Interaction Analytics Helps Companies Hear their Customers
View this article on the publisher’s website.
INTERACTION analytics (IA) is a must-have solution for enterprises that want to understand and enhance their customer experience. These solutions convert unstructured recorded and live-stream audio and digital customer conversations into transcripts and structured output files that can be categorized, searched, and analyzed. Interaction analytics solutions can extract business intelligence from omnichannel interactions captured anywhere in an enterprise, covering both speech and text.
Organizations can use the findings to gain insights into customer and employee needs, wants, trends, and opportunities. Interaction analytics enables companies to measure and quantify the impact of their actions on their customers and prospects. Just as important, these solutions discern nuances in customer/employee communications, including intent, emotion, sentiment, empathy, propensity, and more.
The newer real-time IA applications analyze interactions as they are taking place and use the findings to generate actionable alerts or recommendations to managers, supervisors, and/or agents, to bring their attention to situations that require immediate remediation. These applications have great potential to dramatically improve the performance of contact center agents and other front-line employees (see figure 1).
Here’s a list of standard capabilities that should be included in a full-featured IA solution:
1. Omnichannel recording to capture audio, screens, and text-based interactions and/or integrate with a variety of call recording platforms.
2. Indexing to process the recorded or real-time conversations by breaking them into phonemes (representations of sounds) or words, phrases, and concepts.
3. Search, retrieval, and playback to enable the identification, retrieval, and replay of relevant call/text segments and/or the entire call or conversation.
4. A query builder to mine, categorize, and analyze the content of live and recorded interactions (structured queries are employed to parse the data into words, topics, or categories for root cause, trend, and correlation analyses).
5. Categorization to group interactions automatically by type, based on a contextual analysis of the language, words, phrases, events, rules, etc., to identify underlying trends.
6. Concept extraction to automatically find similar themes expressed in different ways.
7. Speaker separation to distinguish between different speakers (customer and agent) in dual-channel recordings, in order to conduct analysis on the speaker-side audio.
8. Conversation analysis to detect and analyze interruptions, over-talk, cross-talk, and silence.
9. Transcription that applies a speech-to-text engine to simultaneously convert all audio interactions into text, providing a visual map of the call.
10. Emotion detection that uses language patterning and acoustic metrics to identify a wide range of emotions in customer interactions.
11. Sentiment analysis to automatically detect, extract, and classify whether the conversation is positive, negative, or neutral; it should include the ability to differentiate sentiment by agent or customer.
12. Discovery to identify and surface concepts and trends that organizations would not have known to look for and to correlate seemingly disparate but related issues.
13. Trend analysis to identify changes in topic, phrase frequency, and trend direction over time.
14. Root cause analysis to provide visual word clusters, correlation, and context maps of related items, in order to suggest a root cause.
15. Redaction to remove sensitive data from a conversation to comply with regulations or to address security concerns.
16. An analytics-enabled quality management module to identify, classify, and score voice- and text-based interactions based on defined quality criteria.
17. Tuning/training to refine and improve system findings.
18. Reporting and dashboards to deliver and display system findings.
19. Real-time and/or historical alerts or pop-ups to communicate an issue to an agent, supervisor, or manager on a timely basis.
THE BENEFITS OF IA
Companies need a constant flow of insights into customer needs and wants, as well as information about how well employees are performing. The ideal scenario is to receive real-time inputs that can be used to alter outcomes during customer interactions, along with historical data to analyze results and identify patterns and opportunities to improve the company’s performance—its products, services, employees, partners, or systems. Interaction analytics is the only application that can provide this range of information. Sophisticated IA solutions deliver both real-time and historical insights that, when properly analyzed and applied, deliver the enterprise-level findings that corporations expect from business intelligence solutions.
A unique and highly beneficial aspect of IA is its ability to address voice and digital channels and stitch together a comprehensive story of the customer experience. Looking at feedback in each channel has always been important, but gaining visibility into what is happening across channels and business units is essential to understanding the overall customer journey. This is becoming even more critical as activity in digital channels picks up momentum.
Companies that want a complete view of why customers/prospects are reaching out to them should use both speech and text analytics to capture and analyze insights from all interactions. When interaction analytics are performed on a historical basis, conversations are typically analyzed anywhere from a few minutes to 24 hours after they are received, although the timelier the analysis, the more actionable and generally beneficial the findings.
Historical interaction analytics can be put to the following uses:
• Identifying the root causes/underlying reasons why people are contacting the organization.
• Automating the quality management process.
• Ensuring that agents are properly adhering to their scripts and complying with external and internal regulations and requirements.
• Identifying and measuring the level of customer and agent emotion in each interaction.
• Measuring how customer sentiment changes during an interaction.
• Obtaining a complete and accurate transcript of all interactions.
• Automating the wrap-up process by providing summaries of interactions, including what customers requested and what agents committed to do for them.
Historical IA can deliver useful results, but providing real-time feedback to agents and managers while the customer is still on the line is where IA solutions really prove their value. Real-time IA, which captures and processes customer conversations as they occur, uses the derived insights to notify supervisors of issues and opportunities in the form of real-time alerts. These capabilities can identify angry, frustrated, or unhappy callers and allow supervisors to provide advice in the moment and/or join the conversation. To assist agents, real-time IA solutions deliver context-based real-time guidance with information from a knowledge base or instructions on how to perform a task. This is a huge benefit for remote or work-from-home agents who lack in-person supervisory resources, and it supports new agents through the onboarding process.
Real-time interaction analytics can be put to these uses:
• Providing agents with guidance so they can accurately resolve an inquiry during the first contact.
• Identifying and immediately addressing non-compliance or potential fraud situations.
• Providing next-best-action recommendations to agents.
• Providing real-time guidance to agents to help them close a sale or collect funds.
• Identifying customer sentiment and emotion and alerting agents and supervisors when a call is going badly.
Historical speech analytics is an essential application in a growing number of contact centers, and real-time IA (also known as real-time guidance, or RTG) is a new market entrant that is gaining traction. Because it can track various aspects of the customer journey, IA is expanding beyond the contact center and is slowly being applied to other enterprise functions, which will expand its benefits. IA is also being embedded into a variety of third-party applications, including sales, customer relationship management, voice-of-the-customer, and call tracking solutions, which is greatly increasing its value for enterprises.