Enterprise-Grade People Data: Governance, Quality and AI-Ready Foundations
Enterprise-Grade People Data: Governance, Quality and AI-Ready Foundations
How Novartis built a trusted, scalable people-data domain
Novartis spent seven years transforming scattered, inconsistent people data into a coherent, enterprise-grade domain aligned with central data strategy. This session shows how governance roles, standards, lineage and ownership were designed, tested and scaled across regions and functions. It also explains how People Analytics reshaped its operating model and how these foundations now support workforce planning, cost optimisation, decision-making and AI readiness across the organisation.
This session explores
- Establishing governance roles, ownership models and federated stewardship.
- Designing quality standards, definitions, controls and lineage tracking.
- Aligning people data with enterprise data strategy and architecture.
- Restructuring People Analytics for sustainable operations and adoption.
- Preparing people data for high-stakes AI and regulatory requirements.
Learning outcomes
- Understand how to structure governance that scales across HR, Finance and data teams.
- Know which quality standards and controls matter most for audit-ready people data.
- Identify how to align HR data with enterprise platforms without slowing delivery.
- Learn how a federated stewardship model improves ownership and consistency.
- See how trusted people data accelerates planning, cost and AI-readiness decisions.
26 February 2026
13:40-14:10 CET
Ashish Pant
Global Head of People Analytics and Data
Why this matters
Organisations are under pressure to meet CSRD-level reporting standards, deploy AI responsibly, and make faster workforce decisions in volatile markets. These expectations require people data that is accurate, governed, lineage-tracked and interoperable with enterprise systems. Large companies often struggle with fragmented ownership and inconsistent definitions. This session addresses how to build data that is trusted, auditable and ready for both regulatory and AI-driven use.