Cognitive Customer Care – Reference Architecture for Amazon Connect

By and Phani Lingamallu | 9 minute read | October 25, 2021

IBM, together with AWS developed a Cognitive Customer Care Reference Architecture to help clients navigate discussions and design phases during planning or delivery stages of a Cognitive Customer Care implementation.

The reference architecture covers the following areas:

  • Customer interaction channels – synchronous & asynchronous, digital / voice channels
  • Agent experience and workspace – agent desktop
  • Core customer care platform – core capabilities such as routing, skills, flows, IVR, workforce management and others
  • AI services – cognitive services supporting contact automation
  • Data platform – data processing and reporting related to contact centre management and operations
  • Enterprise resources – data and services provided by enterprise specific systems to support contact interactions; typically, this includes customer data, accounts, marketing, sales, identity management and other systems of records
  • External resources – data and services provided by third party providers such as payment gateways, blockchain platforms, traffic information and similar
  • IoT event sources – many organizations consider signals provided by their customer devices as a valuable information; typically, this may include IoT enabled devices such as wash machines, fridges, smartphones, routers and similar.

The diagram below depicts these described areas and can be used as an input to structure and design discussions and activities.

Amazon Connect specific view of the Reference Architecture

AWS provides a portfolio of services which Amazon and IBM mapped to the respective areas of the reference architecture, providing guidelines for implementation of the IBM Cognitive Customer Care service with Amazon Connect and associated services. See the following diagram:

Here are the components used for the Amazon Connect-based IBM Cognitive Customer Care Reference Architecture implementation:

Amazon Connect The core contact centre platform. Detailed info available here.
Connect CCP Agents use the Amazon Connect Contact Control Panel (CCP) to interact with customers. It’s how they receive calls, send chats, make transfers, and perform other key tasks.

It is part of the Amazon Connect platform.

Connect Customer Profile Amazon Connect Customer Profile aggregates customer data with built-in connectors for third-party applications like Salesforce, ServiceNow, Zendesk, and Marketo. When a customer calls in, Customer Profiles scans and matches phone numbers or customer IDs to customer information located in connected applications and surfaces a unified profile to the agent in real-time. Customer Profiles combines contact history information from Amazon Connect (e.g. number of holds, transcripts, customer sentiment) with customer information from CRM systems, and e-commerce and order management applications (e.g. phone number, address, current order status). With Amazon Connect Customer Profiles, agents have all the customer information they need in a single place, so they can deliver more personalized customer service.

It is part of the Amazon Connect platform.

Contact Lens Contact Lens for Amazon Connect enables you to better understand the sentiment and trends of customer conversations to identify crucial company and product feedback. You can also track the compliance of customer conversations in your contact centre to ensure standard greetings and sign-offs are used, help train agents, and replicate successful interactions. Supervisors can conduct fast full-text search on all transcripts to quickly troubleshoot customer issues. In addition, with real-time capabilities, you can get alerted to issues during live customer calls and can deliver proactive assistance to agents while calls are in progress, improving customer satisfaction.
Kinesis Amazon Kinesis makes it easy to collect, process, and analyse real-time, streaming data so you can get timely insights and react quickly to new information. Kinesis will help in real time streaming of Audio data and support any requirements of translation as required.
Lambda AWS serverless compute infrastructure  enables integration to any external systems like CRM, System of Records etc.
Managed blockchain Amazon Managed Blockchain is a fully managed service that makes it easy to join public networks or create and manage scalable private networks using the popular open-source frameworks Hyperledger Fabric and Ethereum.
Directory service AWS Directory Service for Microsoft Active Directory enables your directory-aware workloads and AWS resources to use managed Active Directory (AD) in AWS. It is built on actual Microsoft AD and does not require you to synchronize or replicate data from your existing Active Directory to the cloud. You can use the standard AD administration tools and take advantage of the built-in AD features, such as Group Policy and single sign-on.
QuickSight QuickSight lets you easily create and publish interactive BI dashboards that include Machine Learning-powered insights. QuickSight dashboards can be accessed from any device, and seamlessly embedded into your applications, portals, and websites.  This will help in creating Dashboard to provide an overview of Agent performance, Call centre Performance etc..
Athena Amazon Athena helps in quickly query your data without having to setup and manage any servers or data warehouses. Just point to your data in Amazon S3, define the schema, and start querying using the built-in query editor. Amazon Athena allows you to tap into all your data in S3 without the need to set up complex processes to extract, transform, and load the data (ETL).
Data Lake Amazon data lake is a centralized repository that allows you to store all your structured and unstructured data at any scale. You can store your data as-is, without having to first structure the data, and run different types of analytics—from dashboards and visualizations to big data processing, real-time analytics, and machine learning to guide better decisions.
S3 S3 provides, in the context of a CC solution, functionalities to store Audio and Chat conversations and also act as a repository to store all the data related to Amazon Connect.
Lex Amazon Lex is a service for building conversational interfaces into any application using voice and text. Amazon Lex provides the advanced deep learning functionalities of automatic speech recognition (ASR) for converting speech to text, and natural language understanding (NLU) to recognize the intent of the text, to enable you to build applications with highly engaging user experiences and lifelike conversational interactions.
Poly Amazon Polly, a service that converts text into lifelike speech using Speech Synthesis Markup Language (SSML)  [Amazon Polly Developer Guide].
Textract Amazon Textract is a machine learning service that automatically extracts text, handwriting and data from scanned documents that goes beyond simple optical character recognition (OCR) to identify, understand, and extract data from forms and tables.  This will support in any attachments shared by customer to be read automatically and provide enriched information to the agent.
Connect Wisdom Amazon Connect Wisdom helps agents use ML to search across connected repositories based on phrases and questions exactly as the customer would ask them in order to quickly find answers. Wisdom helps customers connect relevant knowledge repositories with built-in connectors for third-party applications like Salesforce and ServiceNow, also internal wikis, FAQ stores, and file shares. In addition, Wisdom uses real-time speech analytics to detect customer issues during calls and provide agents recommendations and answers. Wisdom helps provide faster issue resolution and improved customer satisfaction.

Unified ML-powered knowledge search and real-time customer recommendations.

SES (email service) Amazon Simple Email Service (SES) is a cost-effective, flexible, and scalable email service that enables developers to send mail from within any application. You can configure Amazon SES quickly to support several email use cases, including transactional, marketing, or mass email communications.
SNS (notification service) Amazon Simple Notification Service (SNS) enables you to send messages or notifications directly to users with SMS text messages to over 200 countries, mobile push on Apple, Android, and other platforms or email (SMTP).
Pinpoint Amazon Pinpoint enables Contact centre to engage with customers by providing support for voice reminders, post call surveys, Two way Session context for SMS and any marketing promotions.

Key Benefits of the IBM Cognitive Customer Care Reference Architecture for Amazon Connect

1. Reduce customer effort
a. AI driven conversational interfaces: Using Amazon Lex with Amazon Connect allows contact center operators to create intelligent conversational “chatbots” to turn their automated interactions into natural conversations.
b. Natural Language Understanding (NLU): Amazon Connect with Amazon Lex automated chatbots allow businesses to qualify intent and guide the customer through the exact journey they’re looking for and gives companies the ability to project a unique experience for every caller.
c. ML Based caller authentication: Machine learning-powered, real-time caller authentication that makes voice interactions in contact centers more secure and efficient by leveraging Amazon Connect Voice ID reducing the need for Knowledge-based authentication process. Depending on the authentication score, contact center managers and agents can decide whether they need to take additional security measures during the call.
d. Self-Service: Callers can perform routine tasks, fully automated by Lex chatbots, such as changing a password, requesting a balance on an account, or scheduling an appointment, without needing to speak to an agent.

2. Improve customer satisfaction
a. Customers don’t have to wait on hold for an available agent to get help with their problem leveraging Amazon Lex chatbots to fully automate and answer simple, routine questions means there is less queue contention for customers who need help from live agents to solve complex challenges.
b. Context transferability : If a call does escalate, from an automated Lex chatbot interaction to an agent, the context of the discussion is passed along, which helps reduce the time to resolution. No more having to repeat your name, account number, reason for your call, etc., when your call escalates from the automated system to the agent.
c. Unique Experience to every caller: Using Amazon Connect and Amazon Lex provide customer context such as previous contact history, purchasing history, previous sentiment, etc. allows businesses to create a unique experience for every caller thereby improving CSAT or NPS scores.

3. Gain operational efficiencies
a. Omni channel experience: Provides a seamless experience across voice and chat for customers and agents.
b. Contact Center administrators don’t need specialized skills to get started incorporating Amazon Lex chatbots into their Amazon Connect contact center. Intuitive graphical interfaces make it easy to create highly engaging user experiences and lifelike conversational interactions without writing code.
c. Amazon Contact Lens give businesses the ability to ensure agents are saying the right things at the right times, across all calls, and helps quality analysts focus on the calls where compliance is in question without having to spend time searching for a needle in a haystack.
d. Amazon Contact Lens real time transcripts give managers the ability to know what agent behaviors impact customer satisfaction, across all agents, and can use that information to better train agents; improving agent performance and reducing agent turnover.
e. Amazon Connect Wisdom delivers agents the information they need, reducing the time spent searching for answers across connected repositories to find answers and quickly resolve customer issues. Wisdom uses real-time speech analytics and natural language processing (NLP) from Contact Lens for Amazon Connect to detect customer issues during calls, and then provide agents recommendations and answers which will help in faster issue resolution and improved customer satisfaction.
f. Automate repetitive tasks: Managers can also automate tasks that don’t require agent interaction, like notifying a customer when their claim is processed and a refund has been issued.
g. Integration with 3rd party Systems: Open platform makes it easy to integrate with existing or other third party systems. Amazon Connect provides integrations for leading customer relationship management (CRM) offerings, such as Salesforce and Zendesk, Workforce Management (WFM), and Analytics tools.

4. Customer Insights:
a. Historical metrics reports: Generate reports to analyze how your contact center has performed over a specified period of time. You can generate granular or aggregated reports pivoted on queues, individual agents, and phone numbers.
b. Real-time metrics reports: Gain insight into how your contact center is performing in real time. You can see reports pivoted on queues, agents, and routing profiles
c. Contact search: View detailed individual contact reports with the option to find and play back call recordings
d. Quicksight Dashboards : Comprehensive dashboard defining and monitor the service levels and agent occupancy performance indicators that are most important to you. You can configure the dashboard so that the metrics you care about are always visible.