AWS Updates Amazon Connect

Today at its AWS re:Invent virtual conference, Amazon Web Services announced five capabilities for its contact center service, Amazon Connect, to improve contact center agent productivity and end-user customer experiences.

The capabilities announced today give agents the right information at the right time, provide more personalized service, help managers impact customer interactions during calls, make it faster to authenticate customers, and make customer follow-up tasks easier to manage. These capabilities are powered by AWS's machine learning technology.

The five Amazon Connect capabilities launching today are the following:

  • Amazon Connect Wisdom, which gives agents the information they need to solve issues in real time. Wisdom ingests and organizes knowledge content from both homegrown databases and third-party knowledge repositories, with pre-built connectors to Salesforce and ServiceNow. It uses natural language processing (NLP) to detect customer issues during the call and subsequently recommends relevant content stored in the knowledge repositories.
  • Amazon Connect Customer Profiles, which ensures agents have a more unified profile of each customer so they can provide more personalized service. When a customer calls, Customer Profiles scans and matches the customer records across multiple applications for unique identifiers like phone numbers or account IDs. Customer Profiles combines contact history information from Amazon Connect (such as the number of holds, transcripts, customer sentiment, etc.) with customer information from CRM, ecommerce, and order management applications into a unified customer profile that is displayed in the Amazon Connect agent application at the moment a call or chat starts. AWS customers can connect to other homegrown applications using Amazon Connect's software development kit and application programming interfaces ). AWS customers can also use pre-built connectors to third-party applications like Marketo, Salesforce, ServiceNow, and Zendesk, directly from the Amazon Connect console.
  • Real-Time Contact Lens for Amazon Connect, to impact customer interactions during a call. Contact Lens for Amazon Connect (announced last year at re:Invent) helps customers analyze contacts by providing a call transcript that shows what the agent and caller said, performing sentiment analysis with a positive or negative score, and detecting key words, phrases, or other analysis criteria, like silence or people talking over one another. Contact center managers can create rules to flag customer issues using keywords or sentiment analysis. Managers receive real-time alerts when their specified conditions are met so they can provide guidance or have the agent transfer the call. When a call is transferred, the agent can pass the real-time transcript with conversational details to the next agent or manager.
  • Amazon Connect Tasks, which automates, tracks, and manages tasks for contact center agents. Tasks provides pre-built connectors to CRM applications and APIs to integrate with homegrown applications. With Tasks, managers can assign and prioritize tasks to agents based on agent availability and skill set, and the tasks are displayed to agents in the same Amazon Connect interface that they use to view their call and chat interactions. When assigned tasks, agents see notifications along with descriptions of the tasks and links to applications that the agents might need to complete the tasks. Agents can also create tasks on their own to track their own follow-up work. Managers can also create workflows to automate tasks that don't require agent interaction.
  • Amazon Connect Voice ID, which provides real-time caller authentication using machine learning-powered voice analysis. With Voice ID, callers can authenticate themselves using their voices. When a caller opts in, Voice ID analyzes his speech attributes (e.g. rhythm, pitch, or tone) during the first few seconds of the call and then creates a digital voiceprint. When the caller calls in again, Voice ID compares the voiceprint to the claimed identity and authenticates him based on a Voice ID confidence score. This all happens in the first few seconds of a call, when an agent gets on the call and asks the caller to state their identity and reason for calling. If the caller doesn't meet the confidence score threshold, the agent can verify the identity through additional screening or transfer the call to a fraud specialist.

"Amazon Connect is one of the fastest-growing services in the history of AWS and has clearly struck a chord with customers," said Pasquale DeMaio, general manager of Amazon Connect at Amazon Web Services, in a statement. "Amazon Connect is the most flexible, scalable, cost-effective, and inventive call center service in part because it was built from the ground up on the cloud with machine learning deeply embedded to automatically manage features like interactive voice response (IVR), chatbots, call analytics, and sentiment analysis. Today's five new Connect features build on this foundation to make it even easier for customer service agents to have the information they need to provide faster and more holistic customer experiences, optimize agents' time based on what matters most, and enable customer service managers to take action in real time to avoid contacts that will do harm to their brands."