case study

JM Bullion

50% Faster Analytics via Data Warehouse Implementation

Manufacturing Ecommerce

Business Impacts

750+

tables migrated from Aurora to Redshift warehouse

50%

reduction in data refresh on dashboards

Customer Key Facts

  • Country : United States
  • Industry : Ecommerce

Problem Context

JM Bullion, an e-commerce retailer specializing in precious metals and a subsidiary of A-Mark Precious Metals, a Fortune 500 company, played a crucial role in both retail and wholesale markets. Operating in the direct-to-consumer sector, JM Bullion managed multiple e-commerce websites, each with its own database. This fragmented structure posed significant challenges in financial reporting, compliance, and leveraging market-driven pricing opportunities. To address these issues, JM Bullion aimed to build a dynamic and adaptable data warehouse pipeline. The primary objective of this pipeline was to facilitate data-driven decision-making and formulate customer-centric strategies.

Challenges

  • Lack of data pipelines and reporting in Quicksight, restricting data driven decisions
  • Unconsolidated data across different sources with columnar storage of data in json and text format
  • Significant delay in refreshing when merging log tables with base tables in QuickSight

Technologies Used

Amazon S3

Amazon S3

Amazon Redshift

Amazon Redshift

AWS Glue

AWS Glue

Amazon Quicksight

Amazon Quicksight

AWS DMS

AWS DMS

AWS Glue Crawler

AWS Glue Crawler

Amazon SNS

Amazon SNS

AWS IAM

AWS IAM

AWS KMS

AWS KMS

AWS Secrets Manager

AWS Secrets Manager

AWS Security Hub

AWS Security Hub

AWS Inspector

AWS Inspector

Solution

Building Data Warehouse Pipeline For Enterprise Analytics

Quantiphi team meticulously designed and implemented a comprehensive data pipeline, encompassing ingestion, transformation, staging, and warehouse layers. Leveraging DMS, Quantiphi efficiently migrated data from the Aurora data source to S3 and applied essential transformations using Glue scripts. Following this, the data was transferred to the Staging layer (Redshift) and organized into Facts and Dimensions tables, as initially defined during the initiation phase. With the warehouse layer in place, a reporting layer was constructed, employing materialized views to support Quicksight dashboards, henceforth facilitating downstream analytics.

Results

  • Single source of truth for data is created by consolidating the information across individual tables and merging the historical information
  • Scalable and secure solution which ensures new data source (post company acquisition) can be added to the warehouse
  • Consolidated dashboards created with governace enabled for materilized views to ensure singularity of reporting data tables
  • Enabled business teams to publish their segregated business insights, allowing them to generate their own reports.
  • Enabled QBR/MBR Leadership reporting

Start Your Next Gen AI Journey Today

Discover how Quantiphi’s AI-powered solutions can transform your business. Fill out the form, and we’ll help you explore tailored AI strategies to unlock new opportunities for growth.

Thank you for reaching out to us!

Our experts will be in touch with you shortly.

In the meantime, explore our insightful blogs and case studies.

Something went wrong!

Please try it again.

Share