Business Overview
Times Square Group, headquartered in Dubai, is a leading retail fashion conglomerate operating across the Middle East. With over 50 stores and 10 business entities, the group offers a wide range of products—from everyday fashion to luxury bridal couture. As part of its digital transformation journey, Times Square partnered with Logesys to modernize its data platform using Microsoft Fabric.
Current Scenario
Times Square’s operational data was spread across systems like Dynamics NAV and Business Central, which were primarily designed for transactional processing—not analytics. Reporting workflows were manual, slow, and error-prone, relying heavily on tools like Excel and PowerPoint. The lack of a unified, analytics-ready data platform made it difficult for business leaders to access timely insights.
Recognizing the need for a scalable, automated, and governed data infrastructure, Times Square decided to adopt Microsoft Fabric to unify its data estate and enable real-time reporting.
Project Objective
Logesys was tasked with designing and implementing a modern data platform that could:
- Consolidate data from NAV and Business Central into a unified lake-centric architecture.
- Automate data ingestion, transformation, and reporting workflows.
- Enable real-time, interactive dashboards using Power BI.
- Ensure data governance, security, and compliance across all layers.
- Build a scalable foundation for future analytics and AI initiatives.
Scope of Work
- Ingest raw data from NAV and Business Central into OneLake using Fabric Data Pipelines.
- Organize data using the Medallion Architecture (Bronze → Silver → Gold).
- Transform and enrich data using PySpark notebooks within Fabric Workspaces.
- Build Semantic Models to define business logic and KPIs.
- Visualize insights through Power BI dashboards connected to the Gold layer.
- Implement role-based access control and data lineage tracking using Microsoft Entra ID and Purview.
Data Flow Design & Technical Solution
The solution was built using the following components:
|
Component |
Purpose |
|
Fabric OneLake |
Unified data lake for storing raw, cleaned, and aggregated datasets. |
|
PySpark Notebooks
|
Automated data transformation, cleansing, and enrichment |
|
Data Pipelines |
Scheduled ingestion from source systems into OneLake. |
|
Semantic Models |
Business logic layer for consistent KPIs and metrics |
|
Power BI |
Interactive dashboards for real-time decision-making.
|
The architecture followed the Medallion pattern:
- Bronze Layer: Raw data ingested from NAV/BC into OneLake as Delta tables.
- Silver Layer: Cleaned and standardized data using PySpark (null handling, deduplication, joins).
- Gold Layer: Aggregated, business-ready datasets with KPIs and metrics for reporting.
Solution Design
Data Extraction
- Tool: Fabric Data Pipelines
- Description: Automated ingestion from NAV and Business Central into OneLake. Delta format used for schema preservation and incremental loads.
Data Transformation
- Tool: PySpark
- Description: Scalable processing of large datasets. Tasks included cleansing, normalization, enrichment, and joining with reference data (e.g., region, currency).
Data Warehousing
- Tool: OneLake (Lakehouse Architecture)
- Description: Centralized storage with support for versioning, partitioning, and schema evolution. Enabled seamless integration with Power BI and Purview.
Reporting & Visualization
- Tool: Power BI + Semantic Models
- Description: Dashboards built on top of Gold layer using semantic definitions. Ensured consistent KPIs across departments and near real-time insights.
Challenges & Solutions
|
Challenge |
Solution |
|
Data Integrity |
Rigorous analysis and normalization using PySpark to ensure schema accuracy. |
|
Data Quality & Consistency |
Validation protocols and metadata tagging across Silver and Gold layers. |
|
Scalability for Large Volumes |
Delta format and OneLake’s architecture enabled efficient handling of data. |
|
Governance & Access Control |
Role-based access via Entra ID and lineage tracking via Purview. |
Business Impact
- Unified data platform with lake-centric architecture.
- Automated data lifecycle management using Medallion Architecture.
- Scalable transformations with PySpark.
- Secure and governed access across departments.
- Real-time reporting with Power BI dashboards.
- Reduced manual effort and improved data accuracy.
Conclusion
Times Square now operates on a modern, automated data platform powered by Microsoft Fabric. Business leaders can access real-time insights, make faster decisions, and rely on consistent, governed data across the organization. The transformation has laid a strong foundation for future analytics, AI, and business growth.