Configure the process once. Take it to the next site.

1A document package rewrites the process. Seal carries it forward.

A transfer usually travels as documents. The sending site writes its process into descriptions, batch records and reports, and the receiving site reads them and builds the process again in its own systems. In Seal, the receiving site starts from the process already configured, and each difference from it becomes owned work with its own evidence and approval.

Technology transfer moves product and process knowledge between development and manufacturing, and within or between manufacturing sites. That knowledge becomes the basis for the manufacturing process, its control strategy, the process validation approach and continual improvement.¹ When it travels as documents, the receiving team rebuilds the process from them: instructions, parameters, calculations and method references are retyped, and the rationale behind each value is separated from the value itself.

The gaps then surface late. A hold limit, a working volume or a method revision that differs between the sites is found during engineering runs, or only after a batch, and each one is reconciled by reading two documents side by side.

In Seal the receiving site starts from a copy of the configured process itself where the receiving site uses Seal; otherwise it works from a structured comparison of the two sites’ records. Each site difference becomes owned work with its own verification and approval. The receiving team gets a process it can execute, and the sending site’s evidence stays attached.

You can start with one transfer, alongside the document, laboratory and manufacturing systems both sites already run.

2Start the receiving site from the configured process.

A permitted copy of the configured process starts the receiving site as a separate draft. It keeps the steps, fields, calculations and references, and none of the sign-offs.

A process copy retains document content, fields, field configuration, referenced records and coupled rows, held as a separate draft with checkpoint completions cleared. The receiving team can assess local changes without rebuilding the structure from documents, and a value it changes is a visible difference from the source rather than a retyped number.

A site-specific process binds the platform process to that site’s equipment, scale and facility constraints, so the differences from the source are recorded rather than implied. One platform process can start a sponsor facility, a GMP pilot plant, a commercial plant or a CDMO, each with a configuration of its own.

Copying is not automatic site readiness. It carries the starting definition forward; destination access, referenced dependencies, local configuration and the required reviews still matter. Destination systems and referenced records must be accessible under the configured permissions, and duplication does not approve equipment equivalence, validation or site readiness.

3Make every site difference explicit.

Ask Neil to compare the source process with the receiving plan. Each finding points to both sources, states what is missing and proposes the work; the transfer team decides what each difference means.

In the fictional example, Neil works from two excerpts. The sending-site process, PROC-014 v04:

Transfer within 45 minutes of completing the mix. The source process uses a 500 L working volume and analytical method AM-014 v04.

The receiving-site plan, SITE-B-014 v01, which is proposed and not yet executed:

The proposed schedule allows 60 minutes from completion of mixing to transfer. The receiving plan specifies a 2,000 L working volume and method AM-014 v03.

Neither excerpt includes the hold study, an equipment comparison or a method revision-impact assessment, and Neil says so. The excerpts establish the differences, not the receiving site’s readiness or the complete transfer scope.

The useful question is what the receiving team needs to understand, assess and demonstrate. A matching method name or a larger vessel does not establish suitability, so each finding names its source version, the evidence that is missing and the work proposed to resolve it. The transfer team reviews the scope, scientific interpretation and acceptance criteria.

Equipment substitutions deserve a structured argument of their own. An equivalency record links the proposed equipment to the parameters and supporting studies it affects, so the case for a substitution can be reviewed against the evidence rather than read from an attachment.

4Turn each difference into owned work.

Each difference becomes a piece of work with a proposed owner, the evidence that would resolve it and a decision still to make. The transfer team decides when it is resolved.

The proposed 60-minute hold exceeds the source limit by 15 minutes. The process owner confirms the time definitions and reviews the receiving schedule against the source hold study, then resolves the schedule or assesses a proposed change. The limit is not extended silently.

Working volume changes from 500 L to 2,000 L, and comparable performance is not established. Receiving-site engineering and MSAT compare vessel geometry, mixing and transfer arrangements against the process needs, and define the equipment assessment and any studies needed to support the receiving configuration.

The receiving plan names AM-014 v03; the source process names v04. Analytical development and receiving QC obtain both versions and their change history, assess the revision difference and agree the applicable method, transfer work and training before use. EU GMP asks for a documented gap analysis before a test method is transferred, and for a transfer protocol that covers the testing, additional training, standards and samples, and acceptance criteria; deviations from the protocol are investigated before the transfer closes.²

Each of these is an adjustment to the process made during transfer, which ICH Q10 asks the change management system to manage and document at this stage.³ The assessment retains the two source excerpts and records proposed owners, evidence requests and unresolved decisions. It does not approve a transfer or recommend extending a process limit.

A completed checklist is not an approved transfer either. The transfer team defines the scope, acceptance criteria, required studies and approvals. Closed actions are evidence for that decision, not a substitute for scientific assessment, qualification, validation or required regulatory work. The MSAT blueprint follows the rest of that acceptance lifecycle, through engineering batches, method transfer, training, qualification and validation.

5Neil prepares the work. Your team decides.

Neil is Seal’s AI agent. It works across the same records your team does, within the permissions of the person asking, and what it produces is work in the system: a comparison, a configured draft or an assessment to review.

Everything in the middle column is preparation. Nothing Neil produces takes effect until a qualified person has reviewed it, and required people and signatures still control the decisions.

In line with the EU’s draft GMP Annex 22 on AI, Neil is not used in GMP execution. It helps set up configuration, which your team verifies and releases under change control. Customer data is not used to train AI models, and Neil reads only what the person asking may read, so the access you configure for each site and each party applies to Neil too. See how Neil fits Annex 22 and more about Neil.

Transfers repeat, so the method can be kept. A transfer skill can ask for the source and receiving versions, compare equipment and methods, identify missing evidence and prepare an assessment in your team’s format. User-authored skills are versioned, so the team can inspect and refine the method; Neil still checks the current sources and process version for each transfer.

6Keep the parameter rationale with the process.

A critical process parameter keeps its justified range, the quality attribute it controls and the studies that established it. When the receiving site proposes a different range or different equipment, those links show which evidence is in question.

The development study, characterisation run or scale-down model that established a range remains linked to it, along with the risk assessment that covers any equipment change. Comparability evidence bridging two process versions or two sites stays linked to the versions it bridges, rather than sitting as a standalone document.

That is the knowledge the receiving site needs to validate the process. EU GMP expects process knowledge from development studies to be accessible to the manufacturing site and to form the basis for validation; for a product transferred from one site to another, existing product knowledge, including the content of the previous validation, should be available.⁴ ICH Q10 counts development studies, technology transfer activities and process validation among the sources of product and process knowledge to be managed through the product’s life.⁵

Existing development documents can still contribute. Neil can draft a structured comparison from supplied PDF, Word, Excel, JMP or PowerPoint files, proposing extracted parameters, ranges and rationale as a changeset for review before they enter the record. Your team checks the extracted values, versions and references before using the assessment.

An established condition recorded on a process element informs the regulatory assessment of a later change: whether it needs a Prior Approval Supplement, a CBE-30, an Annual Report or no filing. Process development supplies the rationale; post-approval change follows a change through its filings.

7Hand the receiving site a working record.

The receiving team has to run the process, so the handover should be something it can execute. The master batch record is built from the site-specific process, so the execution template and the parameter set do not drift apart.

On the batch, the instruction, the recorded mix, hold and transfer times, the lab sample and the investigation into an extended hold sit on one batch record. An operator works to the limit the process defines, and a reviewer follows an exception to the step that produced it.

Process performance qualification batches are planned with the recorded differences between the source and receiving sites in view. Monitoring during scale-up gives a preliminary indication of process performance and of the process’s integration into manufacturing, and what is learned during transfer can further develop the control strategy.⁶ Each batch keeps the actual values, material lots, results, exceptions and signatures recorded under its process version, even after the process changes.

See manufacturing execution and electronic batch records for the execution side.

8Work with both sites’ existing systems.

Neither site has to replace its systems to begin. Bring the source process, supporting studies and receiving proposal from the systems that hold them, or start with supplied records, then agree what Neil should compare.

Each of those records can stay in the system that holds it today.

Agree source authority, access and the records in scope. Available APIs, files and source permissions determine the implementation, and connecting a system does not grant permission to change it: a connected source is not permission to change the receiving MES, LIMS or QMS. See all integrations.

For sponsor and CDMO teams, the contract says who undertakes each step of an outsourced activity, including knowledge management and technology transfer.⁷ Define responsibilities, permitted records and review access for each party, and keep technical questions, evidence requests and decisions together without assuming every participant can see every source. The CDMO blueprint follows the same transfer from the receiving side.

Process Development provides the process rationale. MSAT assesses applicability and manufacturing differences. Change control keeps proposed revisions and implementation evidence connected.

9Keep reviewed history when the process changes.

Each assessment names the source version it compared: PROC-014 v04 in this example. When the source is revised, the team reviews the revision against each receiving configuration; earlier decisions stay on record, and nothing at the receiving site becomes effective until it has been reviewed.

Preserve the versions used in each assessment. When the source process changes, compare the revision against each receiving site’s configuration, identify the affected work and route proposed updates for review. A later source revision does not silently replace the earlier decision or make a new site configuration effective.³

The same discipline applies after the transfer. A deviation at the receiving site links to the parameter, study and batch context needed to investigate it, and a proposed change to the platform process can be assessed against every site configuration that inherits it. ICH Q10 describes CAPA during technology transfer as a system for feedback, feedforward and continual improvement:⁸ what one site learns can return to the platform process as a reviewed change, and each site that inherits it adopts it on its own review.

10Reuse what each transfer establishes.

The lasting work in a transfer is in the process library, the destination model and the rules for differences. Once that foundation exists, each transfer can reuse more of the work already done.

Define activities such as buffer preparation, inoculation, harvest, column packing and sampling with their inputs, outputs and checks. Keep the activity definition separate from the local equipment assignment so it can be assessed for use at another site.

Record each site’s equipment capabilities, working volumes, automation interfaces, material specifications and analytical methods. A matching name is only a mapping candidate; the receiving team records the suitability assessment and any supporting studies.

The destination is one-click preparation of a receiving process: select an approved source version and destination, then generate a manufacturing draft using established mappings. The click prepares a process for the receiving team; scientific assessment, qualification and release remain recorded decisions.

11Evaluate one real transfer question.

Start with one process, a receiving-site proposal and one transfer question, and agree an output such as a gap assessment or evidence-request package. Judge it on your own records rather than a demonstration.

Bring the relevant source version, the receiving-site plan and the supporting records it depends on. Start with sample material or arrange an NDA before sharing confidential records.

Check whether the comparison is accurate, the references are inspectable and the proposed follow-up is useful. Your team defines the full transfer scope and the required technical, quality and regulatory work.

From there, the next site starts further ahead: the platform process, its rationale and the destination mappings are already in place.