We helped a leading provider of telecommunications services in analysing terabytes of data and detect anomalies in near real-time for a quick and accurate root cause analysis. This led to a better network performance and comprehensive improvement in the customer experience that helped them in reducing churn.
Identify data anomalies
Quick root cause analysis
Enhance customer experience
Multiple teams at the carrier analyse billions of records generated every day to understand and tune key drivers of customer experience. The operator also has a complex data pipeline with one of the largest Hadoop environments globally, DataWarehouse and BI reports serving 200+ users. With such data volume, a complex data flow and 1,500+ metrics to be tracked, it was essential for the operator to -
Have an automated & real-time view of data quality across multiple systems and teams
Systemic detection of anomalies in customer experience KPIs (Key Performance Indicators) daily
Automated root-cause analysis for any data qualty issues and anomalies
The consumption of mobile data is exploding, fuelled by video and other applications running primarily on smartphones. This demand is being met by Long Term Evolution (LTE) – the 4G wireless standard which is being deployed at mass scale by telecom operators throughout the world.
At the core of the evolution to latest standards is the overall mobile data experience, which telecom operators monitor by tracking KPIs related to their network. An example of such KPIs is aggregated time spent by users on the slower 3G/2G network frequencies due to any factor related to the network. Network operators treat this as a costly metric as a user ends up spending more time occupying space on the allocated frequency spectrum owing to the lower 3G/2G speed vs 4G speed.
This tier-one telecom operator analyses 12 billion records on network data every day from more than 40 different data sources about where, when, and how many subscribers use mobile services. It would take weeks before data teams found out about anomalies in data and understood the root cause. They needed an automated mechanism to detect data anomalies, understand key drivers and enable customer experience teams to take quicker and informed actions.
This telecom operator deployed PowerMe’s Intelligent Data Catalog (IDC) & Connected Data Quality (CDQ) facilitating an integrated view of the data and quality for their data teams
PowerMe CDQ utilizes a proprietary Machine Learning model in anomaly detection that automatically adjusts thresholds and predicts the future thresholds for the metric of interest. Users benefit by a smarter, faster response to data quality issues with PowerMe’s CDQ. It works with the Big Data and DWH environments and enables data teams at this tier-1 telecom operator to:
Monitor near real time data quality on a complex data pipeline
Configure automated quality rules computation to profile data and assess level of accuracy, validity, completeness & distinctness on attributes of interest
Detect anomalies at scale and perform drill-down analysis
With automatically curated metadata and lineage information from multiple data sources at the enterprise, available in a single platform – PowerMe IDC Data Analysts & Business users can
Trust insights better with a real-time view of underlying data quality for data assets and BI assets in a single connected platform
With Data Lifecycle Visualizer, traverse the Data Lineage for KPIs and view transformations and data quality metrics at multiple stages of data load and aggregations