AWS brings third-party apps, no-code chat experiences, AI-powered contact summaries to Amazon Connect

This syndicated post originally appeared at Zeus Kerravala – SiliconANGLE.

Amazon Web Services Inc. Today announced several enhancements to its contact-center-as-a-service product, Amazon Connect, at the Enterprise Connect 24 in Orlando.

Prior to the event, I was given a preview of the announcements in a prebriefing with two senior managers of product management for Amazon Connect — Himanshu Suri and Trevor Bloking. Details of the announcements are as follows:

Third-party app support for agent workspace

The Amazon Connect agent workspace now supports third-party applications. “This allows customers to take either their existing vendor–built solutions, like calendaring applications and analytics apps, or internally built applications like CRM tools or order management systems and iFrame [inline frame] them into the workspace,” Bloking told me. “The feature has been in preview since October of last year. Over that time, we have added capabilities. In December, we began adding event support, which means that these third–party apps can access the events happening as part of that contact.”

Apps from Tech See, Verint, Joulica and Neuraflash will work in Amazon Connect at launch. “We’re also seeing customers bring their own applications as well,” Bloking told me. “That’s a very common pattern.” Bloking showed me a mock-up of an internally built banking application sitting directly in Connect. Apps are integrated at the user interface level, with further integration on the roadmap, according to Bloking.

AWS says the third-party apps are available in regions in the U.S., Africa, Asia-Pacific, Canada and Europe.

Application integration is one of the most highly requested feature updates for contact center solutions. Without it, contact center agents and other users must constantly flip between screens, creating a disjointed user experience. This hurts the agent’s experience, which is then reflected in the service they give, leading to a poor customer experience.

Gen AI contact summaries and automated agent performance submissions

The Amazon Connect Contact Lens function within Connect now provides a post-contact summarization created using generative AI so that managers can monitor contact quality and agent performance. “With generative AI, we are now providing a succinct, coherent summary of what actually happened, including contextual insights,” Suri told me. “This is our effort to simplify quality management for our customers and provide these post–contact summaries immediately after a call.”

According to AWS, the summaries will be short, even if the call is long, with specific examples, such as “customer didn’t receive reimbursement for last minute flight cancellation, and the agent didn’t offer partial reimbursement as per the SOP.” Managers can reduce their time on contact quality evaluations and agent performance.

This gen AI–powered tool, available in the US West and US East regions, is built on Amazon Bedrock for both voice and chat channels.

Without gen AI, summarizing calls is a highly manual process that can include combining customer feedback, agent notes and even supervisor information if they were listening in. It is also time-consuming, error-prone and, in many cases, not very effective. This is a perfect use case for gen AI because it’s fast, accurate and works for every call and chat interaction.

Adding to Amazon Contact Lens features, the offering now gives companies the ability to automate the submission of agent performance evaluations. Managers can now evaluate 100% of customer interactions, using insights derived from conversational analytics to automatically populate and submit evaluations. This also projects aggregate information on agent performance across a range of agents over time.

Self-service, no-code chat experiences

Amazon Connect Chat customers can now leverage no-code self-service experiences with guides to help resolve their customers’ issues. This new feature gives chat customers a way to use a drag-and-drop interface to create step-by-step guides for end customers. The step-by-step guides and chat are supported wherever AWS offers Amazon Connect across all regions.

Creating bespoke flows for specific use cases can significantly improve customer self-service. For example, offering an end-customer options for potential issues they are trying to solve on their own, or if a customer calls an agent about a product return, the agent could push a step-by-step flow delivered via chat.

It’s important to note that contextual information is retained if the customer needs to talk to an agent. AWS provided an example where, if an airline loses a customer’s luggage, they can go through the process, and at the end, if they transfer out of the self-service to talk to the agent, all the contextual information is retained.

Some final thoughts

These latest enhancements take some important steps.

First, they recognize that contact centers are not homogeneous. Multiple systems and apps are at play. Enabling a simpler view is a step in the right direction. Second, they’re showing the practical application of gen AI in helping make post-call actions more efficient. Finally, enabling self-service experiences will improve experiences across the board.

AWS was a latecomer to the CCaaS industry, but with simplicity in mind and through AI, the company has legitimized itself as now being mainstream.

Author: Zeus Kerravala

Zeus Kerravala is the founder and principal analyst with ZK Research. Kerravala provides a mix of tactical advice to help his clients in the current business climate and long term strategic advice.