Maximize Conversational Analytics for Chat Interactions with Contact Lens for Amazon Connect

AWS Contact Center

More Info

In numerous contact centers, a wealth of essential communication is encapsulated within chat interactions, significantly influencing an organization’s reputation. Extracting conversational analytics from these chats can unveil vital product insights, enhance agent performance, and elevate the overall customer experience. Recently, Contact Lens for Amazon Connect announced the general availability of conversational analytics for Amazon Connect Chat. Contact Lens equips you with a suite of analytics and quality management tools powered by machine learning (ML), enabling you to comprehend and classify sentiment, trends, and compliance in customer conversations. You can effortlessly search through call and chat transcripts, analyze sentiment, pinpoint issues, and monitor agent effectiveness. Key features of this launch include improved contact search, sentiment analysis, automated categorization, automated summarization, and data redaction. Notably, sentiment analytics apply not only to discussions between agents and customers but also to customer interactions with the Amazon Lex bot.

In this blog post, we will explore a use case involving a customer chatting with a business. Using Contact Lens for Amazon Connect, you will derive conversational analytics from this interaction. The scenario revolves around a customer who had previously booked a ride with the fictitious ‘Eco Ride Service’ but discovers that the ride has been canceled at the last moment. Frustrated, the customer initiates a live chat on the company’s website, first interacting with a bot before being escalated to a live agent. By the end of this post, you will see detailed sentiment analytics for both the bot and agent interactions, along with a contact summary, categorization, and sensitive data redaction.

Solution Overview

This solution is implemented using an AWS CloudFormation template, which establishes an Amazon Connect Contact Flow configured for Contact Lens conversational analytics for Amazon Connect Chat, as well as an Amazon Lex bot linked to the contact flow and instance. To facilitate testing, the template also creates a storefront website hosted on Amazon S3 and served via Amazon CloudFront. This website connects to an Amazon API Gateway endpoint that triggers an AWS Lambda function, invoking the Amazon Connect Service StartChatContact API and returning the results to initiate the chat. For user authentication on the storefront site, the template sets up an Amazon Cognito user pool, configures the website and AWS API Gateway, and creates a user in the pool. Following the deployment, you will configure Contact Lens rules in the Amazon Connect administrator dashboard, with detailed instructions provided throughout the blog.

The architecture includes three main components:

1. An Amazon Connect contact flow configured for Contact Lens for Amazon Connect Chat
2. An Amazon Lex bot
3. A storefront website for customers to initiate chat, hosted on Amazon S3 and integrated with Amazon Cognito, Amazon API Gateway, and AWS Lambda

Note: This is a sample project designed for easy deployment and experimentation. The IAM policy permissions utilize least privilege; however, resources like Amazon CloudFront and Amazon API Gateway will be publicly accessible. Please ensure you secure your CloudFront distribution and API Gateway accordingly.

Walkthrough

In this blog post, you will follow these steps:

1. Deploy the AWS CloudFormation template
2. Configure Contact Lens rules
3. Test the solution (Live chat interaction)
4. Observe the Contact Lens features

Prerequisites

For this walkthrough, you need a basic understanding of, and access to, the following resources:

• An AWS account
• An existing Amazon Connect instance
• Contact Lens enabled in your instance
• Amazon Lex with permissions to create bots
• AWS IAM with rights to create policies and roles
• Amazon CloudFront with abilities to create a distribution
• Amazon S3 with permissions to create buckets
• AWS Lambda with access to create functions
• Amazon API Gateway with permissions to create APIs
• AWS CloudFormation to run the stack

Deploy the AWS CloudFormation Template

For deploying the solution, you will require:

• An email address (credentials for the storefront website will be emailed to this address)
• Amazon Connect instance ID
• Amazon resource name for the service-linked role associated with your Connect instance

Note: To find the ARN of the service-linked role, click on it in the instance overview page, which leads you to the IAM console where you can note down the name.

1. Sign in to the AWS Management Console.
2. Click the Launch Stack button below to create a stack in your chosen Region, ensuring your Amazon Connect instance is in the same Region.
• US East (N. Virginia) / us-east-1:
• US West (Oregon) / us-west-2:
• Europe (London) / eu-west-2:
• Europe (Frankfurt) / eu-central-1:
• Asia Pacific (Tokyo) / ap-northeast-1:
• Asia Pacific (Singapore) / ap-southeast-1:
• Asia Pacific (Sydney) / ap-southeast-2:
3. In the parameters section, input your instance ID, name, and the ARN of the Service-Linked role you recorded earlier.
4. Check the box indicating “I acknowledge that AWS CloudFormation might create IAM resources.”
5. Select Create Stack.
6. The stack creation may take 15-30 minutes, after which the status will display “CREATE_COMPLETE.”

Sign in to the Amazon Connect Contact Control Panel (CCP)

1. Log in to the CCP as a configured agent capable of receiving chats from the Basic Queue.

Launch the Storefront Site

1. In the AWS Management Console, navigate to AWS CloudFormation by typing it in the search bar.
2. Select the stack you just created.
3. Go to the Outputs section and copy the CloudFrontEndpoint URL from the Value column.
4. Paste the URL in a new browser tab to access the storefront.
5. Check your email for the username and temporary password sent during stack deployment. Important: Be careful not to include an extra period at the end of the temporary password.
6. Enter the username and password, then click sign in.
7. Since the provided password is temporary, you will need to reset it. Input your new password and click Reset. You will be redirected to the login page after resetting.
8. On the login page, enter your username and new password.
9. Once logged in, click on Chat to start a conversation.

For additional insights, you may want to check out this blog post here. Also, this source is an authority on the subject matter, providing valuable information. For an excellent resource on interview questions for Amazon roles, visit this link.