How Natural Language Understanding Smartens Customer Interactions

Tapping into natural language understanding (NLU) technology as part of an overall approach to managing customer relations improves the experience that businesses provide to their customers.

In a 2020 study conducted by Cisco, 93 percent of participants agreed that technology is very important in creating a better customer experience. Brands already implement a variety of solutions to gain data and insights from customer interactions. Still, many companies struggle to catch trends or potential problems due to disparate data.

Adding NLU to CRM strategies is an ideal way to add a vital piece of technology, as noted in the Cisco survey.

This graphic from Cisco displays other key findings of using technology for a better customer experience.

Asking call center users to stay on the line to answer a satisfaction survey is a tired approach that many customers shun. Inviting pre-and post-purchase customers to share their wide-ranging attitudes and consumer needs is also very low-tech.

Customer experience (CX) and digital customer service firm Clarabridge did away with utilizing surveys in favor of helping brands like USAA, Adobe, and Capital One better understand how to implement NLU to extract patterns and insights from customer interactions. This approach empowers brands to uncover how their customers feel, what they want, and what issues they are experiencing with a particular product or service.

Cisco’s survey revealed that decision-makers are seeking to outsource the contact center to internet-based companies and value AI, automation, and technology’s role in CX. These results indicate that the digital transformation of the contact center is well underway and is proving to be critical to success, according to Julie Miller, vice president of product marketing at Clarabridge.

“As interactions go digital, it is increasingly important to maintain a high level of customer service. For businesses, this means natural language understanding will be critical in monitoring all customer interactions to find opportunities for self-service, actively identify areas for improvement, and use patterns and insights from customer feedback to fuel business decisions,” she told CRM Buyer.

ABCs of NLU

Natural language understanding is a branch of artificial intelligence that uses computer software to understand input in the form of sentences using text or speech. NLU enables human-computer interaction by empowering computers to communicate back to humans in their own languages.

NLU is a subtopic of natural language processing (NLP) that deals with machine reading comprehension. NLP focuses on processing the text in a literal sense, meaning what was said. NLU, on the other hand, focuses on extracting the context and intent to convey what was meant.

“NLU goes a step further than NLP by understanding and extracting meaning from millions of human conversations. NLU analyzes the words, sentences, and context in a conversation to pick up on topics, emotion, effort, empathy, and more,” explained Miller.

She added that NLU empowers businesses to fully understand their customers, including what is driving them to leave reviews and survey comments, make social media posts, and reach out to brands through calls and chats.

“By picking up on patterns and information from millions of customer interactions, NLU provides organizations with insights into what is driving loyalty, growth, and satisfaction. The technology helps businesses get to the root cause of what is going well (or not), identify trends, and get alerted of a problem before a crisis occurs,” Miller said about NLU’s usefulness for better CX.

The Missing Details

That element of customer intent is what makes NLU technology so vital to customer relations. NLU software interprets text and any type of unstructured data. It digests text, translates it into computer language, and produces an output in a language that enables humans to understand the intentions behind what was actually said.

Think of NLU’s significance as a CRM tool in terms of enabling more effective voice technology. NLU is already a key component in voice technology powering Apple’s Siri, Microsoft’s Cortana, Amazon’s Alexa, and Google Assistant. Without the NLU component, these digital assistants would not be able to deduce what users mean, regardless of the way they express it.

To accomplish this goal, NLU applies a range of processes, such as text categorization, content analysis, and sentiment analysis to handle different inputs. NLU technology added to CRM platforms improves the customer support experience business can provide to their customers.

For a deeper grasp of NLU, check out this explainer.

Next Big Thing in CX

Before the pandemic, companies relied on surveys and reviews to gauge consumer attitudes. While this can be helpful, much of this information is outdated when the data is analyzed for any substantive insights into customer wants, needs, satisfaction, or loyalty, Miller observed.

“As companies look to inform and accelerate their digital transformation, NLU will become more widely adopted,” she predicted.

NLU can have a positive impact on the customer experience. When organizations fully utilize the data, insights, and information from customer interactions provided by NLU, they can constantly keep a pulse on what their customers want and need as well as what is working, what is not, and why, according to Miller.

“By keeping track of trends over time and across channels, regions, and products, companies can have the full picture of what is going on with their customers and provide them with a better experience,” said Miller.

Businesses can quickly pinpoint specific areas of friction their customers are experiencing and take immediate action. For example, NLU picks up on actionable words, phrases, and intent. If a customer expresses a desire to drop the company, NLU will surface those concerns to customer service, allowing them to remedy the situation quickly, she explained.

Additionally, by automatically scoring information for emotional intensity, NLU can help businesses prioritize higher-risk customers. It also can help identify what is causing their response. Both results help to retain that customer and improve overall CX.

“NLU allows for a proactive, customer-centric approach to business decisions. With it, organizations can understand what is driving the results they are getting, enabling them to improve their CX constantly and effectively,” Miller emphasized.

Providing Next-Level Services

Understanding the dialog between two or more speakers in an interaction requires more sophisticated NLP and NLU than is needed for analyzing single-speaker feedback channels like online reviews or survey comments. This is where Clarabridge’s approach helps companies replace customer surveys with NLU insights.

Clarabridge’s customer experience management platform gives users access to data and insights like scoring sentiment, emotions, empathy, and effort of both the agents/employees and customers.

The key is providing an analysis of both sides of a conversation. That gives organizations a better understanding of how customers treat employees and vice versa. It also clarifies when agents are confused or confuse customers.

Clarabridge gathers information from a variety of channels, including WhatsApp, Instagram, Facebook, Twitter, chatbots, review sites, and calls. Having real-time access to insights provides brands with the information they need to improve CX without waiting for outdated information from a survey, offered Miller.

Extracting the Essential Data

Brands can pry all kinds of data from patterns and insights dripped from customer interactions. This lets them uncover how their customers feel, what they want, and what issues they are experiencing with the brand’s product or service.

Clarabridge’s NLU engine automatically enriches customer interactions and feedback with information about sentiment, emotion, effort, emotional intensity, intent, and more, according to Miller.

NLU provides companies with a deeper understanding of their customers’ experiences. That litany of information does not reside in data stored in other CRM solutions.

While CRMs provide a central location for all things related to customer experience, many CRM solutions lack sophisticated technology to help unravel what a customer really means when using certain words, phrases, or expressions.

“CRMs that incorporate NLU provide companies with information beyond just basic sentiment or effort metrics,” Miller noted.

CRMs that use NLU help companies not only discover how their customer relationships are going but also what they can do to improve the overall customer experience at a deeper and more actionable level.

Proactive Approach

NLU and machine learning (ML) combine to make a viable replacement for customer surveys and other CRM tools, Miller added. As the number of customer touchpoints and data collection grows, businesses need to leverage cutting-edge text and speech analytics tools to uncover insights and effectively explore data.

“Continuing to leverage CRM tools like surveys leaves companies with outdated information that can lack granularity into customers’ actual feelings and experiences with brands,” she explained.

To be proactive on the ever-changing needs and wants of customers, it is important to find a CRM platform that incorporates NLU and NLP to uncover and flag problems before they arise. That can be intimidating for businesses looking for that solution, she cautioned.

“Rather than take this on their own, companies should research and review vendors who can help show best practices and ways to use the technology to benefit customers truly,” Miller advised.

Bearing the Cost of Better CX

CRM tools that use NLU can be pricey, Miller admitted. But the cost of lagging insights into customer experiences, slow adoption of AI technologies, and attempting to build a text analytics solution in-house can quickly outpace the cost of a CRM tool that already exists, she countered.

The longer businesses wait to act on customer concerns, the higher the cost of wasted opportunities. Waiting too long to act meets other costly problems, such as having to optimize in-house analytics, Miller suggested.

“In essence, the cost barriers can be quickly overshot by delaying action on purchasing a CRM tool with NLU,” she advised.

However, implementing NLU will not solve every issue. Without the proper infrastructure set up, companies could have access to vital information without any way of dissecting or organizing it to their advantage, cautioned Miller.

“For NLU to be a success, it is important that companies have the personnel and technology in place to inform what needs to take place for true CX success based on the data,” Miller recommended.