SAP Datasphere provides a single semantic layer that keeps data where it is, adds business context, and delivers trusted information to analytics, planning and AI — now strengthened by Horizon and Joule.
Modern organisations generate more data than ever, yet too much remains locked in isolated systems. SAP Datasphere provides a single semantic layer that keeps data where it is, adds business context and delivers trusted information to analytics, planning and AI workloads. Recent updates such as the Horizon user experience and the Joule generative-AI assistant make it an even stronger foundation for fast, well-informed decision-making.
The hidden cost of fragmented data
Multiple clouds, on-premises databases and SaaS applications have multiplied data stores. Analysts spend a significant share of their time reconciling figures instead of acting on them, while disconnected definitions encourage 'spreadsheet logic' that undermines trust. Poorly governed data also feeds inaccurate prompts into generative-AI models. Datasphere addresses these risks by exposing a live, context-rich 'Business Data Fabric' without heavy replication.
What sets SAP Datasphere apart
Business Data Fabric — federated access keeps source systems in place and removes mass data movement, while a business semantic layer preserves calculations, hierarchies and currency logic so users query terms like 'Net Sales' rather than table names. Governance and lineage — integrated lineage tracking and optional Microsoft Purview integration give auditors a clear view from a dashboard value back to the original transaction. An expanding connector library — over twenty new connectors, including MongoDB Atlas and Google BigQuery, bring modern data lakes into the platform.
What's new in 2025 — Horizon and Joule
The Horizon interface introduces a cleaner palette, simplified icons and higher-contrast components, shortening navigation time and improving accessibility. Joule gains a 'prompt-to-model' capability that translates a plain-language description of a data domain into a first-pass semantic model in minutes — cutting modelling effort by roughly forty per cent in early adoption programmes — and orchestrates multi-step tasks across engines such as Microsoft Copilot and Google Gemini. Together, these features turn Datasphere into an AI-ready decision layer.
Reference architecture
Source connections (ERP, CRM, IoT feeds, files, data lakes) feed the Business Data Fabric, which handles federation or selective replication and data-quality rules. A semantic model defines business logic, hierarchies and currency conversions once and reuses them everywhere. The consumption layer serves SAP Analytics Cloud via live connection, Joule natural-language queries, and third-party BI tools via ODBC/JDBC — each department getting its own view while data stays consistent.
Conclusion
SAP Datasphere brings scattered data into a single, context-rich fabric that serves real-time analytics and generative AI. The sooner your information speaks the same language, the faster your organisation can listen, learn and act.
