cleanest transfer package is not necessarily the largest one. It is the package in which every current record has an owner, a revision & a purpose.

Where our capability fits

We receive technology-transfer projects through a controlled file index, then review formula identity, raw material grades, process parameters, test methods, pack inputs & approval responsibilities before pilot work starts. We can map open points, prepare a sampling route, record agreed substitutions & carry the approved revision into the batch record without publishing confidential formula detail outside the technical team.

We work with brands on private label (your brand on a proven base), OEM/ODM contract manufacturing & full custom formula development. The same transfer discipline can support a hyaluronic acid serum capsule format, a hydrating face cream programme & a body lotion pump pack, with each formula, process, component & approval tied to its own current code.

Tell us the product form, current formula status, intended markets, pack, first order quantity & target transfer window, & we will come back with the document gap list, sample sequence & acceptance-stage outline. Start at our inquiry page, email capsulebiotech@163.com, or message us on WhatsApp at +86 158-7655-5319.

General manufacturing information, not regulatory or legal advice. Transfer records, retention periods, product files & release duties vary by market, product type & commercial agreement; confirm the applicable requirements with qualified advisers & the relevant competent authority before placing goods on sale.