No Jitter Midroll: Amazon Connect Brings Gen AI-powered Summaries to Agent WorkflowNo Jitter Midroll: Amazon Connect Brings Gen AI-powered Summaries to Agent Workflow

Today, Amazon Connect announced that Amazon Connect Contact Lens now provides generative AI-powered post-contact summaries directly to agents within seconds after a customer contact call complete versus the minutes it previously took to deliver those summaries.

“What we’ve done is taken the Gen AI summary and made it available at the agent layer, what we call the Contact Control Panel [CCP],” said Michael Wallace, AWS’s Solutions Architecture Leader for Customer Experience.

Amazon Connect Contact Lens provides contact center analytics and quality management capabilities for contact center managers and supervisors, so the records it can produce contain all the metrics associated with the call. “The reason it took minutes before had nothing to do with the Gen AI summary [which was] published as part of the entire call record. There's a lot more in [the record] than just the Gen AI summary.”

Wallace noted that Contact Lens already sent the generative AI summary in the contact record to the supervisors. What’s new is that the Gen AI summary is “pushed right to the agents with near immediacy after the call concludes.”

According to Christina McAllister, a senior analyst at Forrester, if an enterprise wants their agents to review and/or edit the Gen AI-produced summary, then it must be available for the agent immediately when the call ends. “A couple seconds of latency is expected for the full transcript to be ready and subsequently processed by the [large language model] LLM, but faster is always better in the contact center.”

As noted, the summary appears in the CCP, the agent workspace, a portion of which is illustrated below.

These summaries can also be accessed natively within Amazon Connect through contact details, as illustrated below. Wallace said this capability is also available via API and Kinesis data streams, enabling integrations with third-party agent workspace or CRM systems.

From the supervisor perspective, the Gen AI summary is useful because those individuals can use it to more easily discover issues, actions and outcomes across the thousands, or even millions, of customer interactions their contact centers receive. Wallace said that Contact Lens provides a rules-based environment so that supervisors can surface interactions based on different metrics such as sentiment scores, keywords, key phrases, etc. “There are lots of opportunities to filter records based on various criteria,” Wallace said. “And I think there will probably be opportunities to use Gen AI to create ‘summary of summary’ reports, but today, it’s rules driven.”

From the agent perspective, by making the Gen AI call summary immediately available, the agents don’t have to manually enter notes. Instead, they can review the Gen AI-created summary, edit if necessary, and then copy it into the final customer contact record.

Wallace noted further that the summaries go with the calls (e.g., when a support call escalates from tier one to tier two) so the next agent can simply read the summary – which means, of course, that customers don’t necessarily have to repeat everything they just said. This ultimately reduces after-call work and may help improve the customer’s overall experience.

“Additionally, by pushing those summaries to the agents, they can avoid that long drawn out note process and get back to doing what they do best – handling customer needs,” Wallace said. “It also helps ensure accuracy since the summary is pulled from the entire transcript.”

Want to know more?