Marchex Releases Conversational AI to Predict Customer Intent
Marchex, a provider of call analytics, today released artificial intelligence technology that can interpret customer intent during sales conversations over the phone or via text. The new predictive AI models feature more than 230 signals.
"Marchex is cracking the code of consumer intent," said William Li, vice president of engineering at Marchex, in a statement. "How a business engages or responds at each moment and touch point in the customer journey can make or break a sale, as well as impact the customer experience. That's why our AI is so powerful and why we believe it goes well beyond today's basic AI offerings on the market."
Marchex Conversational AI is built on Marchex Stream, the conversational data streaming and business intelligence platform. The technology is powered by Marchex's growing base of more than 1 billion minutes of analyzed consumer-to-business conversations and employs deep learning techniques, including convolutional neural network developments pioneered by the Marchex MIND Lab team.
Some of the new features and capabilities in Marchex Conversational AI include the following:
- Deep Consumer Intent Signals--Eighty new deep intent signals identify and organize information that consumers express in phone and text conversations These new signals identify intent, context, urgency, product and service interest, appointment scheduling, and payment preference, all of which help businesses determine when a consumer is evaluating service or vendor options and which issues or products are of greatest importance and then predict when they have decided to purchase.
- Lead Scoring.
- Churn Prediction.
- Deep Outcome Insights--One hundred new conversation outcome signals organize outcomes of sales or service conversations and help business respond to recapture missed sales opportunities. These signals can identify the reason for the outcome with granular detail, alert the business immediately, and inform the recommended next action for the business to take.
- Sales Best Practice and Performance Improvement Insights--A new voice biometric model determines the identity of the salesperson on the call, and 50 new performance-tracking signals analyze the salesperson's responses to the consumer to identify whether he responded to the consumer's actual intent and how effectively he followed sales or service scripts.
- Vertical Industry Models--Pre-trained models can help businesses address sales conversation challenges and move customers forward in the path to purchase. Today Marchex has pre-trained models for the automotive sales and service, communications, insurance, dental/healthcare, and home services industries, as well as a general-purpose model that can be used across other industries.
"Moving beyond the industry hype, the real measure of success for AI is how effectively it solves real world business challenges," said Russell Horowitz, Marchex's executive chairman, in a statement. "Marchex Conversational AI and Marchex Stream represent important technical advancements that will continue to drive future innovation for Marchex and will be difference-makers for our customers."