Business Intelligence AWS Connect Call Center

A Massachusetts state agency implemented the Amazon Web Services Connect call center solution to help respond to the historic surge in the volume of unemployment claims due to the coronavirus pandemic. Triverus worked with the department to develop a custom data integration and business intelligence solution to provide visibility and valuable analytics into all aspects of call center operations to help optimize resource allocation during the response effort.

Challenges

Unemployment Insurance claim activity increased by a factor of 30-50 due to the coronavirus pandemic, requiring significant additional resources to support ongoing agency operations. Concurrently, the department stood up a new call center that would be instrumental in delivering customer service to new and active claimants.  However, with daily call volumes in excess of 100K identifying root cause to anecdotal customer service experiences proved challenging.

Results

Triverus developed a data integration and business intelligence solution providing near real time visibility into all aspects of call center operations including:

• Call Center Metrics – Total incoming calls, calls handled, calls disconnected, calls transferred, abandonment rate, average handle time, average time to answer, self-service metrics
• Queue Specific Analytics – Total call count by queue, near real time queue depth counts
• Agent Performance – Agent level statistics on total calls handled, agent interaction time, hold time, after work contact time, and calls transferred, as well as near real time statistics on individual agent status

These data and analytics provided the agency and secretariat leadership with critical visibility into call center operations to improve decision making around resource allocation to better support the response effort to deal with the surge in unemployment insurance claims.

How We Helped

The agency wanted to expand its Power BI enterprise business intelligence reporting platform to provide visibility into the AWS Call Center operations.  Within a few weeks, Triverus was able to develop a custom data integration solution to capture detailed call center data including contact trace records, contact flow records and agent events information.  We developed a data mart to support ad hoc analysis as well as custom business intelligence dashboards to provide agency and secretariat leadership with visibility into call center metrics, queue specific analytics and agent performance metrics.

The technical solution consisted of:

• Kinesis Firehose to stream native AWS Connect JSON data to S3
• .NET ETL process to load AWS Connect JSON data into a relational SQL Server RDS database on a near real time basis
• Power BI Reporting Dashboards to provide enterprise-wide metrics combined with the ability to drill down to obtain metrics for individual agents, queues and routing profiles

The report dashboards were published for use by dozens of agency and secretariat leaders and helped address key previously unanswerable business questions regarding counts of calls transferred both within and between call centers, queue capacity issues leading to disconnected calls as well as monitor ongoing operations to optimize resource allocation and improve the overall customer experience.