30% improvement in Agent Efficiency for our Banking client using NLP

Use Case: Automated email categorization and management

Challenges: The customer support receives a large volume of emails daily that needs to be distributed among different departments based on the addressed issue. The manual reading and segregation of emails by support agent not only took a lot of time, typically 4 to 6 hours but also affect accuracy. This, in turn, hampers agent productivity and increases the response time resulting in customer dissatisfaction.

Solution:

We proposed a solution that incorporated Natural Language Processing (NLP) to read categorize emails and push them into the SFDC.

The incoming support emails are fed to the Machine Learning algorithms that classify them into different categories.

The Engagely email bot reads and understands the incoming support emails to classify whether the email is urgent, normal or junk.

Urgent emails are identified by the bot by recognizing the intent of the email, the email addresses, and keywords as specified by the client and the respective team will get alerts for the receipt of the same. The services requests, product enquiry, status enquiry, etc. are classified as normal.

These emails were then allocated to the respective team members based on their departments, which were pushed into the client’s preferred CRM system.

Results & Impact: The manual process of reading and categorizing emails and assigning it to the respective departments is now completely automated reducing hours of efforts to seconds. Engagely has enhanced agent productivity by 30% and improved customer satisfaction considerably with accurate and faster responses.

Engagely email bot also enabled the client to respond to 20% of the emails without any human intervention.

30% improvement in agent productivity using NLP
Enhancing banking customer support by automating email management and improving customer satisfaction