No Jitter Roll: Generative AI Leads This Week with Updates from Cresta, Omilia, CallMiner and Marchex

Welcome to this week’s No Jitter Roll, our regular roundup of product news in the communication and collaboration spaces. This week: Cresta leverages generative AI to improve business outcomes, provide in-conversation agent assistance and rules-based triggering for contact center events. Omilia uses generative AI to provide in-call suggestions to agents as well as sentiment analysis and other capabilities. CallMiner launched generative AI-based features in several of products to provide contextual advice to agents and more customizability to supervisors. Finally, Marchex has launched generative AI-powered call summary and customer sentiment features.

Cresta Includes New Generative AI Capabilities for the Contact Center

The provider of generative AI for contact centers enhanced three of its products with new large language model (LLM) capabilities:

• Cresta Outcome Insights: A diagnostic tool that identifies and quantifies the connection between behaviors and business outcomes, Outcome Insights includes the ability to measure outcomes at the team and individual level, to identify how multiple behaviors work together, and the ability to infer outcomes when that data is not available/inaccessible.
• Cresta Knowledge Assist: This reference tool uses real-time conversation context to identify information and generate a response that links back to the source. It also includes connectors to third-party solutions (e.g., Salesforce, Sharepoint) and the ability for agents to provide feedback on the quality of the generated responses and content.
• Cresta Opera: Geared toward providing insights to improve agent performance, this product now includes a streamlined rule creation process, the ability to create more nuanced rules that can be validated and more configuration options to tailor agent coaching.

“By applying AI to areas such as knowledge management and analytics, companies can see drastic improvements, including a 20% boost in revenue and 21% jump in customer satisfaction," said Robin Gareiss, CEO & Principal Analyst at Metrigy.

Omilia Launches Agent Assist

The conversational AI provider launched Agent Assist which uses Omilia's proprietary large language model (LLM) to provide agents with an intelligent virtual assistant that advises and guides them based on caller interests and specific issues. Some of the features include real-time sentiment analysis, action suggestion, script adherence, content pushing to the agent from a knowledge base, and real-time call transcription, summary and storage after the call.

CallMiner Enhances its AI-Powered Conversation Intelligence Platform

The provider of conversation intelligence solutions has announced new and updated platform features. These include a new, OpenAI-based capability of searching via natural language which will return results where exact or similar language appears in omnichannel customer interactions. Specific enhancements include:

• CallMiner RealTime provides real-time contextual guidance for agents on multiple chat interactions and different alerts can be assigned based on interaction channels.
• CallMiner Coach has new features that improve ease-of-use, filtering capabilities and search. Supervisors can also now customize the metrics they see.
• CallMiner Analyze includes more reporting options, ability to create multiple dashboards based on role and need, as well as edit, duplicate and share those dashboards.

These new and enhanced features will be available by Q1 2024. In addition, CallMiner now supports audio acquisition from CCaaS providers such as NICE, Genesys, Five9, TalkDesk and Amazon Connect.

Marchex Launches Generative AI-powered Call Summary and Sentiment

Marchex provides a conversation intelligence platform to companies in the automotive, home services, senior living, healthcare and other verticals. It has launched Call Summary and Sentiment Suite capabilities powered by generative AI. "We are currently using a mix of commercial LLM powered by Open AI and Marchex's proprietary conversation intelligence technologies," said Cullen Davidson, VP of product management for Marchex.

Call Summary provides two different types of call summaries: outcome focused (what occurred) and agent-focused (assessments of agent performance). The Sentiment Suite is comprised of three components: Customer Emotion (positive, negative, or neutral), Emotion Categories (assigns specific sentiments such as satisfied, frustrated, or confused), and View of Business, which combines emotion grading and call context to infer how the customer perceives the business (positive, negative, or neutral). These structured data pieces are then paired with natural language explanations of why callers reacted the way they did.

"Marchex has fine-tuned its conversational intelligence technology in specific verticals such as automotive, where we have worked with thousands of car dealerships and several of the nation's largest original equipment manufacturers (OEMs), processing hundreds of millions of consumer-to-business conversations annually into actionable data to deliver insights specific to this vertical," Davidson said. "Our ability to  process a vast amount of data in specific verticals helps us deliver unique and mission critical insights tailored to the sales process for verticals like home services and others where Marchex works with industry leaders and processes millions of conversations every month."

This Week on No Jitter

In case you missed them, here are some of our top stories:

• New Technology and New Metrics for Offices: Configure: Instead of looking at offices in terms of cost per square foot, start looking at metrics that can tell you how people are spending their time.
• Ten Questions to Ask Your Provider About Your Communications Technology Bills: Investing time in thoroughly examining what services you’re paying for can reap big dividends later – let these questions guide your investigation into reducing what your organization pays for telecom services.
• Why Network and IT Management are Failing: Nobody’s happy with how their network and IT management is handled. Application performance monitoring (APM) or “observability” could be the key to everything.