Big Data Analytics & Real-time Reporting

Robert Half

Adanto significantly improves customer service and reduces operating costs for the global leader in the HR Consulting field by deploying real-time big data analytics and reporting for data streaming from Contact Centers in 15 regions around the world.

Environment

  • 15 global regions with Contact Centers in North America, EMEA and A/P.
  • 2,500,000 events monthly (chat, voice mail, telephone, service tickets)
  • Very complex data center environment with 30 disparate databases (cloud & on premise).
  • ShoreTel system (PBX Business Phone Systems) operated by 15 regions across the world with
  • CIC system (Customer Interaction Center) where calls are routed to the dedicated support queue.
  • Service Now Cloud (service management system)

Challenge

  • A lot of manual work analyzing what is happening inside each system in each Contact Center
  • Poor manual reporting or none-existing reporting from each system
  • Poor customer service not meeting given SLAs
  • Not understanding what is going on at Contact Centers
  • Hard to find information
  • Extensive delays in getting issues resolved
  • Customer dissatisfaction
  • Data Center complexity

Services performed

Big Data

Custom Application Development

DevOps

Security

Infrastructure Services

Administration Services

Salesforce

Amazon Cloud

Azure Cloud

Key goals

Real-time Analysis of the caller behavior for the Management

Real-time Reports based on caller behavior for call center management and stakeholders to make decisions

Real-time executive Dashboards with cloud access from mobile and PC

Solution

The main challenge to creating the requested result was the correlation of results between the source systems.  We have utilized time as the common key between systems.   We have deployed additional tools which will collect information from 3 non-integrated system sources and used our analytics algorithms to ensure the accuracy of reports. The hit ratio which we were able to achieve was around 89%, meaning that for a given set of 100 events from the first system we were able to correctly identify 89 correlated events from other systems.  With a daily average of around 700 calls, this hit ratio was well accepted.

Solution Architecture

Result

  • Reduced Abandoned Call rate from 18% to 5%
  • Reduced hold time from an average of 15 minutes to 4 minutes
  • Improved Management reporting accuracy, workload and reduced operating cost

Technologies used

  • ETL with Data Warehouse based on Pentaho ETL data integrator & MySQL
  • REST APIs for data access/refresh
  • Machine learning algorithms for data correlation and analytics
  • Amazon AWS & Microsoft Azure Cloud
  • Microsoft Power BI reporting tool from data stored in Azure cloud.