The State of Speech Analytics

Article Featured Image

Research firm MarketsandMarkets late last year valued the global speech analytics market at $1.5 billion and projects it to reach $3.8 billion by 2025, growing at a compound annual rate of 20.2 percent.

Another research firm, Data Bridge Market Research, projects the market to reach $4.2 billion by 2027, growing at a compound annual rate of 17.8 percent.

While the numbers vary slightly, both companies cite businesses’ desires to improve agent performance, combat contact center fraud, extract insights from customer interactions, identify customer emotions during calls, and increase customer satisfaction among the factors that will accelerate the growth of the speech analytics market. “Positive impact created by artificial intelligence will further create new opportunities for the growth of the voice and speech analytics market in the forecast period,” Data Bridge concluded in its report.

Year in Review

In 2020, speech analytics technology continued to improve, particularly as vendors incorporated artificial intelligence into their solutions. Speech technology providers not only updated their solutions, but some added completely new offerings in the past year.

CallMiner, for example, released Illuminate, an AI-driven search feature to help users spot trends in customer interaction data. “Illuminate speeds the time to value in insight for our customers and creates a lower lift in terms of analysts’ time required to use it,” said company president and CEO Paul Bernard in a statement.

CallMiner also released a Continuum Maturity Model, which assesses companies’ maturity levels related to their use of engagement analytics and offers insight on next steps. “It gives customers a road map on how to go forward, where they should be investing, and what the real yields they can get are,” Bernard said.

In October, NICE inContact released CXone Performance Analytics integrated with Salesforce Einstein, an AI-powered analytics application that analyzes blended contact center and CRM data to pinpoint specific recommendations to improve key performance indicators (KPIs). The new optional add-on to NICE inContact CXone is available on Salesforce AppExchange.

CXone with Einstein blends contact center omnichannel reporting data with Salesforce data to create a 360-degree analysis and provide data-driven predictions designed to boost first-contact resolution, agent occupancy efficiency, and sales productivity.

“COVID-19 has presented new challenges to traditional benchmark and KPI tracking tools due to sudden and substantial shifts in customer needs and agent work environments and assignments, including working from home, and increased interaction volumes,” NICE inContact said in a statement. “CXone Performance Analytics for Salesforce Einstein helps leaders quickly understand the new normal and identify new trends and opportunities for improvement beyond what is possible with existing KPI management tools.”

Earlier this year, Verint introduced a Microsoft Teams recording integration, designed to help businesses centrally capture, retain, analyze, and retrieve all communications from Microsoft Teams calling and meeting scenarios, including voice calling, chat, online meetings, screen sharing, and more. Verint’s native integration with Microsoft Teams helps businesses leverage the latest digital collaboration channels while remaining compliant with evolving regulations.

The speech analytics industry also saw a slew of integrations, with a significant number of key acquisitions also taking place in 2020. These include Medallia’s acquisitions of Voci Technologies in April and Stella Connect in September, Astute’s acquisition of iPerceptions in July, Pegasystems’ acquisition of Qurious.io, and RingCentral’s deal with speech analytics startup DeepAffects. RingCentral plans to integrate DeepAffects AI speech analytics into the RingCentral unified communications and contact center platforms. In announcing the acquisition, Kira Makagon, RingCentral’s chief innovation officer, called DeepAffects’ AI models “incredibly rich. They collect details on caller emotions, topics discussed, and talk-to-listen ratios.”

A Look Ahead

In addition to the uses mentioned in the Data Bridge report, companies are also using speech analytics to help generate marketing data, identify causes of caller frustration, track silences, and expand service to other channels.

Forrester Research principal analyst Brandon Purcell recommended in a report earlier last year that companies rely more on speech analytics and less on location-based analytics, which he says has become less relevant as a result of COVID-19 lockdowns and consumers staying away from brick-and-mortar locations.

“Text and speech analytics are more accurate and insightful than ever before thanks to advances in deep learning,” Purcell said at the time. “Leverage these tools to sift through internal customer feedback data as well as external data, such as social media, to decipher emerging issues and respond to changing customer sensitivities.”

Companies already had expanded use of speech analytics in their plans, but the pandemic prompted them to accelerate their timelines, observes Daniel Ziv, vice president of customer analytics and speech analytics at Verint Systems.

“There’s been a big transition,” Ziv says, pointing to both the increase in calls to contact centers and the movement of contact center agents to remote work. “For some companies, calls doubled or tripled.”

Another analytics vendor, Clarabridge, recently found that 51 percent of consumers and a full 92 percent of businesses identified the phone as their preferred channel of customer/business interaction, showing the importance of analyzing these contacts for customer feedback.

“Companies are relying on speech analytics to aggregate the voice of the customer,” adds Chris Tranquill, founder and CEO of Topbox, a conversational analytics platform provider acquired by Khoros in January. Tranquill notes that there was already an uptick in companies using speech analytics for voice-of-the-customer (VoC) purposes in 2019, and it grew even more in 2020 as companies started to look for ways to complement or replace more traditional surveys.

“Companies have to perform better. They need to have access to broader data. There are more and more eyes on the customer experience,” he states. “The traditional survey platform is not necessarily the best indicator of customer experience. There isn’t enough granularity. They don’t provide enough depth and context. Most customers won’t even fill them out. You can’t rely simply on surveys for actionable data. There are thousands of phone conversations.”

By relying on speech analytics rather than or in addition to surveys, companies can more accurately capture customer sentiment not just about contact center experiences but also for product reviews, Tranquill says, adding that chat information should also be included to get the best handle on the voice of the customer.

Beyond VoC-related information, speech analytics, particularly when coupled with artificial intelligence, can help increase the effectiveness of self-service on calls, Ziv says. Speech analytics can alert agents to upsell and cross-sell opportunities while the caller is on the phone, providing excellent sales opportunities that weren’t possible a few years ago when the technology was much slower.

“Speech analytics provides a lot of rich insight on the customer engagement side,” Ziv adds. “It’s also huge for quality monitoring.”

Quality monitoring is important not just to grade and coach agents but increasingly for compliance, particularly in debt collection, financial services, or industries that require disclosures during the course of conversations. Without speech analytics, contact centers couldn’t keep up with the increasing amount of calls requiring compliance-related quality monitoring, Ziv states.

The increased merger, acquisition, and integration activity signals just how much of an emphasis companies—both user companies and technology providers—are placing on the technology, and with good reason.

“The pressure is on speech analytics to get better and better. The need for speech analytics is clear; it won’t slow down,” Ziv says. “It can help people upsell and cross-sell. It’s huge for compliance.” 

Phillip Britt is a freelance writer based in the Chicago area. He can be reached at spenterprises@wowway.

SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues