Text analytics is the process of analysing unstructured text, extracting relevant information, and transforming it into useful business intelligence.
Why is text analytics important?
Emails, online reviews, tweets, call center agent notes, and the vast array of other written feedback, all hold insight into customer wants and needs. But only if you can unlock it. Text analytics is the way to extract meaning from this unstructured text, and to uncover patterns and themes.
Speech Analytics is the process of extracting meaning from audio recordings and analysing it to find relevant business intelligence. Related to audio mining, speech analytics is often performed using specialised speech analytics software that can understand the spoken word of many dialects and translate it into text.
Why is Speech Analytics important?
51% of consumers and a full 92% of businesses say that the phone is their preferred channel of customer/business interaction. Obviously, call recordings are a major source of customer feedback, particularly about areas that are causing dissatisfaction for your customers.
How do you capture and process the data from your customer calls efficiently? Speech analytics software automatically parses call records so that critical business insights are not lost.
Highly regulated industries such as financial services and healthcare can particularly benefit from speech analytics, as they have compliance requirements regarding storing and searching for customer data. They also have a pressing need for the earliest possible indication of compliance violations – and people calling with complaints or questions can definitely be a red flag. Audio mining is also a key factor in the discovery process in the event of litigation.
Sentiment Analysis is the measurement of positive and negative language.
It is a way to evaluate written or spoken language to determine if the expression is favorable, unfavourable, or neutral, and to what degree.
Today’s algorithm-based sentiment analysis tools can handle huge volumes of customer feedback consistently and accurately. Paired with text analytics, sentiment analysis reveals the customer’s opinion about topics ranging from your products and services to your location, your advertisements, or even your competitors.
Why is Sentiment Analysis important?
Sentiment analysis is critical because helps you see what customers like and dislike about you and your brand.
Customer feedback—from social media, your website, your call center agents, or any other source—contains a treasure trove of useful business information. But, it isn’t enough to know what customers are talking about. You must also know how they feel. Sentiment analysis is one way to uncover those feelings.
Sometimes known as “opinion mining,” sentiment analysis can let you know if there has been a change in public opinion toward any aspect of your business. Peaks or valleys in sentiment scores give you a place to start if you want to make product improvements, train sales or customer care agents, or create new marketing campaigns.
Sentiment analysis is not a once and done effort. By reviewing your customer’s feedback on your business regularly you can be more proactive regarding the changing dynamics in the market place.
Interaction Analytics are the processes of uncovering meaningful insights from unstructured natural language conversations.
Why do you need Interaction Analytics?
Conversations are gold mines of information about your customers and their perception of your organisation. While traditional surveys are sampled and biased, leveraging Interaction Analytics to evaluate phone conversations, social media exchanges, or Live person chats reveals the holistic, actual voice of the customer. Taking advantage of the treasure trove of unsolicited feedback within their contact centres empowers organisations to make more informed decisions more quickly, thus saving time, money, and headaches.
Intelligent Scoring is a patent-pending, proprietary feature that allows customers to distill many complex criteria into a customisable, unifying measurement. It is is a breakthrough in omnichannel interaction analytics that can automatically evaluate and score both conversational and non-conversational data sources.
What’s a score?
We see scores all the time, often without questioning their origin. For example, consider safety inspection ratings at restaurants; we’ve been socialised to understand that an A is better than a B, which is better than a C, and so on. This understanding helps us decide which ratings we’re willing to risk, and which we might pass on when choosing a restaurant for a quick lunch. Let’s dissect why these ratings are so useful.
Behind the scenes, over fifty different criteria—each weighted according to their likelihood of causing food-borne illness—are distilled into a single score, which then is ranked on a scale and placed in public for potential diners to see. Rather than forcing diners to sift through pages of information, the single measurement quickly displays a cohesive story about the state of the facility so restaurant-goers can make their own informed decisions about where to eat.
What counts as an Intelligent Score?
Intelligent Scoring automatically distills many complex criteria into a single unifying score, harnessing features across the entire Discover product suite. You can leverage enrichments like effort and sentiment to create customer experience scores without surveys, build complex conditional logic, and deploy root cause features like Drivers and Outliers to ensure your index captures what you’re aiming to track.
In a nutshell: if you can ingest it into Qualtrics, you can score it, regardless of the data source. Intelligent scoring can be used to measure agent quality, customer experience, sales efficacy, legal risks, or even to monitor consumer reactions to current events.
Intelligent Scoring answers questions like:
- Which contact center agents are struggling to answer questions about new products?
- Which agents are better equipped for chat customer service as opposed to phone channels?
- How can I improve my agent training materials based on score discrepancies between new agents and more tenured agents?
- Which interactions have multiple important indicators of legal risk?
- Which customers are good candidates for proactive outreach based on poor experiences with company policy?
- Which customers are having poor experiences with my brand, despite having interacted with highly trained agents?
- Which current events are my customer base reacting most strongly about, and how can I future-proof my organization’s policies accordingly?
- Which agent behaviours most correlate to high customer satisfaction?
Quality Assurance (QA) is the practice of maintaining consistent standards when delivering a product or a service to a customer.
QA is crucial for contact centres as it ensures a consistent level of customer service that matches predefined requirements. Customers increasingly view the experience they have when interacting with a company to be equally as important as its products. Therefore, ensuring both quality and consistency in customer experience is key to making customer care a competitive differentiator.
QA also improves contact center performance by enforcing consistent brand messaging, accelerating conflict resolution, and expediting identification of recurring customer issues.
What processes do QA programs follow?
To maintain brand standards, most QA programs follow these main processes:
Traditional rubrics or scorecards often evaluate performance by examining the presence of desired behaviours, some of which may be easier to consistently measure than others. For example, a rubric might evaluate the presence of a required disclosure statement or whether a representative showed soft skills like empathy. While a disclosure statement is relatively straightforward to both communicate and identify, soft skills requirements are often fuzzier and leave much up to the subjective interpretation of both the auditors and the representatives themselves.
Contact centres evaluate a percentage of their representatives’ calls. Auditors evaluate adherence to expected behaviours; however, when done manually, this process introduces the potential for subjective variation and human error. Extreme sampling can also make it harder to find significant trends that inform smart coaching.
Coaches may work with representatives to review customer interactions and correct undesired behaviours. However, this approach focuses on punitive corrections and does not reinforce desired behaviours. Additionally, coaches are limited by the sample size, hours in a day, and the audit timeframe. Coaches may not give feedback to an agent until a month after a call, thereby making it very difficult for the agent to remember the interaction and learn from the situation, Additionally, a lack of historical records makes it challenging for agents and supervisors to track improvement over time.
Quality Management (QM) is the process of continuous improvement based on setting goals, identifying deviations from these goals, and adjusting processes and behaviours accordingly.
Quality Management is the combination of QA and process improvement, making it a critical component of successful contact center operations. Potentially a customer’s first point of contact, the contact center presents a unique opportunity for an organisation to directly and often permanently impact a customer’s experience and overall perception of the company. For example, a representative who fails to disclose a privacy statement or who is rude to a customer could inadvertently land the organisation in legal trouble, elicit bad press, or both.
QM programs improve contact center performance by clearly defining desired representative behaviours, identifying interactions that miss the mark, and providing mentorship based on these “coachable moments.” A mature QM program not only improves customer perceptions of an organisation, but also empowers representatives to become and remain effective frontline drivers of both positive CX and consistent regulatory compliance.
What do current approaches to Quality Assurance look like?
Customers’ expectations are higher than ever, and companies who successfully meet these rising expectations have a tremendous opportunity to differentiate themselves. Heightened customer expectations and increasing standards mean that QA must also be optimised accordingly. For example, “customer-obsessed” Zappos describes its purpose as “to live and deliver ‘WOW!’” Therefore, a Zappos contact center representative who doesn’t deliver “wow” to a customer is not acting in alignment with the brand’s mission.
What do current approaches to Quality Management look like?
While there is a spectrum of maturity for QM programs, most follow these core processes:
Traditional rubrics or scorecards often evaluate performance by examining the presence of desired behaviours, some of which may be easier to consistently measure than others. For example, a rubric might evaluate the presence of a required disclosure statement or whether a representative showed soft skills like empathy. While a disclosure statement is relatively straightforward to both communicate and identify, soft skills requirements are often fuzzier and leave much up to the subjective interpretation of both the auditors and the representatives themselves, leading to inconsistent scoring.
Contact centres evaluate a percentage of their representatives’ calls. Auditors evaluate adherence to expected behaviours; however, manual quality management processes introduce the potential for subjective variation and human error. Extreme sampling can also make it harder to find significant trends that inform smart coaching.
Coach on behaviours:
Coaches review representatives’ performance to reinforce organisational standards and expectations. However, coaches are limited by the number of audited calls, hours in a day, and the audit timeframe. Coaches may not give feedback to an agent until a month after a call, thereby making it very difficult for the agent to remember the interaction and learn from the situation, Additionally, a lack of historical records makes it challenging for agents and supervisors to track improvement over time.
QM programs build on QA processes by uncovering macro-level trends across contact center interactions and incorporating these discoveries into operation-wide improvements. However, if an organization only considers a sample of interactions, the insights it finds may be misleading or inaccurate.
Natural Language Processing
Natural Language Processing (NLP) is a field of computer science that deals with applying linguistic and statistical algorithms to text in order to extract meaning in a way that is very similar to how the human brain understands language.
Why is Natural Language Processing important?
About 95% of customer data is found in the form of unstructured text – in emails, survey write-in answers, Twitter posts, online reviews, comments in forums, and more.
Reading through all of this text is next to impossible: assuming that the average person can process 50 items of unstructured data an hour, it would take nearly seven years for one person to read through one million items.
Natural Language Processing automates the reading of text using sophisticated algorithms. Fast, consistent, and programmable, NLP engines identify words and grammar to find meaning in large amounts of text.
Natural Language Understanding
Natural Language Understanding (NLU) is a field of computer science which analyses what language means, rather than simply what individual words say.
This area of research and development relies on foundational elements from natural language processing (NLP) systems, which map out linguistic elements and structures. NLU seeks to intuit many of the connotations and implications that are innate in human communication such as the emotion, effort, intent, or goal behind a speaker’s statement.
Why is Natural Language Understanding important?
Human language is fluid, complex and full of subtleties. Two people may read or listen to the same passage and walk away with completely different interpretations. If humans struggle to develop perfectly aligned understanding due to these congenital linguistic challenges, it stands to reason that machines will struggle as well.
The field of NLU is dedicated to developing strategies and techniques for understanding context in individual records and at scale. In order to categorise or tag texts with humanistic dimensions such as emotion, effort, intent, motive, intensity, and more, NLU systems may leverage both rules based and statistical machine learning approaches. Together, these techniques are applied to resolve tasks such as content analysis, topic modelling, machine translation, and question answering. At scale, NLU systems empower analysts to distill large volumes of text into coherent groups without reading them one by one.
Natural Language Generation
Natural Language Generation, otherwise known as NLG, is a software process that utilises Natural Language Processing (NLP) to produce natural written or spoken language from structured and unstructured data. The most common methods of NLG are extractive and abstractive. An extractive approach takes a large body of text, pulls out sentences that are most representative of key points, and concatenates them to generate a summary of the larger text. An abstractive approach creates novel text by identifying key concepts and then generating new language that attempts to capture the key points of a larger body of text intelligibly.
Why is Natural Language Generation important?
Unbeknownst to the reader, NLG techniques are commonly used in popular data analysis and reporting outlets, such as sports and financial reports. In these applications, numeric values from a box score or financial statement are populated into a machine-generated article that describes those figures or scores. This format helps make quantitative, structured data more storylike. Advancements in NLP and NLU have opened the door to introducing NLG processes into the world of Customer Experience Management. Now, analysis of millions of customer conversations can be transformed into machine-generated summaries tailored for specific use cases making data consumption more convenient and palatable for larger audiences.
Social Engagement is the process of communicating (engaging) in an online community
The conversation can take place on individual social media platforms such as Instagram, Twitter, Facebook and LinkedIn, or in blogs, forums and third-party review sites. A strong social engagement strategy allows businesses to remain in constant contact with their customers by advancing brand interests and responding to feedback in a seamless cycle.
Why is Social Engagement important?
When companies engage and respond to customer service requests over social media, those customers end up spending 20% to 40% more money with the company.* Conversations regarding your organisation, your industry and your interests are constantly occurring online, whether or not you choose to participate. Only once you join a conversation can you have an effect on its direction and its outcome. The use of social engagement tools, along with a sound strategy and strong messaging, allow you to interact with your audience and shape the conversation surrounding your brand or product.
*Bain & Company
Customer Intent is often understood as just buyer intent. However, in the customer experience and service space, it can mean much more than just the reason for a call or a chat, or a purchase. Discover how to use consumer intent to enhance your business.
Customer Intent is often understood as buyer intent, or the purpose or reason behind a statement or action as part of a customer’s journey toward a purchase. However, in the customer experience and service space, it can mean much more than just the reason for a call or a chat or a purchase. Detecting consumer intent helps you understand and classify what a customer is trying to communicate beyond what they are saying or talking about, whether it’s a cry for help, a question about a feature, or a heartfelt thanks.
Why is customer intent detection important to businesses?
When analysing customer feedback, being able to isolate certain kinds of intents can lead to productive actionability. With basic NLP, businesses are able to categorise the topics and analyse trends from a customer interaction. And with advanced machine learning techniques and intent classification, business analysts and customer support representations can better understand the subtext behind what customers are communicating to better determine the next best action.
Compare “Your website sucks!” and “Your website would be so much easier to use if the chat box didn’t cover up the login area!” While we might be drawn to the obvious negativity in the first sentence, it is the second one that we would deem as actionable. It offers an explicit suggestion that unlocks valuable information we can use to identify specific pain points and to design customer-centric solutions.
Intent detection goes beyond hearing what customers say and enables active listening to customers’ needs through customer intent data.