Shaman Atlas
Your most valuable content sits locked inside flat PDFs — value you can't reuse, govern, or scale. Atlas unlocks it as structured, machine-readable MLR knowledge — what every modern content operation is built on. As AI reshapes how content gets made, the teams that put it in place first will move fastest.
The insight
In pharma, approved content is treated as a flat PDF — a finished file to store and retrieve. But underneath, every asset is structured data: Medical, Legal and Regulatory claims, each with its references and footnotes.
Capture that as machine-readable knowledge and you have the foundation for content excellence — your MLR Knowledge Engine: the structured knowledge that powers authoring, control and mass update, AI content creation, and MLR pre-checks. More than somewhere to keep finished assets: it's the knowledge they're built from.
How it works
Building the MLR Knowledge Engine
A structured MLR library has always been too manual to create and keep current. Shaman solves that — turning the approved content you already own into a living knowledge engine.
Your approved content — from thousands of pieces to hundreds of thousands — sits in Veeva Vault PromoMats, but the data is locked inside flat layout files. Shaman Atlas Extraction Services re-engineers those PDFs into structured data — text blocks, grouping, classification, relationships and visual extraction. Start by migrating all your approved content; from there it stays current, as every newly approved piece is extracted automatically.
Extracted data flows into the MLR Knowledge Engine, which pulls authority data, legal requirements and your internal SOPs on a per-market basis. Approved content is merged in and matched to local requirements — canonicalised, deduplicated, enriched, with new variants added. Partly automated, always with a human in the loop.
The engine stays in lockstep with Veeva Vault PromoMats, so every element carries an approved status grounded in Veeva.
Using the MLR Knowledge Engine
Once built, the MLR Knowledge Engine is exposed across Shaman:
The outcome
By treating pharma content as structured data, Shaman automates everything that can be automated, assures compliance, and simplifies the process — so marketing and MLR teams focus on what adds value and what matters.
Trusted by the world’s leading life sciences organisations
FAQ
A governed store of the knowledge inside approved content — claims, references and mandatory statements — extracted from Veeva Vault PromoMats into structured, machine-readable data, scoped per market and brand.
Atlas Extraction re-engineers approved flat files into structured elements: classifying text and visuals, grouping claims with their evidence, and resolving references and footnotes — each with provenance back to its source Veeva document.
No. Veeva Vault PromoMats remains the single source of truth. Atlas holds knowledge derived from approved documents and stays in lockstep with Veeva as content is approved and expires.
AI composes from the Knowledge Engine — approved claims with bonded references and market-matched mandatory statements — so output is traceable assembly with provenance, not plausible free text.
Discover how to produce better content, faster — and take back control.
