The label is the truth.
Pharma product data includes SmPC sections, patient leaflets, FMD serialisation references, and MDR device identifiers (UDI). Every change is a signed change request, not a merchandiser click.
Your PIM must support signed workflows, full audit trails, and retention windows that match your pharmacovigilance SOPs. The general rule: 10-year retention for human medicines, 7+ years for medical devices, longer in some jurisdictions.
The SmPC (Summary of Product Characteristics) and PIL (Patient Information Leaflet) are structured documents with numbered sections. Treat them in the PIM as section-level attributes — section 4.3 Contraindications is one attribute, section 4.8 Undesirable effects is another — not as a single blob of text per locale.
- Signed approvals per attribute change, with reviewer role, timestamp, and reason code.
- Change control — every edit reviewable as an auditable workflow artifact.
- Retention — minimum 10 years for human medicines, enforced automatically.
- E-signatures — Part 11 / EU Annex 11 compliant signing for regulated flows.
Machine-readable regulatory content.
EMA's electronic product information (ePI) initiative is progressing from pilot to mandatory during 2026–2028. ePI is published as FHIR-structured content — the patient leaflet in a machine-readable form that pharmacy dispensing systems and patient apps can consume.
MDR (EU Medical Device Regulation) similarly mandates UDI (Unique Device Identification) on every device, registered in EUDAMED. Both of these mean: your PIM has to emit structured regulatory output, not just PDFs. The RFP question is whether this is native or an export script.
The SKU is not the unit — the pack is.
Under FMD (EU Falsified Medicines Directive) and the US DSCSA, every individual pack of a prescription medicine carries a unique serial number. These numbers are aggregated into cartons, cases, pallets — a full track-and-trace chain from manufacturer to pharmacy.
This is not strictly a PIM problem (serialisation lives in a Level-4 system like TraceLink or SAP ATTP), but the PIM has to own the product-level identifiers — GTIN, NTIN, MPN, batch-lot template — that anchor the serialisation chain. Mismatches between PIM and serialisation system are the most common root cause of pack-level failures.
Two very different content disciplines in one catalog.
A brand selling both OTC and prescription products runs two parallel content disciplines. OTC operates under consumer-goods rules — marketing copy, retailer content scoring, Amazon syndication — while Rx operates under full pharmacovigilance governance with every word requiring medical-affairs sign-off.
Your PIM should support a dual-lane workflow: the OTC lane moves quickly, the Rx lane moves deliberately, and attribute-level rules prevent OTC-style copy from reaching a prescription channel by accident.
- Product class as a first-class attribute — OTC / Rx / device / cosmetic / supplement — drives workflow routing.
- Reviewer roles differ by class: OTC approved by brand ops, Rx by medical affairs plus regulatory.
- Channel filters — Rx products must not be syndicated to Amazon/Bol, enforced by the platform not by memory.
- Retention and audit inherited from the class, not set per SKU.
EMA requires 24 languages.
An EMA-authorised medicine must carry its SmPC and PIL in all 24 official EU languages. Translation here is not a marketing task — it is a regulatory deliverable, with strict terminological fidelity and legal review in each target language.
The PIM's job: hold the structured master (usually EN), track a per-language translation state per section, route translations through approved linguistic service providers, and enforce that a locale ships only when every required section is approved. TMS integration is table stakes; regulatory-grade TMS integration (Lionbridge, TransPerfect, Welocalize) is often the differentiator.
Medical devices, same catalog, different rules.
If your catalog includes medical devices (a class-IIa wearable, a class-III implant), MDR applies instead of the medicines regime. UDI-DI / UDI-PI identifiers, EUDAMED registration, a technical documentation file per device, and post-market surveillance data all intersect with the PIM record.
Device classes I through III carry escalating documentation requirements. Your PIM model should represent device class explicitly and drive required-attribute sets based on it. Devices get tripped up most often on the basic UDI-DI / UDI-PI distinction; model that field clearly and enforce its format.
Vendors with real pharma references.
Pharma shortlists favour vendors with genuine life-sciences reference customers and MDM heritage. Confirm vendor fit via the PIM Shortlist tool and ask any candidate for three regulated-industry references before demo.