Case Study

Sense and Response Bot

CASE STUDIES

Sense and Response Bot

Service Center operational efficiencies which impact a provider’s NPS scores are critical for any enterprise. Service Center/Contact Center Analytics is contributing significantly towards metrics-driven optimization in this space. Though technology is a non-trivial investment in this space, the bulk of the cost is attributed to people. Hiring, training representatives to answer those calls and represent the company is a continuous challenge as attrition is high very high. Any improvements a company can make to improve the job satisfaction of their service center representatives can pay back huge dividends in terms of cost savings and NPS score.

Most of us who have ever called a Service Center before probably heard the representative complain about their computers. “My computer is slow”, “our systems are acting up”, “I don’t understand why this is taking this long” etc are common phrases we hear. Some of that might just be a way to fill the awkward silence while they try to get go through the process, but there might be instances where they are truly facing issues. What if we can find patterns and proactively sense system issues and take corrective action before it becomes a real problem?

SENSE AND RESPONSE BOT

SENSE AND RESPONSE BOT:

Qualigy Tech got an opportunity to help a customer in this space by building a Bot capable of interpreting internal communication between representatives to identify patterns and check the systems they are using. Bot we built will passively monitor the messages and apply NLP and NLU algorithms to trigger on keywords and identify the sentiment/mood of the conversation. If they detect a concern about a particular process or system, Bot will check against systems availability metrics and monitoring endpoints to test and verify. Bot will take corrective action, where permitted and available, to fix issues or create a ticket for support teams to investigate and mitigate.

Bot built with cloud-native services

  • Amazon Lex
  • AWS Amplify for UI, access to supervisors and admins
  • Integration with OAuth provider for authentication
  • Hook into messaging streams
  • Social channels for external chatter as an option
  • Amazon Polly for NLP
  • Amazon Comprehend for NLU
  • Integration with monitoring API and endpoints for systems
  • Trigger remediation scripts as defined by admins
  • Integration with ServiceNow to create tickets for action and reference
  • Dashboards with metrics

Get In Touch