Smart Data Platform

The Smart Data Platform is an end-to-end analytics platform built on Microsoft Fabric that transforms raw data from Dynamics AX, Dataverse, Vesper market APIs, SharePoint, and internal data collection systems into business-ready reports and exports. It replaces a legacy PySpark notebook system with a modern, dbt-driven data warehouse deployed across three environments (DEV, UAT, PROD) using infrastructure-as-code.

What Does This Platform Do?

For business users, this platform is the single source of truth for operational and financial reporting at Geris. Instead of relying on scattered Excel files or manually maintained dashboards, all data flows through a governed pipeline that ensures consistency, traceability, and freshness.

Reports are delivered through a Fabric Workspace App -- a curated portal organized by business domain (Logistics, Trade, Finance, Supply, Management). You access it through your browser, and the data behind every chart and table is automatically refreshed from source systems.

The platform also provides automated Excel exports -- scheduled or on-demand reports delivered by email or downloadable from a web interface, with optional password protection.

How Data Flows Through the Platform

graph LR
    Sources["External Sources<br/><small>AX, Dataverse, Vesper,<br/>SharePoint, Datacollect</small>"]
    Bronze["Bronze Lakehouse<br/><small>Raw data via shortcuts<br/>and API ingestion</small>"]
    Gold["Gold Warehouse<br/><small>dbt transforms:<br/>staging, intermediate, marts</small>"]
    Semantic["Semantic Layer<br/><small>8 DirectLake models</small>"]
    Reports["Reports & Exports<br/><small>22 Power BI reports<br/>+ Excel exports</small>"]

    Sources --> Bronze --> Gold --> Semantic --> Reports

Data enters through the Bronze layer as raw tables, gets transformed in the Gold warehouse by dbt into business-ready dimensions and facts, is exposed through Semantic Models for self-service analytics, and finally consumed via Power BI Reports and automated exports.

Platform in Numbers

Where Should I Start?

Depending on your role, different parts of this documentation will be most relevant.

Developer

You build and maintain dbt models, semantic models, and reports. You work with feature branches, local DuckDB builds, and Fabric workspaces.

Start here:

• Developer Onboarding -- set up your local environment
• System Architecture -- understand the 4-layer data flow
• Data Flow Pipeline -- how each source system feeds the warehouse
• Technology Stack -- tools, versions, and key decisions

IT Admin

You manage infrastructure, security, deployments, and service principals. You care about Terraform, Key Vault, RBAC, and pipeline health.

Start here:

• System Architecture -- full system overview with all components
• Workspace Layout -- 4-workspace model and management methods
• Technology Stack -- infrastructure tooling and auth model
• FAQ -- secrets management, SPN details, credential rotation

Business User

You consume reports and exports to make operational and strategic decisions. You access the platform through the Fabric Workspace App.

Start here:

• FAQ -- where to find reports, data freshness, access requests
• Platform Overview -- this page, the section above

IT Manager

You need to understand the platform's architecture, security posture, and operational model without diving into implementation details.

Start here:

• Platform Overview -- high-level data flow
• System Architecture -- component overview and what does NOT exist
• Workspace Layout -- environment isolation model
• FAQ -- secrets, access control, audit trails

Quick Links