How AI is Changing Consumer Behavior and Decision Making

10 min read.

Will Alexa be deciding which Netflix movie should be your favorite?

From analog to digital, has technology reshaped your brain anatomy and decision-making process? Are companies using you to generate more profits??Is AI enhancing your life or casting a shadow? Or are you soon to be replaced by the unstoppable rise of AI? In this article, we will dive into the influence of AI on human behavior and judgment.

The Psychology of Decision Making

While neuroscientists have been investigating the brain’s decision-making mechanism for decades, Daniel Kahneman’s recent hypothesis outlined in his book “Thinking Fast and Slow” stands out the most. The Nobel Prize-winning psychologist suggests the brain consists of two distinct systems: intuitive and analytical. Making a coffee cup is an example of quick routine decisions performed by the former, whilst the latter is responsible for complex tasks that require heavy information processing such as learning a new language (Kahneman, 2011).

The intuitive system relies on detecting patterns and using mental shortcuts commonly referred to as “heuristics” to accelerate the decision-making process (Richard H. Thaler, 2007). These heuristics are subconsciously influenced by our six senses, previous experiences, or impulsive emotions that shape our choices (Kahneman, 2011). For instance, look at the two cookie jars below, which one would you consider buying? Despite their similar shape and price, most people believe B must taste better than A because the emptier jar indicates higher demand. However, it is possible that A was recently restocked due to higher sales. Failing to take the time to consider additional information can result in misjudgments and ultimately, poor decision-making.

Consequently, the intuitive brain can be manipulated by the external environment to alter its preferences. For instance, marketers have been leveraging heuristics to influence consumers’ buying decisions to drive more sales. In the previous example, Scarcity is one of Cialdini’s persuasion techniques that provokes the availability heuristic which entails if something is rare then it must be good (Cialdini, 2007). Additionally, customers respond positively to free delivery promotions and loyalty programs even though they are not actually free. We unconsciously continue to fall for these marketing strategies because heuristics are prompted by our brains based on context and environment. (Dale, 2015).

The Impact of Technology on Consumer Behaviour

Despite the potential to exploit heuristics, humans were traditionally responsible for making their own decisions until the advent of modern technology breakthroughs that partially integrated with our decision-making processes. AI and digital technologies have transformed our choices’ architecture by increasing information accessibility, personalizing the wide range of choices available to us, and enhancing our social exposure to other people's preferences. Such smart nudging helped our intuitive brains to make faster, more rational, and informed decisions with greater cognitive ease, leading to increased trust and adoption of AI by humans (Cristina Mele, 2021).

Today, customers rely heavily on AI recommenders to discover their next best Netflix movie, Amazon product, Facebook friend, or LinkedIn job. Brands that offer personalized experiences are likely to drive 76% of consumers to make a purchase compared to others (Nidhi Arora, 2021). This is reflected in how customers behave towards sharing personal data and rating products, as they anticipate being rewarded in the future for improving the quality of recommendation systems (Shen, 2014).

However, excessive exposure to digital screens in the past 20 years has had profound consequences on our brain anatomy. As many as 46% of Americans spend between 5 and 6 hours daily on their phones, which is more than one-third of their waking time (Statista, 2022). This has reduced our attention and concentration, degraded our memory capacity, and shifted our social interactions to digital instead of physical. In extreme cases, this has led to serious associations with forms of addiction, such as the Internet and Gaming Disorder (Margret R. Hoehe, 2020). Thus, reduced cognitive capacity has made artificial intelligence more capable of taking over intellectual and creative tasks that were previously handled by humans (Stefano Puntoni, 2021).

While technology has become an indispensable part of our lives, I believe it is crucial for consumers to use it moderately to avoid any adverse effects. It is also essential to embrace the integration of technology in our decision-making process to improve the overall quality of our choices.

Is it Personal or Profitable?

The influence of technology has gone beyond just the customers' experience to transform the entire business landscape. According to McKinsey, companies that excel at personalization are projected to generate 40% more revenue compared to their average counterparts (Nidhi Arora, 2021).

Nowadays, 76% of customers feel frustrated if they do not receive the expected personalized experience, while 62% of consumers have a higher level of trust in companies that use ethical AI (Nidhi Arora, 2021; Anne-Laure Thieullent, 2019).

Between the promising ROI and being customer-centric, the question remains: are businesses developing AI engines to sell the most profitable or the best-fit product for their customers?

To answer this, we must first examine how Marketers build personalized promotions. The data used to train AI models can be classified into two types: individual-specific, and non-individual specific. Private data that can identify a person, such as their name, ID number, or facial recognition, fall into the first category. Whilst non-individual specific data, such as location, purchases, and online behavior, can be shared among multiple people (Sophie C. Boerman, 2021). The key to creating the most effective marketing campaigns is to target unique personalization elements by combining private and behavioral data. Studies have shown that personal interests, location, and age are the three most influential factors that drive the highest conversions and sales (Freya De Keyzer, 2022). For example, in 2018, Facebook generated $56 billion from the sales of tailored ads alone (Stefano Puntoni, 2021).

Nonetheless, customers are sensitive to the type of data collected, how it is shared, and how pricing is determined accordingly. They have presented negative reactions to personalized ads that use their name or photo. Even stronger reactions were captured against personalized pricing ads where customers feel it is unfair to be charged more based on their social status, geography, and other personal factors beyond their control. For instance, Uber uses a person's location to determine ride costs, while hotels use browsing cookies and postcodes to adjust room rates. Such negative emotions lead to fewer conversions and sometimes losing complete trust in the brand (Sophie C. Boerman, 2021). ?

In my opinion, it is crucial for businesses to prioritize their customers’ privacy and maintain their brand reputation by adopting transparent marketing communication and ensuring customers’ consent is granted when necessary. Additionally, marketers should create promotions that align with customers’ preferences rather than solely based on profitability.

AI Can be More Emotional and Creative Than You!

?In late 2022, a new era of AI advancements was marked by the launch of ChatGPT. This cutting-edge technology is capable of interpreting humans’ emotions, and possibly exceeding their creativity in tasks such as creating original art, generating marketing content, and producing music. These breakthroughs are possible due to a form of unsupervised learning, which differs from traditional machine learning models that categorize outcomes based on input data. Instead, deep learning models are self-trained to generate new patterns that were previously unimagined by humans. With the help of emotional AI, marketers can now anticipate consumers' moods and tailor the advertising content using generative AI to persuade them to make purchases (Erik Brynjofsson, 2017).

Ever wondered how Amazon guarantees fast deliveries? This is because it has already predicted you will need this product before you even know it.

The concept of anticipatory shipping discovered by Amazon has transformed how online retailers operate, by introducing a new dimension to forecasting supply and demand. Through utilizing machine learning algorithms, retailers can now ship products to the nearest distribution center even before a customer place an order. Amazon, which has long been known for its customer-centric approach, regards long delivery times as one of its most significant challenges. Nevertheless, anticipatory shipping not only improved customer satisfaction but also reduced fulfillment and shipping costs, which accounted for nearly 26.4% of its total sales in 2017 (sfadavi, 2018).

However, it is possible that such powerful algorithms can exploit human vulnerabilities and manipulate their decisions. In a recently conducted experiment, AI succeeded in changing customers’ preferences in about 74% of cases. The study further revealed that an AI trustee persuaded a human investor to make payments that favored the AI's maximum profit rather than the investor's interest (Amir Dezfoulia, 2020). This is concerning, since we experienced a similar incident through the famous Facebook scandal in 2018, in which politicians manipulated citizens’ votes for their favor (BBC, 2018). Also, it has been recently suggested by Elon Musk that Twitter might use its personalized engine to influence users' political views based on their personal data (Hogg, 2022).

In the future, customers that rely on virtual assistants to make their decisions could potentially be directed towards Alexa’s favorite products rather than items that match their preferences (NIRAJ DAWAR, 2018).

Are You in the Driving Seat Anymore?

Perhaps Humans vs Machines is the longest-standing controversial debate we have faced recently. Transhumanists advocate for the complete adoption of AI-driven decision-making processes, suggesting that humans should step aside and let machines do all the work. They argue that human cognitive biases, lack of objectivity, and incapability of deducing patterns of large information disqualify them from dealing with data. Instead, they believe that humans should only make the final decisions based on the AI's identified options (Colson, 2019).

However, biases are also an AI challenge that could emerge from self-learning algorithms and not intentionally from humans who train them. For example, unsupervised learning methods used to assess creditworthiness may consider gender as a factor, assuming that females are more likely to earn less and thus limit their credit. Similarly, they may fail to consider the needs of minority groups, such as those with disabilities in a queueing system?(Stefano Puntoni, 2021). These shortcomings and the failure to meet social expectations can result in customer resistance to AI. Therefore, humans tend to support augmentation over delegation, as it enables them to have some level of control over the choices they receive.

I find it remarkable how far technology has progressed to date, outperforming humans in some instances, and reserving a creativity seat. However, the ultimate goal of generative AI is to create a human-like interaction that will promote the adoption of technology by humans. While machines may excel at specific tasks, humans will always play a critical role in the decision-making chain, not just as the final step. For example, a marketer must initiate a ChatGPT prompt, guide the conversation to enhance the outcome and make final tweaks to ensure quality creative content is created.

The Final Verdict

In conclusion, the key takeaway for marketers is to engage customers in creating personalized experiences and to constantly find ways to gather their input.

Never Stop Listening to Customers

Ever since we started using analytical tools to understand customers’ behaviour such as Google Analytics and Hotjar, we distanced ourselves from having live interactions with customers which enable listening empathetically to their experiences.

Challenge the AI

Always remember that you know your customers better. Therefore, it is important to evaluate the effectiveness of AI-generated outcomes and ensure that social norms and biases are considered.

?

Further Readings

Amir Dezfoulia, R. N. (2020). Adversarial vulnerabilities of human decision-making. Proc Natl Acad Sci U S A.

Anne-Laure Thieullent, Z. J. (2019). AI and the Ethical Conundrum. Capgemini Research Institute.

BBC. (2018, July 27). Facebook: Explaining the company's massive share slump. Retrieved from BBC: https://www.bbc.com/news/world-us-canada-44978452, Accessed 17th April 2023.

Cialdini, R. B. (2007). Influence: The Psychology of Persuasion. In R. B. Cialdini, Influence: The Psychology of Persuasion. New York, Collins.

Colson, E. (2019). What AI-Driven Decision Making Looks Like. Harvard Business Review.

Cristina Mele, T. R. (2021). Smart nudging: How cognitive technologies enable choice architectures for. Journal of Business Research, 12.

Dale, S. (2015). Heuristics and biases: The science of decision-making. Business Information Review, 14.

Erik Brynjofsson, A. M. (2017). The Business of Artificial Intelligence . Harvard Business Review .

Freya De Keyzer, N. D. (2022). Let’s get personal: Which elements elicit perceived personalization in social media advertising? Electronic Commerce Research and Applications.

Hogg, R. (2022, 5 15). Elon Musk called Twitter newsfeed algorithms 'manipulative,'. Retrieved from Business Insider: https://www.businessinsider.com/musk-and-dorsey-argue-over-manipulative-twitter-algorithms-2022-5?r=US&IR=T, Accessed 17th April 2023.

Kahneman, D. (2011). Thinking Fast and Slow . New York: Farrar, Straus and Giroux.

Margret R. Hoehe, F. T. (2020). Going digital: how technology use may influence human brains and behavior. Dialogues Clin Neurosci.

Nidhi Arora, D. E. (2021, 11 12). The value of getting personalization right—or wrong—is multiplying. Retrieved from McKinsey and Company: https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/the-value-of-getting-personalization-right-or-wrong-is-multiplying, Accessed 17th April 2023.

NIRAJ DAWAR, N. B. (2018). Marketing in the Age of Alexa. Harvard Business Review.

Richard H. Thaler, S. B. (2007). Heuristics and Biases in Retirement Savings Behavior. Journal of Economic Perspectives .

sfadavi. (2018, 11 15). Anticipatory shipping—retail’s crystal ball? Retrieved from Digital, Data, and Design Institute at Harvard: https://d3.harvard.edu/platform-rctom/submission/anticipatory-shipping-retails-crystal-ball/, Accessed 17th April 2023.

Shen, A. (2014). Recommendations as personalized marketing: insights from customer experiences. Journal of Services Marketing.

Sophie C. Boerman, S. K. (2021). When is personalized advertising crossing personal boundaries? . Computers in Human Behavior Reports.

Statista. (2022, June 14). How much time on average do you spend on your phone on a daily basis? Retrieved from Statista: https://www.statista.com/statistics/1224510/time-spent-per-day-on-smartphone-us/, Accessed 17th April 2023.

Stefano Puntoni, R. W. (2021). Consumers and Artificial Intelligence: An Experiential Perspective. Journal of Marketing.