Saving time, eliminating hassles, and reducing TCO by 61%

This telecommunications giant saw an opportunity to enhance its customer care experience with better data-driven insights. The company was committed to reducing costs while analyzing its structured, semi-structured, and unstructured data to drive improved customer intelligence, loyalty, and predictive analysis. 

Unfortunately, the company’s data was distributed across several sources, which was less than ideal from an access management perspective.

The customer care department handled over 500,000 daily inbound calls and had to query 360 customer data, aggregated sentiment scores, device and accessory-related data, and more, from different platforms. The telecommunications company’s customer care data resided in multiple cloud and on-premise platforms, including Snowflake, Amazon S3, Azure Synapse, and Microsoft SQL Server. 

The data architecture team spent too much time copying the data from different platforms into Teradata so it could be analyzed. Copying and migrating data instead of connecting all of it directly to a single source system drove up operational and performance overhead costs and delayed time-to-insight. This delay reduced the company’s ability to effectively leverage customer care data and negatively impacted its NPS scores and customer churn rate.

The company looked at several data virtualization tools over three years to find the best data analytics engine. Most options were discarded because they required an overly complex data architecture, created unnecessary data duplication, or didn’t meet the time-to-insight requirements. 

The data architecture team felt open source Trino was their best option because they had success with it in the past. While Trino had many features the team needed, it lacked enterprise-level connectors and security.

The telecommunications organization selected Starburst Enterprise, a more feature-rich version of open-source Trino, which provided them with the ability to scale performance and provide a low-latency query access layer across several data sources.

Starburst also ensured secure access to data with strict global access control. Starburst’s enterprise-grade security includes features such as query auditing, data masking, and role-based access control. These allow the telecommunications company to control user access granularly, by data source, table, and row.

Starburst Enterprise gave the telecom provider the following results: 

Faster time-to-insight. Starburst Enterprise is more than seven times faster than open-source Trino. Queries that once took two hours now take only nine minutes. 

Insights without ETL. With 50+ enterprise connectors, Starburst enabled the telecom organization to connect to all of its data sources — Teradata, S3, MS SQL, ADLS, and Snowflake — through a single abstraction layer without moving any data at all.

TCO savings. By implementing Starburst Enterprise, the telecom giant reduced enterprise data warehouse spend, ETL, and labor costs, leading to total cost of ownership savings of 61%.

Improved business insights. The company expanded its Starburst-powered analytics engine across its supply chain, finance, and marketing departments, helping those business units optimize how they deliver insights back to the business. 

With Starburst Enterprise, the telecom provider achieved data virtualization and federation at scale and with the performance that it wanted, along with a low latency query access layer across its data sources. The faster, more efficient enterprise-grade technology gives the company the insights they need to meet their goals of reducing customer churn and improving their NPS scores.