Viewership Dashboard helped Telekom Malaysia Enable Near Real-Time Insights for Better Decision-Making

Telekom Malaysia Berhad (TM) is nation’s leading integrated telecommunications provider that is on a continuous journey of enhancing and elevating the nation’s telecommunications technology and services toward delivering connectivity to all Malaysians. They approached Blazeclan for help in building a reporting solution that performs sentiment analysis on Twitter and community forum data in near real-time.

The Challenge

TM was looking into a comprehensive solution to enable near real-time & better visibility of TV viewership in view of their ever-increasing data. By understanding the customers’ viewing behavior, TM intended to make data-driven decisions to grow their business.

Key Analysis Areas Included

• Actual user engagement at a particular hour which can be drilled down to respective channels. 
• Visibility of new or repetitive users based on daily interval.
• Time vs. day user engagement to understand user behavior & identify peak viewing hour/day. 
• User engagement duration with a particular channel (average time spent)

The Solution

TM partnered with Blazeclan in developing a Near Real-time Viewership Dashboard. Blazeclan came up with a solution leveraging AWS Glue & AWS Lambda for processing the voluminous data. The approach involved the use of multi-node AWS Redshift cluster. This cluster brought scalability & efficiency in query resolutions for TM. For visualization purposes, AWS QuickSight was used in combination with SPICE for a rapid, real-time, in-memory calculation & analysis. This was engineered for rapid, advanced calculations & providing near real-time data for the dashboard.

The Approach

• AWS infrastructure implementation.
• Data ingestion and data preparation.
• Dashboard development.
• Data warehousing.
• Iteration, evaluation, and streamlining.

Architecture Diagram

Benefits Achieved by TM

Actionable Insights: The cloud-based data pipeline and BI dashboard enabled TM  to get actionable insights on their user viewing patterns and trends for making better decisions and bringing overall improvement in end-user experience.

Serverless Architecture: TM was offered with a serverless architecture that enabled greater scalability, high flexibility, and faster time-to-release. They were able to achieve all this while maintaining cost-effectiveness.

High Performance: The solution offered TM with the ability to perform complex calculations at scale. This was done by using an in-memory engine, SPICE, for faster load times and automated, periodic data refresh.

Tech Stack