Organizations receive communication from their customers through various channels like emails, phone calls, and job applications. It’s important for these organizations to act on these communications quickly and accurately. Manually handling this scenario can be time consuming and error prone. Building an automated, intelligent system to handle customer communications is needed for customer satisfaction and speedier resolutions. This code pattern provides a methodology for handling emails from customers in an automated and smarter way.
In the Provide automated customer support for emails code pattern, we use customer support in the telecommunications domain as an example. We look at request scenarios for enabling a service, disabling a service, changing plans, and adding family members to a plan.
With customer support, you should:
- Know the intent of an email
- Know the information available in these emails
- Identify missing information
- Auto-compose and send responses
This code pattern provides an automatic and cognitive way to achieve these requirements. It uses natural language processing for emails, understanding the intents of emails, auto-composing responses, and providing a dashboard with a high-level summary of intents and emails. While the use case shown is for a telecom domain, it can be applied to any domain. It integrates with a database that acts as a CRM to pull customer information to validate emails and requests. It uses Watson Knowledge Studio for custom domain natural language processing and Watson Natural Language Understanding to deploy custom domain models. It uses Watson Natural Language Classifier to get the intents of emails, an IBM Cloudant database to store emails and customer data, and Node-RED to integrate with emails.
Take a look and let us know what you think!