Data at Work

The left view illustrates approximately 1,200 systems and 6,000 interfaces. Finance, Sales, Marketing and HR depend on a heavily interconnected data landscape. A central Data Platform connects the landscape to two detailed lineage views on the right. The example demonstrates why uncontrolled connections make data difficult to understand and trust, slow business activities and increase operational costs.

Lineage explains where data comes from, where it goes and how it changes. This example traces ERP data through Azure and the GDAH platform's Bronze, Silver, Gold, Shared and Semantic layers into Power BI reporting. System layers provide an end-to-end view across ERP, data warehouses, master-data systems, platforms and reporting. Individual metadata objects link to glossary definitions, owners, data-quality results, systems, projects, ingestion schedules and classifications. Business processes connect to layers and objects to show how processes affect data and vice versa.

Connected One Data exchanges information to establish one definition, one quality and one truth. Source systems feed quality checks, the catalogue and configuration metadata. Configuration supplies ownership metadata to the platform. Quality checks and results flow between quality and platform solutions. The platform exchanges technical metadata and enrichment with the central catalogue and provides data to process analytics and reporting. Business intelligence supplies report metadata, process management supplies process metadata, enterprise architecture supplies business metadata, and protection supplies privacy and retention metadata to the catalogue. Enterprise architecture and IT service management exchange systems and interfaces, while architecture supplies process and retention context to protection. The catalogue supplies trusted metadata to data products combining catalogue, quality, lineage, and classification and retention. Metadata from connected systems supports a common end-to-end lineage tool. The source explicitly marks the lineage-tool context as planned and not yet approved.

The architecture connects nine layers. Value streams describe end-to-end operational flows supporting divisions. Business domains describe organisational areas and functions. Teams include Sales Analytics, Payroll, Talent Acquisition and Tax. Business capabilities describe core abilities, including data capabilities. Business processes describe workflows and procedures. Data domains group related data assets, and data entities are logical objects within those domains. IT systems, applications, interfaces and services connect technology assets, tables, attributes and services. Ownership and governance establish accountability and controls. Data as a Product spans the complete chain, supported by Governance, Data Quality, Data Operations and Data Protection.

Business Processes: Understand the process and information requirements. Data Dictionary: Agree business terms and definitions. Data Model: Define entities, relationships and structures. Create & Modify: Capture and update data at the source. Analyse & Profile: Identify quality patterns, issues and gaps. Ingest: Bring data into the required platforms. Harmonise: Align, validate and transform for consistent use. Consume: Use data in operations, reporting, analytics and AI. Retain & Dispose: Archive or delete under approved requirements. A simple view of how data moves from operational sources to business use, with controls throughout the Data Journey.

The ingestion process begins with Data Product Owners and Tactical Data Owners agreeing the need, ownership and controls. The ingestion team loads data into GDAH. Tactical Data Owners and Data Stewards confirm definitions, owners and tags in DAWISO. The ingestion team and Data Stewards set up checks and monitoring in Anomalo. The Data Product Owner confirms that data is ready for use. Data Stewards and the ingestion team then monitor issues, fix problems and update metadata. Ongoing maintenance spans governance, quality, operations and protection.