Introduction
Every batch of medicine that reaches a patient passes through one final gate: the release decision. A Qualified Person (QP) or QA lead reviews the evidence, batch records, lab results, deviations, certificates, regulatory approvals and signs a legally binding statement that the product is fit for market. That decision is the most consequential moment in pharmaceutical manufacturing. It is also, in most organisations, the least digitalised. Production has been automated. Labs have been instrumented. Serialisation has been mandated and delivered. Yet the release step itself still runs on portals, spreadsheets and the institutional memory of overloaded QPs.
SAP Batch Release Hub (BRH) for Life Sciences Digitalisation was built to close that gap – bringing structure, automation and AI-assisted decision support to the one step that still runs on spreadsheets and institutional memory. This article explains why batch release has become the bottleneck, how BRH reframes the problem, and where AI is taking the release process next.
The Industry Challenge: Release Has Not Kept Pace with Production
Ask a QP what a release actually involves and the answer is rarely “one decision”. It is closer to a scavenger hunt across five to ten systems: the ERP for batch and inspection-lot data, the MES for execution records, the LIMS for analytical results, the QMS for deviations and change control, the DMS for the inspection plan and SOPs, the RIMS for market authorisations, the track-and-trace system for serialisation status, plus emails, PDFs and paper for everything that falls between them.
The process typically looks like this: locate the right inspection plan in the ERP or DMS; log into each source system in turn to check the individual release criteria; record the outcome on a paper or PDF form under good documentation practice; log into the ERP to post the usage decision and stock movement; scan and file the completed record so it can be retrieved during the next audit.
The consequences are well documented across the industry: labour-intensive, repetitive work for highly qualified people; error-prone manual transcription that creates compliance risk; long qualification periods for new hires because the process lives in people’s heads rather than in a system; release lead times measured in days or weeks, which inflate safety stock and inventory carrying costs; and no scalability for the smaller, more frequent batches that cell and gene therapies, personalised medicine and clinical supply demand.
The paradox is stark. Pharma 4.0 initiatives have cut manufacturing cycle times substantially — yet a sterile injectable batch whose MES record closes in 24 hours can still wait 20 days for release while lab results consolidate, deviations close and cross-system reconciliation happens by hand. Manufacturing got faster. Release did not.
Market Trends Reshaping Batch Release
Several forces are converging to make this status quo untenable.
Regulators are rewriting the digital rulebook. In July 2025 the European Commission published drafts of a revised EU GMP Annex 11 (Computerised Systems), a revised Chapter 4 (Documentation) and an entirely new Annex 22 on Artificial Intelligence. Annex 11 grows from five pages to nineteen, with a sharp new emphasis on identity and access management, audit trails, electronic signatures and data integrity. Annex 22 sets out, for the first time, how AI models may be used in critical GMP applications — intended use, acceptance criteria, independent test data, explainability and confidence thresholds. The message is clear: digital, auditable, well-governed systems are no longer a nice-to-have.
Annex 16 has moved supply-chain transparency onto the QP’s desk. QP certification now presumes visibility of the entire authorised supply chain for the product – every site, every step, every market. Jurisdiction control and supply chain maps are becoming a release-time requirement, not a regulatory-affairs afterthought.
Serialisation is mature, and its data is finally usable. A decade of DSCSA, EU FMD and national track-and-trace mandates means most manufacturers now hold rich, structured serialisation data in systems such as SAP ATTP. The next step is obvious: use that data as a release check rather than leaving it in a silo.
Portfolios are fragmenting. Advanced therapies, personalised medicines and clinical trial supply mean more products, smaller batches and far more release events per unit of revenue. A release process designed for large commercial lots simply cannot absorb this volume.
The ERP backbone is moving to the cloud. S/4HANA migrations and SAP’s Cloud ERP strategy are pushing life sciences companies to reconsider which quality processes belong in the core and which belong in purpose-built industry applications on SAP Business Technology Platform (BTP).
Talent is scarce. Experienced QPs are hard to find and expensive to train. Organisations cannot keep solving throughput problems by adding headcount.
The Shift: From Data Collection to Decision Support
The real shift is conceptual. Batch release has historically been treated as a documentation problem: gather the evidence, file it correctly, sign it. The leading organisations now treat it as a decision problem: what does the QP actually need to know, and how quickly can the system put it in front of them with confidence that it is complete and correct?
That reframing changes everything about how the process should be designed:
- Evidence comes to the decision-maker. Data from ERP, MES, LIMS, QMS, DMS, RIMS and serialisation systems is pulled into a single release cockpit mapped to a standard set of release checks, not scattered across logins.
- Deterministic checks run automatically. Certificate of analysis present? Component batches approved? Deviation closed? Market authorisation valid? These are rules, not judgements. Business rules evaluate them the moment data arrives so by the time a human looks, the green checks are already green.
- Humans focus on genuine judgement. Review by exception means the QP spends their time on the two checks that are amber, not the eighteen that are fine.
- The decision flows back automatically. A release is only useful if it becomes a usage decision, a stock posting and a certificate in the ERP without re-keying.
This is precisely the model SAP Batch Release Hub implements. Co-innovated with leading pharmaceutical companies and delivered as a SaaS application on SAP BTP, BRH is the central release platform, tightly integrated with the S/4HANA backbone and open via OData APIs to non-SAP systems.
BRH’s release check portfolio covers the full spectrum of what a QP verifies, with each check evaluated by configurable business rules so a policy change is a rule update, not a development project:
- Batch record review and certificate checks
- Quality control data and deviation / change control
- Component and batch genealogy (down to sub-component level)
- Regulatory data, jurisdiction control and market authorisations
- Serialisation data and temperature / transport records
- Artwork, safety labels and annual product quality review
- Clinical trial regulatory and technical checks
- Fully custom release checks for organisation-specific requirements
BRH covers every release scenario in the real supply chain across multiple business entities and QA departments, with role-based worklists, GxP-grade e-signature, full audit logs, and data freezing after release:
- Incoming goods release
- Bulk and semi-finished release
- Finished goods and CMO release
- Market release and distribution release
Once the QP signs, the release decision is sent back to S/4HANA automatically, updating batch characteristics, posting inspection-lot stock and generating the certificate.
The outcome is not just a faster process but a fundamentally more reliable one: standardised across the organisation, transparent to auditors, and scalable to batch sizes that manual release could never economically handle.
AI and Technology Transformation: The Batch Release Assistant
Automating deterministic checks solves most of the volume. It does not solve the part of release that is buried in unstructured documents the 200-page batch record, the CMO’s release dossier, the scanned COA, the paper checklist that a site has used for fifteen years.
This is where generative AI is now entering the release process, and where SAP has been notably careful to keep the human in the loop.
AI-assisted custom release checks. SAP Business AI in BRH analyses the documents attached to a custom release check and proposes results for each check item directly in the QP’s interface, with references back to the source. The QP reviews, verifies and signs — the role of final decision-maker never moves. SAP’s own estimate, based on customer feedback, is that a custom release check that took a QP around two hours can be completed in roughly forty-five minutes, an improvement of up to sixty percent in QP productivity. Just as importantly, it turns paper checklists into structured digital checks, using AI as the bridge to data sources that were never going to be integrated by API.
Fully automated release for low-risk scenarios. For release types where every check is deterministic and the risk profile allows it, BRH now supports configurable automated outcomes — a business-rules-driven release decision without human intervention and, where appropriate, without e-signature. This is the “zero-time” release that makes high-volume, small-batch models viable.
Serialisation as a release check. The evolving integration between BRH and SAP ATTP brings serialisation and aggregation status into the release cockpit, so that a batch cannot be released to a market whose track-and-trace requirements it does not yet satisfy.
Towards an agentic Batch Release Assistant. SAP’s stated direction is to evolve BRH into an AI-first release platform — with specialised agents for SOP interpretation, compliance checking, and release automation working alongside the QP. The design principle is scalable autonomy: the organisation decides, scenario by scenario, how much the system may do on its own.
Future Outlook
Looking towards the end of the decade, three developments seem likely.
- Release becomes continuous rather than terminal. Instead of a retrospective verification at the end of the batch, release readiness will be assessed in near real time as data arrives — a batch will carry a live “distance to release” throughout its lifecycle, and quality teams will intervene when a check turns amber, not when the batch is already sitting in quarantine.
- The QP becomes an exception manager and risk owner. As deterministic checks and AI-assisted document review absorb the routine workload, the QP’s role concentrates on judgement, risk assessment and the governance of the automation itself — deciding what the system may release autonomously and validating that it does so correctly. That is a more valuable use of a scarce skill, and arguably a more defensible one under Annex 16.
- Release, serialisation and supply chain converge. Jurisdiction control, supply chain maps, serialisation status and market authorisation data will be evaluated together at the moment of release, closing the loop between what a company is authorised to sell, what it has actually made, and where it is allowed to ship it. The companies that treat batch release, track-and-trace and regulatory data as one connected process will have a structural advantage in speed to market and inspection readiness.
The direction of travel is clear. What differs between organisations is how quickly, and how safely, they get there.
How Novatio Can Help
Novatio specialises in SAP track & trace, serialisation and regulated supply chain for life sciences — with batch release at the intersection of everything we do: the S/4HANA quality backbone, serialisation data in SAP ATTP, and industry cloud applications on SAP BTP.
- Assessment and business case. We map your current release process, quantify cycle time and inventory impact, and build the value case, identifying which release types are candidates for automation and which need AI-assisted or manual checks.
- Solution design and integration architecture. We design the end-to-end integration between BRH and your landscape – S/4HANA, LIMS, MES, QMS, DMS and SAP ATTP – bringing specific experience in the master-data and configuration decisions that determine whether a rollout is smooth or painful.
- Business rules and release check configuration. We translate your SOPs into configurable business rules on SAP BTP extending SAP’s preconfigured rule projects without touching standard content, so your logic survives SAP’s twice-yearly releases.
- GxP validation and compliance. We deliver the validation documentation, e-signature setup, role and access design, and audit-trail evidence required under Annex 11, and prepare your AI governance approach ahead of Annex 22.
- AI adoption and global rollout. We introduce AI-assisted release checks and automated outcomes in a controlled, risk-based sequence then scale from pilot site to multi-entity, multi-jurisdiction deployment with the change management and hypercare to make it stick.
Get in touch with Novatio to start the conversation.
Novatio — SAP Track & Trace, Serialisation and Batch Release specialists for Life Sciences.