Is Emotion Analytics the Holy Grail in CX?
Customer Journey Mining
Is Emotion Analytics the Holy Grail in CX?
Recently a new report was published by Credence Research in which they predict that the market for Emotion Analytics is expected to expand at a CAGR of 15.6% during the forecast period from 2019 to 2027. This confirms figures published earlier this year by MarketsandMarkets: they predicted that the global emotion analytics market size would grow from USD 2.2 billion in 2019 to USD 4.6 billion by 2024.
These figures would suggest that emotion analytics is indeed the next Holy Grail in CX, but what is emotion analytics, and why do analysts predict a huge growth?
How are companies capturing emotions today?
Today companies are combining customer journey techniques with qualitative and quantitative research methods to try and get an understanding of customer emotions at key points in their journeys.
But the process of representing emotion in customer journey maps is not without its challenges. For a start, accurately capturing the emotional state of the customer along their various pathways on the journey can be problematic. Customers can share their emotions if asked, but are they asked at the right time or do they feel like being honest?
“Today companies are combining customer journey techniques with qualitative and quantitative research methods to try and get an understanding of customer emotions at key points in their journeys.”
So how do you capture customer emotions?
Emotions can be captured using a range of qualitative and quantitative research techniques. Qualitative research is an excellent way to capture in-the-moment emotions – positive, neutral, or negative – at different touchpoints along the journey. It helps you capture the intensity of those emotions and the behaviors they drive, which will help you assess which touchpoints are really ‘make or break’ moments.
Some of the key qualitative journey research methods include:
- Contextual interviews
- Observational interviews
- Digital diary studies
- Focus groups
- Community boards
As well as qualitative approaches, there are also quantitative ones that can be applied.
Researchers often use quantitative approaches to back up qualitative insights with ‘hard data’, they are a powerful way to learn what customers ‘are really doing’ and prioritize which areas of the journey are critical for deeper exploration.
Quantitative research methods include:
- Text mining open texts and surveys
- CSAT and Net Promoter Score (NPS) assessments
- Website tracking data, CRM data, telephony data etc.
- Customer journey analytics*
*Customer journey analytics is an approach that analyses customer behaviors and motivations across touchpoints and over time to optimize customer interactions and predict future behavior. Customer journey analytics can reveal which stages of your customers’ experience matter most. This way, your survey can ask your customers to rate or describe their emotional reaction to that area of their experience.
Emotion Analytics; the next steps
Experts describe Emotion Analytics (EA) as software which collects data on how a person communicates verbally and nonverbally to understand the person’s mood or attitude.
In the illustration below all types of Emotion Detection and Recognition Techniques have been described:
“Customer journey analytics is an approach that analyses customer behaviors and motivations across touchpoints and over time to optimize customer interactions and predict future behavior.”
All the techniques outlined above are being evaluated and used to a greater and lesser extent today.
Open texts, speech and spoken language communication cues are an important means for measuring and modelling human behavior; with speech, it is not just what a person says, it is “how” they say it. Part of the “how” includes emotion: for example, was a customer angry, happy, or neutral? And will they buy/stay-on as a customer
The technology provides insights into how a customer perceives a product, the presentation of a product or their interactions with a customer service representative. Scientific advances in how we quantify unstructured signal data – like text, speech/audio, and video – allow us to quickly gain understanding from large datasets.
Much is experimental at present, but within the next 5 to 10 years, this technology might surpass humans in understanding human emotions because it will tap into data that humans can not perceive on their own: using biometrics, brain waves, subtle cues from body language and facial expressions, and more.
“Open texts, speech and spoken language communication cues are an important means for measuring and modelling human behavior; with speech, it is not just what a person says, it is “how” they say it.”
The evolution in emotional text analytics
Zooming into the field of emotional text analytics huge steps have been made up to date. But much is still to be explored and trained, using AI and deep learning techniques.
Building on the developments in sentiment analysis in the past decade, which is increasingly being adopted by leaders in CX, the race is on to developing the necessary next steps to achieving maximum emotion detection using full integration of text analysis with facial coding, tone, EGG, biometrics etc.:
“Emotion Analytics technology promises to deliver fascinating new types of data, which we will need to learn to use and refine over time.”
- Sentiment analysis, current techniques
Sentiment analysis is the automated process that uses AI to identify positive, negative, and neutral opinions from text. Sentiment analysis is widely used for getting insights from social media comments, survey responses, and product reviews, and making data-driven decisions.
- Finely grained emotion recognition
This is a step up from sentiment analysis in the field of emotion classification, which is a more fine-grained form of sentiment analysis that focuses on extracting emotions from text like joy, anger, and fear.
- Contextual emotion comprehension
Recent discoveries in the field of emotional intelligence show that emotions should be perceived as context-sensitive engagements with the world. This leads to a need to specify whether the emotions conveyed in a conversation are appropriate for a situation they are expressed in.
- Full EQ making decisions
Empathetic technology is the next big thing. And it boils down to technology that ‘reads’ emotional states and well-being but also responds to it. It opens a new dimension in the relationship between people and technology. Using AI and deep learning techniques, technology will become emotionally sensitive.
- Full integration of text, speech, tone, facial coding etc.
This integration step will happen along the way but is the Holy Grail and the end game in emotion analytics. Experts say this step is on the near horizon, set to happen 5-8 years down the line.
d
Emotion analytics in customer experience management is here to stay
Emotion Analytics technology promises to deliver fascinating new types of data, which we will need to learn to use and refine over time. As with all of the data-informed marketing developments that have preceded it, the true test of emotion analytics will be in how insights are drawn and acted upon.
Many companies across the world are harnessing the power of emotional intelligence to improve their business processes and customer experience today. Emotion analytics is here to stay, and huge investments are currently being made and are predicted to explode.
Underlined does continuous research with customers’ and academic partners into emotion analytics. One major partner is the JADS (Jheronimus Academy of Data Science). Goal: to develop new techniques that constantly gain better (emotional) insights from data. This research focuses on how we can help the increasingly mature CX initiatives steer towards more data driven working and to gain insight into increasingly complex customer relationships and behavior.
Our current studies focus on emotion detection in data based on Artificial Intelligence applications. We are also working on techniques to make journeys transparent with process mining products. Traditionally these products have the process as a starting principle, which is really a different starting point (inward looking) than a customer journey (outward looking). How this can come together is a key area of research for us.