When I tell people that I work on text analytics products, their first question is often “But, how do you handle sarcasm?” My typical response—“About as well as a human, which is to say not very well at all”— is equally sincere and sarcastic, a poetic homage to the difficulty of the linguistic problem.
Sarcasm is mired in a complex web of deep cultural knowledge, emotional sensitivity and individual awareness. Some cultures (looking at you, my British friends) find comfort in the dryness of the humor, whereas others find it distasteful or even disrespectful. Sarcasm can be subtle or obtuse; flattering or insulting; funny or offensive. In sarcastic expressions, the words used are only a small piece of the puzzle. Tone matters, body language matters, context matters. Given the high degree of tangible and intangible awareness necessary for sarcasm to succeed, it is perhaps unsurprising that non-native speakers struggle to master both delivery and understanding of sarcasm in learned languages.
All of these components make sarcasm an extremely difficult linguistic problem to study. Linguists have classified sarcasm into specific sub-types including irony, satire, passive aggression and flattery. They’ve determined that your brain works differently when processing sarcastic comments in comparison to sincere ones. Others have identified the facial tics that betray an otherwise earnest face. Most of this research, though, is conducted through individual face-to-face analysis. Conducting broad analyses of sarcasm through text analytics is nearly impossible. Text-only communication lacks tonal and visual cues, making it highly susceptible to misunderstanding and misinterpretation. Other clues for correct interpretation of a comment may be baked directly into the medium and shared among some of its participants, but these clues may be imperceptible or opaque to outsiders.
An NLP engine is not a native speaker of any human language. It understands the rules or the patterns that its human overlords have programmed into it, but it lacks any linguistic intuition. It can understand words used, relationships between those words, and maybe even emotion and intent, but it fails when meaning transcends content. It may understand its context or its purpose but will undoubtedly fail when other niche cultural or societal knowledge is injected into a witty retort. As humans, we could all listen to or read the same passage and walk away with different understandings of its intent. An NLP engine fares about the same. In some situations, it will interpret an ironic comment correctly; in other cases, it will completely miss the mark and produce the exact opposite sentiment as a native speaker would otherwise expect.
The Clarabridge NLP engine errs on the side of sincerity, but, given the flexibility in our sentiment engine, users have the power to customize rules to support common sarcastic phrases that appear in their dataset. I’ve found success in customizing rules for two specific types of sarcasm.
1. Speakers, in an attempt to underscore their emotions, may associate positive actions with negative aspects (or vice versa) such as in the sentences “I love sitting in traffic” or “going to the dentist is the best.” Clarabridge understands word associations, parts of speech and sentiment and allows users to leverage this word-level metadata in the construction of sentiment rules. Users could construct a rule that negated every positive verb associated with “dentist” or “traffic” or “[insert emotionally charged word from your industry here].”
2. Social media posts are now often suffixed with #sarcasm, #sarcastic or #not to aid a reader in interpretation of an otherwise ambiguous post. Users can tune sentiment based off of specific hashtags and positions of these hashtags within posts.
Sarcasm is, without a doubt, an important part of our communicative tools as social beings. However, sarcastic expressions are relatively rare in most types of text. The ability to detect sarcasm through computational means should not be a make-or-break point when deciding which NLP tool to use. We at Clarabridge will continue to investigate sarcasm in customer feedback and how we may be able to improve sentiment accuracy for these expressions, but for now I’ll quote the Comic Book Guy from “The Simpsons”: “Sarcasm detector? Now that’s a really useful invention.”
For this week’s topic of spam, I want to remind you of this Monty Python sketch from 1970. It starts with customers trying to order a breakfast in a diner. The waitress details what’s on the menu:
“Well, there’s egg and bacon,
Egg, sausage and bacon,
Egg and spam,
Egg, bacon and spam,
Egg, bacon, sausage and spam,
Spam, bacon, sausage and spam,
Spam, egg, spam, spam, bacon and spam,
Spam, sausage, spam, spam, spam, bacon, spam, tomato and spam,
Spam, spam, spam, egg and spam,
Spam, spam, spam, spam, spam, spam, baked beans, spam, spam, spam and spam.”
Ah, Monty Python at its finest. And, recently, an apt description of my email inbox and my voicemail. Spam, Hawaii’s favorite processed meat and Western culture’s favorite protein to mock, has an impressive résumé. Hormel offers 15 flavors of Spam and has sold over 8 billion cans across 44 countries since its introduction in 1937. But, Spam’s influence isn’t just culinary; it has also had an understated influence on our technological world. Inspired by the Monty Python sketch, certain abusive users of Bulletin Board Systems and Multi User Dungeons would repeat “spam” a massive number of times to scroll other previous messages off the screen. Soon, the term became a moniker for the unwanted junk we find on the Internet that obfuscates the content that we actually want to see.
Spam content can be a real burden to those trying to find meaning from their customer feedback data. It can increase the effort needed for analysis and can cause misinterpretation of customer needs. In most cases, customer experience analysts want to analyze the customer’s organic voice, not the inorganic voice of automated bots or fake customers. Looking at volumes and sentiment of mentions of specific products or topics without regard to whether the content is spam can be very misleading.
Not all spam, though, is created equal. By classifying the types of inorganic messages that appear frequently in customer feedback data, we can gain a better picture of how the market views a brand or product. The types of spam present in each dataset may vary. On social media, auto-generated messages in content such as news headlines, reviews and articles abound. In email sources, job requests or solicitations for corporate sponsorship may get in the way. Analyzing the language used in the inorganic messages bares its own utility. When these spam documents mention a brand or product, they may reveal how customers and potential customers use, advertise and perceive that brand or product.
A tool that blindly looks at words and phrases is limiting itself to parsing of discrete words; a tool focused on understanding will tease out the organic versus inorganic messages and classify them into their associated types. By exposing these types to end users, such a tool empowers analysts to isolate the true feedback and determine meaningful insights. The Clarabridge proprietary Content Type Detection feature uses a machine learning algorithm to identify and tag spam content and then classify it into subtypes: advertisement, coupon, link or undefined. Users can choose either to purge any spam documents upon ingesting or to retain them for analysis. Analysts can also customize this feature by training the algorithm on their own data and with custom subtypes. With the Clarabridge Content Type Detection, users can go beyond basic topic and sentiment analysis. Considering the integrity of a message provides a different dimension and a unique analytical lens for many stakeholders that would be missed or misinterpreted if all documents were viewed the same way or were viewed purely by their independent words. Just the same as how “egg and spam” is not the same as “spam, spam, spam, egg and spam.” Bon appetit.
It’s important to provide a stellar experience every time someone wants to engage with your brand. And as the avenues to engage with customers continue to expand, you will want to be accessible and ready to serve on these new channels. Research indicates that customers who start and end service requests using digital channels have a satisfaction rate that is significantly higher than those using traditional channels. Additionally, the costs per digital customer interaction can be between 5 to 12 times cheaper than when they engage you via the phone. It is clear that improving your digital efficiency can be an opportunity to minimize costs for your contact center and improve the overall experience for your customers.
Here are three ways that today’s organizations can boost their digital effectiveness:
1.Listen to every feedback source to prioritize opportunities that will have the highest impact on the customer experience As the old adage goes—you cannot fix what you cannot measure. Similarly, you cannot enhance what you don’t fully understand. When embarking on the journey to enhance digital effectiveness, it is essential to listen to all sources of feedback in order to prioritize opportunities that will have the highest impact on the customer experience. Calls are very important but they’re not the only way your customers are reaching out or looking for support. Look at emails, social media posts, review sites, surveys, and chats and integrate those sources with CRM data to reveal the full picture.
Clarabridge clients will often analyze unstructured text feedback from these channels by overlaying the Clarabridge sentiment and effort scores to understand high friction points in the digital experience. They will also look for language around “suggestions” when your customers are telling you what they wish could be better. When analyzing phone calls, our clients typically aggregate all mentions of failures on online channels to understand pain points. They also analyze short duration calls that typically have a singular call driver and prioritize these for digitization.
Using the techniques above, action-oriented organizations can create a priority matrix that ranks each digital opportunity along the two dimensions of customer impact and level of effort. An opportunity that is deemed to have a high customer impact but will only require a low to medium level of effort to implement quickly jumps up the prioritization list.
2.Understand the full customer journey and measure the impact of channel switching Today’s customer wants to communicate via their channel of choice, and increasingly, these channels are digital in nature. When your customers start a transaction online or via a mobile app, they want to be able to complete their transaction within that channel. However, many times they are unable to do so and are forced to contact you via a phone or chat channel for further assistance. We often hear comments like “I don’t know why we get so many calls related to buying a ticket when our customers can easily do so via our mobile app or our website.” Upon deeper analysis it becomes clear that these transactions are often more complicated and multi-faceted than the simple act of purchasing a ticket. There is often another related event that complicates the transaction and causes a channel switch. In the example above, it may be that the customer is indeed trying to buy a ticket but is trying to do so via loyalty miles or perhaps they are trying to apply a discount code they were provided that is not working. Understanding the co-occurrence of such related events to the main transaction is key to designing a digital solution that solves for not just the most basic intents but the more complicated scenarios that may result in channel switching.
3.Explore emerging AI-powered technologies such as chatbots Chatbots provide a way for customers to self-serve on known issues, or to collect important information that facilitates a seamless transition to a contact center agent. This type of innovation is experiencing significant growth and is a growing topic of conversation with clients at Clarabridge. As suggested in a 2016 report by Creative Virtual, introducing a virtual assistant for customer service can improve chat and phone service levels by 10-15%.
You can train your chatbots to improve your customer experience in a variety of ways. By understanding customer “intent” during a live chat interaction or phone call, you can start identifying opportunities for chatbot automation. You can also listen for the words that customers use to express frustration and high effort while accomplishing a task. Once you understand these linguistic patterns, you can train chatbots to express empathy, and to route a frustrated customer to an agent with a skill set that specializes in the particular topic that is causing frustration.
The Clarabridge Marketing Team recently sat down with Dimitri Callens, Director of Product Management at Clarabridge, and asked him to answer 5 questions on the state of the social media management industry. Here’s his take:
1.What do you believe are the key ingredients for a brand to engage successfully with their customers on social media?
First and foremost, you have to listen to what is being said about you out there. There are plenty of tools that will help you understand the trends, your audience, and what they like and dislike about what you offer. Then adapt your messaging to your audiences. Don’t try to push your product—today’s generations see right through that. Be honest and post content that can spark their interest about what your product can do for them.
And if you are offering customer service on social, make sure it is in the DNA of your company to see customer service as an investment, not as a cost. If everyone in your company believes helping customers is a positive thing, instead of a burden, it will shine through in how you appear online—for anyone to see!
2. As an industry insider, what are you most excited about concerning innovation in the field?
This is not a technical innovation but one that is happening in the minds: more and more executives understand that social customer service should not be seen as a cost, but as an investment.
I recently had this discussion with a CEO of a rapidly growing online retailer that specializes in car luggage racks. “It literally costs me money having my people answer customer questions, and it does not gain me a penny as the product is already sold,” he said. Of course, from an accountant’s point of view, he was right. But even from a broader view, he told me it was a cost. I did not feel like angering him more at that point, so I left it at that and we had a fun evening. However, the next day he was in the same line as I was waiting for our high-speed train, which was delayed. We decided to have a quick meal together. He ordered his but forgot to ask whether there were any mushrooms in there, as he really disliked those. Upon getting served, he asked the employee, but the employee did not know. “There might be some in the sauce” was the reply, to which he responded “If you don’t know this basic knowledge, you will not see me here anymore.” A second later, he rolled his eyes, glanced at me and told me “Okay, you win.” At that moment, I pushed him over the edge by telling him, “Imagine if everyone could read how this conversation went between you and this business, like on social media.”
It is good to see that more and more businesses understand that a penny not lost is a penny won. Definitely in these economic times, when services that come with a product are becoming more and more the core of many businesses, it is even more important to retain customers than win new ones.
The great technical innovations of today are the huge steps taken in having computers understand human language. The better applications out there offer real-time analytics on incoming customer queries that understand whether a question is urgent or not, whether a customer has to make a lot of effort in getting what they want or not. This allows for the most important questions to be routed immediately to agents, while less urgent ones can wait a bit longer. We all know that some of our questions to brands are not as urgent as others. The issue before was that brands were unable to make that distinction without having all incoming queries read immediately by humans (impossible) or channeling their service to specific channels (often not those that consumers like to use). Text analytics help brands in prioritizing what is most important.
For the complete article continue reading on CX Social.
The Clarabridge Marketing Team recently sat down with Sid Banerjee, Founder and Vice Chairman of Clarabridge and asked him to answer 5 questions on the state of the social media management industry. Here’s his take:
1.What do you believe are the key ingredients for a brand to engage successfully with their customers on social media?
Listen everywhere—don’t depend on a single channel, or a narrow slice of the internet. Engage often—use technology to understand a sincere interest in engagement, and engage. But also take the time to process all interactions, suggestions, ideas, market trends, competitive insights, and build an always-on view of your customers by tracking, trending, alerting, and acting on insights even for those customers you don’t engage with.
2. As an industry insider what are you most excited about concerning innovation in the field?
I’m very excited by the application of deep learning and AI technologies to understand not just what people are saying on the internet, but to understand how questions get answered. How problems are solved. How best to react and respond to threats and challenges. These learning technologies are making increasingly possible to not just find and engage with customers via social media, but to predict how best to respond, and to optimize the timing, and content of responses to ensure your customers feel listened to, and responded to. In time I expect more and more conversations via social media will be man (or woman) to machine, and if the machines are smart, empathetic, and responsive, they will create efficiencies and better engagement outcomes for more organizations who use them.
For the complete article continue reading on CX Social.
CXPA’s 6th Annual Worldwide CX Day presents a fantastic opportunity for Clarabridge to celebrate the increasing significance of customer experience for businesses across the globe. Highlighting the efforts of customer experience professionals who continuously strive to raise the bar for achievement and innovation in the CX field, this day marks an opportunity for Clarabridge to reflect upon our company’s role in the space. Sid Banerjee, Founder and Vice Chairman, and Lorraine Schumacher, Executive CX Advisor and CEM Evangelist at Clarabridge agreed to share their thoughts on what customer experience means to them and the progress that Clarabridge has made operating in this area.
What does CX Day mean to you?
Lorraine: As a board member of CXPA, a certified CCXP and a Clarabridge CX Evangelist, I know that today is a day worthy of celebration. Seeing my personal passion for customer experience reflected by Clarabridge’s core values and commitment to both our customers AND our customers’ customers is truly inspiring. Some may view Clarabridge as text analytics vendor, but you’ll hear from Sid about how the strategic commitment to CX that we have had from the very beginning continues to set us apart from companies that are newer to the discipline.
Sid: As Lorraine noted, Clarabridge has spent over 10 years building a high scalability, high accuracy, high value analytical platform that extracts insights from text and sentiment analytics. Very early on, we recognized core principles that caused us to singularly focus on applying our platform and business to the cause of customer experience promotion. On CX Day, we commemorate this history and the culmination of progress working toward enabling the delivery of superior customer experiences across industries.
What have you learned from working as a CX professional?
Sid: Over the years, we’ve learned three important lessons:
The first is that in order to understand the customer experience, a company must truly understand the voice of the customer. The ability to incorporate data-driven insights into corporate strategy remains an essential component of providing the best customer experience possible.
A second lesson is that that the voice of the customer is everywhere and focusing on just one source of data is not good enough anymore. Businesses may look to surveys, social media or CRM to gain insight into the customer experience, but the range of channels that today’s consumers use to communicate with brands means that failure to consider each and every data source will result in an incomplete understanding of the customer perspective. The amount of customer feedback and interaction data that exists today across emails, calls, chats, survey responses and social media posts is unbelievable, yet many businesses fail to leverage all available information to actively inform customer experience improvements.
Lastly, in order to understand the experiences of customers, you need to obtain a high fidelity understanding of the issues, emotions, cries for help, complaints and stated intents that customers express. To reach this understanding, a company needs best-in-class, AI-powered text analytics. The ability to listen everywhere, analyze everything and engage everyone inside and outside an enterprise is essential. At Clarabridge, we’ve built a business, product and culture that works to ensure this comprehensive understanding and all of their wants, needs, frustrations, passions, values and loyalties so that companies can create authentic, lasting relationships with their customers.
How are you celebrating CX day this year?
Lorraine: CXPA members are celebrating around the world and embracing the opportunity to reflect on their work promoting the customer experience. I will be attending the local CX Day event in SoCal, and many of my colleagues will attend other local CXPA events in their respective locations. As part of the week’s festivities, we are hosting two webinars featuring Bruce Temkin, the “Godfather of CX.” Congratulations to everyone who focuses on the customer experience every day and keep up the good work!