The InfoCepts’ team empowered the customer’s users with a high-performance, self-service analytics platform. It gives them access to certified data sets, near real-time data feeds, and a unified semantic layer with self-service visualization capabilities.
InfoCepts started the engagement by conducting a two-week analytics assessment workshop with the customer’s business and IT teams. Our goal was to:
- Understand the current and desired states
- Get a walk-through of the business process
- Collect information about key challenges users face
- Get acquainted with the current technology landscape
- Review current data extraction and Excel generation processes
Our team collaborated with customer IT stakeholders to finalize the reference architecture and overall solution approach (e.g., define the target data model and data flow, integration techniques for disparate data sources). A cloud-first strategy enables agility with the best balance of cost, performance, and long-term analytics. As part of its execution, our InfoCepts team ingested and processed data using dynamic clusters, then built and loaded the data warehouse as per the data model.
In parallel, we created a KPI catalog for each subject area and initiated reviews with business users. This was to ensure 100% coverage of both the base and derived fields required for analysis. From there we created a unified semantic layer in a reporting tool that provides canned reports and self-service analytics.
The InfoCepts team also leveraged its proprietary accelerator, AQuA (Automated Quality Analyzer) to improve the speed and accuracy of testing. With wider testing coverage and ability to proactively identify data inconsistencies, the KPI owners gained more trust in the solution.
We then enabled users (consumer and analysts) with multiple knowledge elements to democratize insights. Here we:
- Conducted virtual sessions to educate all users about the system, data, and reports
- Conducted a two-day training workshop for analysts focused on self-service capabilities
- Published report templates for guided ad hoc analysis
- Provided a walk-through of the data dictionary and model available for analysis
Additionally, we set up a sandbox environment for data science users. Here they have ~36 business views and can process additional data, integrating it with an R tool for their advanced analytics needs.
Overall, the new platform processed and transformed 50+ tables, ~4 TB of data, and three years of history across four subject areas for the customer’s business needs. The reporting semantic layer enabled ~110 fields (in the form of dimension and reporting measures) for approximately 50 users.