Emotion AI Opens Up New Possibilities for Consumer Research

The idea of using Artificial Intelligence to understand consumer behavior has been around for a while now. From speculating its accuracy to debating its methods, researchers have always had a keen interest in discussing AI’s future in consumer research. Emotion AI is the most noticeable development in this regard.

Emotion AI is the sub-set of Artificial Intelligence that tries to understand human expressions, both verbal and non-verbal. Also known as Affective Computing, Emotion AI is the science of recognizing, interpreting, processing, and simulating human expressions. Affective Computing was first coined in 1995 by Rosalind Picard’s paper of the same name, published by the MIT Press1. Since then, innovators and researchers have been developing research methods coupled with DeepTech solutions to measure and quantify human feelings, emotions, and moods. Biometric sensors, voice analysis, text analysis, and computer vision have been used for data collection.

This article investigates the possibilities Emotion AI brings, augmenting existing research methods to solve the problems of accuracy, bias, and scalability. We discuss Emotion AI technologies like Facial Coding and Eye Tracking; look at their accuracy levels, current-day adoption, and the opportunities they introduce to consumer research.

 

Consumer Behavior and Research: The Status Quo

An average customer is exposed to more than 4000 ads every day and has 12 alternatives for the same product. Moreover, according to Harvard Business School Professor Gerald Zaltman, 95% of our purchase decision-making takes place in the subconscious mind. So, the modern-day customer is not only overwhelmed by choices but also makes these choices subconsciously; relying on body memory and favoring those brands that tend to their subconscious and emotional preferences. Understanding this modern and tech-driven consumer behavior demands new and innovative research methods.

On the other hand, we are halfway through 2021, slowly recovering from a pandemic. The research fraternity has emerged with new realizations, taking cognizance of the agility and scalability that online research tools bring into consumer research. The GRIT Report of 2020 noticed a 12% increase in Online IDIs and a 20% increase in Online FGDs, in 2020 alone. Slow but steady tech disruption is seeping into the field of consumer research with increased adoption of AI and Analytics. A 2021 Gartner report3 cited 27% of the survey respondents seeing AI as the next big gamechanger for organizations to improve their customer experiences.

 

 

Facial Coding

Facial emotion recognition is one of the most popular applications of Emotion AI, with computer vision-based Facial Coding the most scalable one. This technique is based on Paul Ekman’s Facial Action Coding System (FACS)4, a system used to classify facial expressions and movements as universally accepted emotions. FACS breaks down facial micro-expressions into individual components of muscle movement, called Action Units (AUs) and there are 7 emotion-related facial actions classified into 7 universal emotions: Happiness, Sadness, Surprise, Fear, Anger, Disgust, and Contempt. Machine Learning algorithms are trained with each emotion separately to increase the precision and allow them to detect multiple emotions on the same face. Entropik Tech’s Affect Lab Facial Coding has an accuracy of more than 90% in identifying all these emotions by just using a webcam/mobile cam.

 

Applications: Facial Coding is already widely used for the second-by-second analysis of bumper ads, video ads, movie trailers, and TV show pilots. The analysis helps creative directors and content creators to identify high and low engagement points. These insights can be used to optimize their creatives and improve overall virality, creative efficacy, media reach, and targeting before release. Facial Coding can also be used to track the emotional engagement of a shopper throughout their path to purchase. It can also be coupled with Eye Tracking and Surveys for more accuracy and deeper insights.

 

Eye Tracking

Eye Tracking is another computer vision-based Emotion AI technology that has found widespread acceptance and application in consumer research. Eye Tracking is essentially the process of mapping eye gaze/movements to understand where the viewer is looking, what they are looking at, and for how long. Eye gaze heatmaps, gaze plots, and Area of Interests (AOI) are plotted to measure attention, engagement, and noticeability metrics. The technology is so advanced and accurate today that it is possible to use a mere mobile cam to tell where a person is looking at and for how long within minutes. In fact, Entropik Tech’s Affect Lab Eye Tracking is capable of delivering 96% accuracy in a controlled environment and 91% in a wild environment.

All product and company names are trademarks™ or registered® trademarks of their respective holders. Use of them does not imply any affiliation with or endorsement by them.

 

Applications: Eye Tracking is already widely used to test and optimize media, shopper, and digital experiences. Be it second-by-second testing of ads or videos; be it to analyze package designs or shelf placements/store layouts; be it testing an app/website/e-commerce store for goal completions and conversions; Eye Tracking can be used in any consumer behavior study that involves measuring the customer’s attention, engagement and noticeability. It can also be coupled with Facial Coding and Surveys for more accuracy and deeper insights.

 

Emotion AI Driven Consumer Insights

Consumer insights that are driven by Emotion AI definitely add more value and stand out when compared to consumer insights by traditional methods like surveys, in-depth interviews and focus group discussions. Here’s how:

• Accurate and Real: Technologies like Facial Coding and Eye Tracking have more than 90% accuracy when it comes to mapping non-verbal cues of human behavior. They easily measure metrics like emotions, attention, engagement, and noticeability in real-time, accurately capturing every human expression.
• Rich and Actionable: Not only Emotion AI-powered insights are real/accurate, but they are also incredibly richer and actionable. Firstly, they make it possible for researchers to effortlessly quantify human emotions and subconscious behavior. Secondly, these insights can be interpreted and visualized as quantified data for agile decision-making.
• Bias-free: Emotion AI marginally removes the inherent bias of a moderator or respondent because emotions are measured in real-time. However, surveys and interviews can always be added for precision, again coupled with Facial Coding and Eye Tracking to ensure the genuineness of responses.
• Scalable: Despite the complexity of tech involved, computer-vision-based Facial Coding and Eye Tracking are incredibly easy to use and implement. With no additional hardware requirement and readily available online respondent panels, researchers can remotely generate consumer insights 4X faster than traditional methods.

 

The Way Forward with Emotion AI

If we rewind back to a decade, probably the skepticism around Emotion AI technologies seemed valid. But technological advancements have brought about radical shifts in consumer behavior and research in the past decade. Along with technological advancements, the providers of these Deeptech solutions have also gained the market’s confidence in terms of data privacy and security. Moreover, highly accurate technologies like Facial Coding and Eye Tracking have brought about agility, scalability, and cost-efficiency with their computer-vision-based data collection methods.

First-mover brands and market research agencies have recognized the potential of Emotion AI. They are adopting this AI-based, analytical approach to augment their existing research methods and elevate their consumer insights teams. Emotion AI has certainly arrived and opened up a plethora of opportunities for decision-makers to humanize their consumer insights. To deliver meaningful customer experiences that can help them stand out and connect with their distracted and overwhelmed, modern-day customers.

 

References:

1. “Affective Computing”, MIT Press, 1995: https://affect.media.mit.edu/pdfs/95.picard.pdf
2. “How to use AI to Improve the Customer Experience”, Gartner, 2020: https://www.gartner.com/en/documents/3983583/how-to-use-ai-to-improve-the-customer-experience
3. “GRIT Report – Insights Practice Edition 2020”, Greenbook, 2020: https://www.greenbook.org/mr/grit/insights-practice-edition/
4. “FACS”, Paul Ekman Group, 1978: https://www.paulekman.com/facial-action-coding-system/

 

Photo by Andrea Piacquadio from Pexels