Speech Analytics: The Fastest Growing App in Contact Centers
Speech Analytics: The Fastest Growing App in Contact Centers
Audio mining finds hidden insights, implicit customer needs and wants, and the root causes of issues embedded in conversations.
By Donna Fluss
Speech analytics was first introduced into the contact center market in 2004 and is still considered an emerging application. Within 5 to 8 years it will become a “must have,” as it is the only solution that gives enterprises insight into customer needs and wants in near-real time.
Contact center speech analytics solutions are used to literally “listen to” customer conversations to identify the reasons why customers call. The conversations are analyzed to provide input about all aspects of a company’s performance: sales, marketing, product development, operations, fraud, collections — whatever is on customers’ minds. Within two years of its introduction, speech analytics has developed into the fastest growing contact center application.
What is Speech Analytics?
Speech analytics, also known as audio mining, finds hidden insights, implicit customer needs and wants, and the root causes of issues embedded in conversations. It can monitor how well agents comply with scripts and/or regulations.
Speech analytics applications capture customer conversations and structure them, transforming the content into metadata that can be searched. The structured conversations are then analyzed using a variety of techniques, including key word, phrase, concept and contextual search. Some speech analytics applications can detect concepts and trends that end users didn’t even know existed. This enables contact center managers and executives throughout the enterprise to address the issues that generate call volume and identify competitive challenges and new revenue opportunities. See Figure 1.
Speech analytics applications are compelling because they deliver benefits to an enterprise’s bottom line; these applications reduce costs and increase revenue while improving the customer experience. The benefits of speech analytics are far-reaching, extending throughout the enterprise and beyond, to customers.
Benefits of Speech Analytics
For contact center managers, speech analytics uncovers valuable customer insights and identifies the root cause of calls on a timely basis. This enables managers to reduce call volume, improve quality and reduce customer attrition. It’s also a great tool for making sure agents meet regulatory requirements and adhere to scripts.
For sales organizations, speech analytics can be used to identify ways to increase sales conversion rates — both for new customers and for up-sell/cross-sell opportunities. Speech analytics can also be used to segment customers and to identify the most appropriate channel for selling to them. Customers often tell contact center agents their preferred form of communication and what types of products and services they favor.
A speech analytics application can capture this information and use it to better satisfy customers by interacting with them in their channel of choice, while also increasing sales conversion rates. Additionally, customers will be pleased that an enterprise listened to them, which will help to build customer loyalty.
Speech analytics applications contribute to marketing organizations by identifying competitive challenges and opportunities on a timely basis, while it’s still early enough for the enterprise to create a campaign to respond appropriately. Speech analytics can be used to identify “at-risk” customers and successful retention methods. It can detect a marketing campaign or product collateral that is causing negative reactions or confusing customers.
Speech analytics can be used to identify new product ideas and enhancements and to gain an appreciation of general customer insights about a company, its products and the competitive environment. It is also a great tool for understanding customer and market trends. Speech analytics is also very important from a public relations perspective. While it may be rare, companies do occasionally make a misstep and generate negative publicity. Speech analytics functions as an early warning system and can notify management of an issue before it impacts a large segment of the customer base or becomes an issue in the press.
Contact centers handle all customer inquiries, regardless of which department in the company causes the underlying issue that generates the calls. Speech analytics rapidly identifies call trends and their root causes. It identifies performance issues throughout the enterprise and is the only tool that can directly and consistently attribute calls to the group that triggers them. Speech analytics provides timely awareness of operational issues. It monitors the performance of all operational areas and evaluates customer satisfaction with policies and procedures.
Speech analytics also provides benefits for many functional areas not reflected in Figure 2. For fraud departments, it can quickly pick up on new schemes and prevent major losses. It can also be used to identify individuals who are engaged in fraudulent activity. For collections groups, speech analytics emotion detection can be used to measure stress levels to determine if customers are telling the truth when they make a promise to pay. By identifying customers at risk of defaulting on a loan or payment early on, an enterprise can intervene promptly and reduce their losses.
Among the most important beneficiaries of speech analytics are enterprise customers. Today, much of what customers share with enterprises is ignored because agents do not have a system or process for formally gathering and communicating this information. Wrap-up systems are limited and generally invite agent feedback on only one of many pre-specified items. CRM applications capture transactions, but generally do not reflect customer opinions. Speech analytics does not miss anything and its input is most valuable if the enterprise builds a formal process for putting its output to work in a timely manner.
It’s one thing to hear about the value of speech analytics and another to see examples where it was applied to benefit an enterprise and its customers. DMG Consulting has tracked numerous speech analytics implementations over the past two years. We are aware of many success stories, and also of some troubled implementations that either didn’t yield the expected results or took too long. The good news is that there are substantially more successes than failures and the proportion of failures has decreased substantially during the last year as the offerings have improved and more best practices have emerged.
Case Study: Auto Insurance
Problem: An auto insurance company decided to improve the performance of its contact center by identifying the real reasons why customers called (root cause analysis), improving agent training and reducing risk and legal exposure.
Benefits: The speech analytics application met all of the company’s needs. Within the first three months it identified 11 fraudulent callers, representing a cost avoidance of $500,000. It gave management the information they needed to improve procedures and train their staff. This resulted in a 2% increase in directing accident victims to bring their cars into a company body shop, a savings of $180,000/month or $2.16 million annualized. It also resulted in a 40% — 50% reduction in escalation calls, happier customers and improved accuracy of agent catastrophe coding for the home office.
Secondary uses and benefits: The company also used the information from the speech analytics application to pinpoint agent training needs and increase the effectiveness of its cross-selling programs.
In over 23 years, I have never seen an application with greater potential to benefit enterprises. However, end users have to distinguish between vendor hype and reality. Speech analytics is capable of providing useful and actionable data. If used properly and on a timely basis, this information can improve an enterprise’s bottom line by reducing operating expenses and increasing revenue, while improving the customer experience.
Speech analytics solutions have improved substantially in a short period of time due to the millions of dollars invested in enhancing their accuracy and ease of use. There is still opportunity to improve the underlying technology, but the most significant challenge in a speech analytics initiative is organizational. Enterprises that implement speech analytics must be prepared to change processes and retrain their staff to take advantage of the information that the application discovers. They must also build formal processes that allow the contact center to share data with other operating areas on a timely basis.
Any company that wants a strategic differentiator should take a close look at speech analytics. Speech analytics gives all enterprises, no matter how large or complex, a way to listen and, ultimately, respond to their customers on an institutional basis.