Online sentiment analysis can improve CX

If you have a million customers, do you know how they feel? Would such insight help you improve your service to reach a million more? Can you tell the difference between the things you think they need and those they do desire? Online sentiment analysis, also known as opinion mining, may help you find out the answers to all these questions.

What is online sentiment analysis?

Online sentiment analysis differs from traditional surveys in that it works with already available data, such as internal operations data, social media feeds, online comments and news reports. It then uses artificial intelligence (AI) and natural language processing (NLP) to yield insights into people’s emotions.

The big advantage of sentiment analysis is that it can deliver insights more quickly than traditional surveys and user groups. While a conventional customer survey may take weeks, sentiment analysis can yield results in hours.

The tools use a combination of different techniques to place messages on a spectrum of sentiment, identifying emotions and intentions. Most available packages rely on lexicons, which consist of words chosen to reflect positive, negative and neutral reactions. They use advanced algorithms and deep learning techniques to classify text, exploring intent and context. This contextual understanding approach means that tools can identify conversations that are about price even if the specific word isn’t used.

However, it is important to constantly curate machine-learning models to ensure systems do not deliver flawed results that perpetuate prejudice or encourage cultural bias. Executives recognize this problem: 59% of EMEA brands have observed AI-driven bias. Tools should remove bad elements of customer experience (CX), rather than create them. Like so much in digital transformation, they should be used to augment objective and empathic human input.

Does sentiment analysis work?

It seems effective. Researchers at the National College of Ireland School of Computing ran a Syuzhet sentiment analysis on tweets to identify which of the 15 movies would succeed. The results yielded 60% predictive accuracy. Investors and traders use these tools alongside machine learning and predictive tools to understand emotional sentiment better to predict stock prices. CNN’s Fear & Greed Index suggests how such insights work.

Investors have found that Twitter sentiments are subsequently reflected in stock market activity. Researchers have also found that a nation’s Spotify music choices reveal national mood and correlate with stock market activity.

Sentiment analysis can deliver valuable insights across national and local politics. It can also generate great insights into customer relationships, reactions to new products and services, and your customer journey.

There’s a strong correlation between CX and growth. Gartner claims 80% of growth organizations use CX data, compared to just 58% of non-growth companies. Customers like companies that help build stronger CX. In fact, 77% of consumers have a more favorable view of brands that ask for and accept customer feedback. The 2022 Adobe Trust report shows 71% of customers think it’s important for brands to demonstrate empathy by showing they can see things from their perspective.

The bottom line? Using sentiment analysis can help you test the impact of your CX in real-time – thereby helping you improve customer loyalty and satisfaction. Happy customers generate word-of-mouth marketing, positive reviews and recommendations to help your company achieve business success.

Stopping problems in their tracks

Getting things right when they go wrong is one of the most critical moments in the customer journey. Sentiment analysis can help ensure better customer experiences at such times. These tools are powerful enough to analyze the text of a support request sent via email or chat to identify a customer’s current emotional state. The agent can then be provided with the appropriate empathic response to a customer in that state, accelerating customer success.

One Spanish banking group receives over 80,000 customer comments each month. It analyzed a million of these to identify the most frequent complaints. This helped it identify which improvements to prioritize, and the work generated a 10-point increase in customer satisfaction. Sentiment analysis can help quickly identify high-priority issues in existing CX and deliver the insights you need to resolve them.

McKinsey explains that 85% of customers will purchase more following a positive customer experience, but 70% purchase less after a bad one. In another application, brands use automated sentiment analysis tools to monitor social media for criticism, alerting social media teams as it appears. They can then swiftly reach out to the unhappy customer, resolving their complaint on the same social media feed.

These tools also help separate truth from the noise. Nike suffered from loud criticism when it featured NFL quarterback and civil rights activist Colin Kaepernick in a marketing campaign. Despite the well-publicized clamor, sentiment analysis revealed that positive sentiment increased dramatically, and sales grew 31%. To get a sense of what these tools may do for your business, MonkeyLearn, MeaningCloud, and IBM Watson offer free trials, though many other services also exist.

Playing a key role in CX

Sentiment analysis is one of many tools and technologies available to companies seeking to improve CX. In use, sentiment analysis helps meet a need to replace slow-moving traditional customer surveys with faster tools. Surveys only reach limited customer subsets, are reactive rather than real time and can be unfocused and provide ambiguous results. Just 15% of business leaders are fully satisfied with them.

Sentiment analysis digs deeper into larger and more diverse customer groups, answers questions in near real time and can reveal insights that may take months to find with survey data. It can also be used alongside predictive CX data analytics models to inform new product development and future trends.

AI presents the biggest opportunity for companies to evolve their CX. In the best contact centers, bots enable fast service when it’s needed and enable agents to give detailed answers to their customers, de-escalate emotional engagement and offer extra care when the situation demands it. Read this ebook to discover chatbots and RPA.

Jon Evans

Jon Evans is a highly experienced technology journalist and editor. He has been writing for a living since 1994. These days you might read his daily regular Computerworld AppleHolic and opinion columns. Jon is also technology editor for men's interest magazine, Calibre Quarterly, and news editor for MacFormat magazine, which is the biggest UK Mac title. He's really interested in the impact of technology on the creative spark at the heart of the human experience. In 2010 he won an American Society of Business Publication Editors (Azbee) Award for his work at Computerworld.