This PhD thesis Literature Review chapter sample focuses on the challenges of ethical AI-driven hyper-personalisation in marketing digital content in Canada. The author provides an overview of the hyper-personalisation concept and its constituent elements and reviews existing debates on the ethical considerations and key barriers to the adoption of such instruments.
Chapter 2: Literature Review
2.1. The Advancement of Hyper-Personalisation
The concept of hyper-personalisation has evolved over the years as one of the unique features in contemporary digital marketing (Jain et al., 2021; Singh & Kakkar, 2025). Traditional approaches in this sphere were based on the analysis of past customer data to generate insights, allowing practitioners to develop the most lucrative offerings for specific clients.
However, such tools as contextual awareness, deep learning, predictive psychographics, and behavioural biometrics potentially allow companies to make such propositions in real-time communication (Hess et al., 2020). By dynamically adapting the marketing communication style, content, and offerings, hyper-personalised marketing allows practitioners to tailor these elements to specific consumer characteristics (Hari et al., 2025). Potentially, this approach can increase conversion rates and help firms establish more effective relationships with their clients. Specific tools used in hyper-personalisation include collaborative filtering, 360-degree customer profiling across different devices, and real-time decision support.
As noted by Mendia and Flores-Cuautle (2022), hyper-personalisation strategies can be generally based on three dimensions of customer data. Identity included personal characteristics such as first and last name, date of birth, tax ID, and other similar information, allowing automated systems to authenticate individuals. Contactability reflected the contact data shared by the customer including their phone numbers, social network accounts or address (Micu et al., 2022). Finally, traceability was related to the history of interactions with a certain client including their visits to stores and/or e-stores and transactions or receipts (Ngoc Thang et al., 2020).
Both types of contact could be useful to produce insights into products of potential interest and contribute to hyper-personalised offerings development. The combination of these three dimensions for campaign development and ongoing communication was found to produce a variety of positive effects on the growth of customer numbers and customer retention over time (Kalia & Paul, 2021; Mendia & Flores-Cuautle, 2022).
In the content marketing niche, the promise of hyper-personalisation is primarily related to several key areas (Florido-Benitez, 2024). First, the analysis of consumer behaviours can provide deeper insights into the types of content potentially interesting to certain clients. This can improve the quality and accuracy of recommendations and stimulate up-selling and cross-selling. Second, the analysis of interactions with content can help firms personalise user experiences and make them more convenient and unique (Yildiz et al., 2023). This can increase the time spent on the websites and overall satisfaction levels as well as customer engagement levels. Third, the sheer power of machine-learning-based personalisation potentially allows marketers to achieve the long-sought 1:1 segmentation (Rosenbaum et al., 2021). This hyper-level was not available previously but is presently experimented with by such companies as Tesco introducing individual discounts based on specific buyers’ behaviours and purchase histories within the scope of the existing Clubcard loyalty programme (Davenport, 2023).
According to Netflix, as one of the pioneers of hyper-personalisation in digital content marketing, more than 80% of content items viewed by platform users were based on prior personalised recommendations (Kaponis et al., 2024). This figure clearly demonstrates that well-tuned prediction algorithms can stimulate up-selling and cross-selling activities of customers. However, it may not be clear if similar effects could be noted in services based on per-item payments rather than subscription-based content platforms.
The difficulty of realising the hyper-personalisation strategy without AI instruments is further supported by the fact that consumers contact brands via multiple touchpoints and channels (Naz & Kashif, 2025). As a result, it may be impossible to provide individualised approach if the history of such interactions is not recorded, analysed, and transformed into marketing insights. As also noted by Chandra et al. (2022), personalisation can be realised in various formats including personalised recommendations, relationships, advertising, and discourse. Hence, it may influence both the style and content of promotional messages, which provides substantial flexibility and creativity, as well as the reduction of ‘customer fatigue’ caused by sub-optimal suggestions and/or the inability to quickly find interesting information, products or services.
2.2. Ethical Personalisation and AI
One of the main problems related to ethical personalisation in digital marketing is called the privacy paradox (Dinulescu et al., 2022; Saura, 2024). It is associated with the fact that most users are concerned about the safety of their personal data they share with websites, social media, and apps. At the same time, they are generally willing to sacrifice some part of their privacy in exchange for free services or intelligently personalised content. The main problem with this paradox is associated with the fact that the two goals may be mutually contradictory in some cases where greater and more effective adaptation requires an increase in the quality and quantity of data for processing (Guendouz, 2023). This situation is further complicated by the increasing interest of national governments towards user data privacy (Naqvi et al., 2024). The introduction of such regulations as GDPR in the EU or the Data Protection Act in the UK has already led to several high-profile scandals involving market leaders such as Facebook and Google. These issues are drawing the attention of the general public towards the problems of personal data privacy and further reduce the willingness of regular customers to share their data with companies.
AI technologies are presently used by major corporations including Amazon, Netflix, YouTube, Facebook, and Yahoo to optimise their ad targeting approaches (Gao et al., 2023; Zywiolek et al., 2022). This involves the analysis of user data to produce personalised promotional messages aligned with their interests and needs. As shown in the following figure based on the analysis of these companies, this process involves the use of a/b testing procedures, programmatic advertising, target analysis, and other elements involving the collection of user data.
On the one hand, this raises the questions of ethicality in relation to the collection of such information (Kouroupis et al., 2022). Canadian regulations including the Personal Information Protection and Electronic Documents Act (PIPEDA) suggest that activities, such as user behaviour tracking across multiple platforms, require explicit informed consent (Saura et al., 2024). This makes it difficult to implement real-time hyper-personalisation tactics without violating established principles of ethical communication and data handling. On the other hand, these regulations suggest that users must possess the right to withdraw their consent and completely delete their private data from such platforms at any moment. This further complicates the use of longitudinal big data tools relying on large-scale customer information samples.
Figure 1
AI-Powered Advertising Elements
Source: Gao et al. (2023, 11)
Similar ideas were voiced by Vishwakarma et al. (2025) noting that ethical marketers must prioritise the opt-in strategy over the opt-out one in the sphere of personalisation and hyper-personalisation. While past practices in this sphere frequently involved the collection of customer data without explicit consent, the development of public opinions and regulatory practices clearly implies that this approach is considered sub-optimal and dishonest. Additionally, the study by Sameen (2025) concluded that the willingness to share additional information can be moderated by trust towards certain brands. The activities of companies associated with unethical behaviours and the overall lack of transparency in the past were frequently considered more intrusive and similar to surveillance than similar actions of ethical brands. Considering the fact that many consumers had adverse past experiences in this sphere, the promotion of openness can be seen as a crucial element of digital marketing focused on hyper-personalisation (Hardcastle et al., 2025). In this scenario, full transparency regarding the aims of data collection and data handling and security procedures could win customer trust and convince users to share their information with companies demonstrating ethical integrity.
2.3. Key Challenges of AI-Driven Hyper-Personalisation in Digital Content Marketing
While hyper-personalisation offers a wide range of benefits, this technology is also associated with a number of implementation barriers related to regulatory, technical, ethical, and human resource management spheres (Malgieri, 2023; Wong & Floridi, 2023). Real-time consumer sentiment processing and 360-degree profiling require access to customer data and the use of behavioural tracking. From the standpoint of local regulatory acts such as Quebec’s Law 25 in Canada, such algorithmic analyses can be partially or fully prohibited. Since such activities can interfere with consumer decision-making in a manipulative manner, they are also associated with ethical and reputational risks (El Emam et al., 2024).
Past examples of failed personalisation attempts included Mother’s Lounge attempting to target women on the basis of their location and age to promote pregnancy-related products. With one of the recipients receiving the message on the one-year anniversary of a miscarriage, the resulting backlash reflects the risks of using incomplete or inaccurate data for marketing messages customisation (Olson et al., 2021). From the standpoint of hyper-personalised digital marketing, this creates a dilemma where explicit information about customers allowing firms to make accurate algorithmic adjustments may not be available due to privacy regulations while partial information can lead to marketing communication failures.
Similarly, time-limited offerings and other instruments capitalising on the fear of missing out could be equally controversial from an ethical standpoint according to Singh and Kakkar (2025). Another challenge mentioned by Davenport (2023) was associated with technological barriers to hyper-personalisation adoption in modern firms. To realise the economies of scale in this sphere, firms had to automate the processing of data and use machine learning models. On the one hand, this required large volumes of high-quality data (Liu et al., 2020). With some consumers opting to not disclose their full personal details due to privacy concerns and some states setting limits to the collection of such information, this poses a serious challenge.
On the other hand, machine learning models are susceptible to bias and erroneous judgment, which requires human supervision and expertise in this area (Saura et al., 2022). Skilled personnel possessing these capabilities was not available to many firms, which acted as a barrier to the adoption of this strategy. Since its implementation requires the use of big data, advanced technological instruments, and highly qualified experts, this makes it available only to large companies with substantial resources. Moreover, the high risks of privacy laws violations can further decrease the willingness to adopt hyper-personalisation even in the case of corporations.
Additional challenges faced by digital content providers are related to the categorisation of content (Davenport, 2023; Rosario & Dias, 2023). Netflix can be seen as one of the pioneers of this approach using its proprietary recommendations engine for many years. The company explored the content in terms of overall tone, directors, topics raised, and combined this information with users’ watch histories (Liu & Tao, 2022). This hyper-personalisation system also took into account the time of the day, the average lengths of past viewing sessions, and device types to make highly specific recommendations tailored to the preferences of individual customers. With that being said, this degree of consumer centricity may be difficult to achieve for smaller platforms (Lee & Hancock, 2024). Additionally, some types of content such as educational courses may be more difficult to promote in this manner in comparison with books, videos, games or apps. This calls for a more systemic approach to hyper-personalisation suggesting a uniform strategy that can be applied by most firms.
Finally, one of the technological challenges highlighted by Chen et al. (2023) involved the need to integrate hyper-personalisation into customer relationship management platforms. With digital content marketing being a longitudinally-oriented strategy, this implies that customers may need to engage in multiple interactions with brands. As shown by the earlier analysed examples from Netflix and other market leaders, these points of contact must be used to collect and consolidate data about their preferences and match it with brand offerings (Dwivedi et al., 2021).
With that being said, this approach can create controversies between item-centric and customer-centric approaches. The first one implies that filtering should be based on content properties where clients interested in certain types of offerings get recommendations of similar products (Chandra et al., 2022). On the contrary, the customer-centric approach starts with the analysis of customer characteristics that are matched with specific items in a unique combination. While the latter strategy is used by Netflix and other technologically advanced brands, it requires substantial client data as well as unique expertise in this sphere (Rosenbaum et al., 2021). This creates a dilemma where firms lacking these resources cannot use third-party services due to increased risks of user data privacy violations and data breaches. Moreover, the provision of internal product information for external analysis can also be deemed problematic and lead to copyright-related issues.
References
Chandra, S., Verma, S., Lim, W., Kumar, S., & Donthu, N. (2022). Personalization in personalized marketing: Trends and ways forward. Psychology & Marketing, 39(8), 1529-1562. https://doi.org/10.1002/mar.21670
Chen, X., Miao, S., & Wang, Y. (2023). Differential privacy in personalized pricing with nonparametric demand models. Operations Research, 71(2), 581-602. https://doi.org/10.1287/opre.2022.2347
Davenport, T. (2023). Hyper-personalization for customer engagement with artificial intelligence. Management and Business Review, 3(1-2), 29-36. https://doi.org/10.1177/2694105820230301006
Dinulescu, C., Visinescu, L., Prybutok, V., & Sivitanides, M. (2022). Customer Relationships, Privacy, and Security in Social Commerce. Journal of Computer Information Systems, 62(3), 1-18. https://doi.org/10.1080/08874417.2021.1975172
Dwivedi, Y., Ismagilova, E., Hughes, D., Carlson, J., Filieri, R., Jacobson, J., Jain, V., Karjaluoto, H., Kefi, H., Krishen, A., Kumar, V., Rahman, M., Raman, R., Rauschnabel, P., Rowley, J., Salo, J., Tran, G., & Wang,Y. (2021). Setting the future of digital and social media marketing research: Perspectives and research propositions. International Journal of Information Management, 59(1), 1-20. https://doi.org/10.1016/j.ijinfomgt.2020.102168
El Emam, K., Fineberg, A., Jonker, E., & Pilgram, L. (2024). Perspectives of Canadian privacy regulators on anonymization practices and anonymized information: A qualitative study. International Data Privacy Law, 14(4), 391-403. https://doi.org/10.1093/idpl/ipae017
Florido-Benitez, L. (2024). Generative artificial intelligence: A proactive and creative tool to achieve hyper-segmentation and hyper-personalization in the tourism industry. International Journal of Tourism Cities, 11(1), 83-103. http://dx.doi.org/10.1108/IJTC-05-2024-0111
Gao, B., Wang, Y., Xie, H., Hu, Y., & Hu, Y. (2023). Artificial intelligence in advertising: Advancements, challenges, and ethical considerations in targeting, personalization, content creation, and ad optimization. Sage Open, 13(4), 1-19. https://doi.org/10.1177/21582440231210759
Guendouz, T. (2023). Artificial intelligence-powered customer experience management (moving from mass to hyper-personalization in light of relationship marketing). International Journal for Scientific Research, 3(6), 247-306. https://doi.org/10.59992/IJSR.2024.v3n6p9
Hardcastle, K., Vorster, L., & Brown, D. (2025). Understanding customer responses to AI-Driven personalized journeys: Impacts on the customer experience. Journal of Advertising, 54(2), 176-195. https://doi.org/10.1080/00913367.2025.2460985
Hari, H., Sharma, A., Verma, S., & Chaturvedi, R. (2025). Exploring ethical frontiers of artificial intelligence in marketing. Journal of Responsible Technology, 21(1), 1-18. https://doi.org/10.1016/j.jrt.2024.100103
Hess, N., Kelley, C., Scott, M., Mende, M., & Schumann, J. (2020). Getting personal in public!? How consumers respond to public personalized advertising in retail stores. Journal of Retailing, 96(3), 344-361. https://doi.org/10.1016/j.jretai.2019.11.005
Jain, G., Paul, J., & Shrivastava, A. (2021). Hyper-personalization, co-creation, digital clienteling and transformation. Journal of Business Research, 124(1), 12-23. http://dx.doi.org/10.1016/j.jbusres.2020.11.034
Kalia, P., & Paul, J. (2021). E-service quality and e-retailers: Attribute-based multi-dimensional scaling. Computers in Human Behavior, 115(1), 1-20. https://doi.org/10.1016/j.chb.2020.106608
Kaponis, A., Maragoudakis, M., & Sofianos, K. (2024). Enhancing User Experiences in Digital Marketing Through Machine Learning: Cases, Trends, and Challenges. Computers, 14(6), 1-17. https://doi.org/10.3390/computers14060211
Kouroupis, K., Vagianos, D., & Totka, A. (2022). Artificial Intelligence and Customer Relationship Management. East European Yearbook on Human Rights, 4(1), 5-24. http://dx.doi.org/10.5553/EEYHR/258977642021004001002
Lee, A., & Hancock, J. (2024). Social media mindsets: A new approach to understanding social media use and psychological well-being. Journal of Computer-Mediated Communication, 29(1), 1-48. https://doi.org/10.1093/jcmc/zmad048
Liu, J., Tian, Y., Zhou, Y., Xiao, Y., & Ansari, N. (2020). Privacy preserving distributed data mining based on secure multi-party computation. Computer Communications, 153(1), 208-216. https://doi.org/10.1016/j.comcom.2020.02.014
Liu, K., & Tao, D. (2022). The roles of trust, personalization, loss of privacy, and anthropomorphism in public acceptance of smart healthcare services. Computers in Human Behavior, 127(1), 1-22. https://doi.org/10.1016/j.chb.2021.107026
Malgieri, G. (2023). In/acceptable marketing and consumers’ privacy expectations: Four tests from EU data protection law. Journal of Consumer Marketing, 40(2), 209-223. https://doi.org/10.1108/JCM-03-2021-4571
Mendia, J., & Flores-Cuautle, J. (2022). Toward customer hyper-personalization experience—A data-driven approach. Cogent Business & Management, 9(1), 1-19. https://doi.org/10.1080/23311975.2022.2041384
Micu, A., Capatina, A., Cristea, D., Munteanu, D., Micu, A., & Sarpe, D. (2022). Assessing an on-site customer profiling and hyper-personalization system prototype based on a deep learning approach. Technological Forecasting and Social Change, 174(1), 1-17. https://doi.org/10.1016/j.techfore.2021.121289
Naqvi, M., Hongyu, Z., Naqvi, M., & Kun, L. (2024). Impact of service agents on customer satisfaction and loyalty: Mediating role of Chatbots. Journal of Modelling in Management, 19(2), 470-491. http://dx.doi.org/10.1108/JM2-01-2023-0004
Naz, H., & Kashif, M. (2025). Artificial intelligence and predictive marketing: An ethical framework from managers’ perspective. Spanish Journal of Marketing-ESIC, 29(1), 22-45. https://doi.org/10.1108/SJME-06-2023-0154
Ngoc Thang, N., Anh Tuan, P., & Moses, Y. (2020). Knowledge acquisition, knowledge management strategy and innovation: An empirical study of Vietnamese firms. Cogent Business & Management, 7(1), 1-18. https://doi.org/10.1080/23311975.2020.1786314
Olson, E., Olson, K., Czaplewski, A., & Key, T. (2021). Business strategy and the management of digital marketing. Business Horizons, 64(2), 285-293. https://doi.org/10.1016/j.bushor.2020.12.004
Rosario, A., & Dias, J. (2023). How has data-driven marketing evolved: Challenges and opportunities with emerging technologies. International Journal of Information Management Data Insights, 3(2), 1-19. https://doi.org/10.1016/j.jjimei.2023.100203
Rosenbaum, M., Ramirez, G., Campbell, J., & Klaus, P. (2021). The product is me: Hyper-personalized consumer goods as unconventional luxury. Journal of Business Research, 129(1), 446-454. https://doi.org/10.1016/j.jbusres.2019.05.017
Sameen, T. (2025). The Role of Ethical Marketing Issues in Consumer-Brand Relationships in the Context of Social Media Marketing. European Journal of Business and Management Research, 10(1), 108-117. https://doi.org/10.24018/ejbmr.2025.10.1.2565
Saura, J. R. (2024). Algorithms in digital marketing: Does smart personalization promote a privacy paradox?. FIIB Business Review, 13(5), 499-502. https://doi.org/10.1177/23197145241276898
Saura, J., Ribeiro-Soriano, D., & Palacios-Marqués, D. (2022). Assessing behavioral data science privacy issues in government artificial intelligence deployment. Government Information Quarterly, 39(4), 1-15. https://doi.org/10.1016/j.giq.2022.101679
Saura, J., Škare, V., & Dosen, D. (2024). Is AI-based digital marketing ethical? Assessing a new data privacy paradox. Journal of Innovation & Knowledge, 9(4), 1-23. https://doi.org/10.1016/j.jik.2024.100597
Singh, A., & Kakkar, A. (2025). Marketing in the age of hyper-personalisation: Lessons from analytics-driven brand success. Applied Marketing Analytics, 11(1), 72-86. https://doi.org/10.69554/XQVW6761
Vishwakarma, R., Pandey, A., Kundnani, M., Yadav, A., Singh, M., & Yadav, M. (2025). Personalization vs. Privacy: Marketing Strategies in the Digital Age. Journal of Marketing & Social Research, 2(1), 177-191. https://doi.org/10.1016/j.intmar.2024.03.004
Wong, D., & Floridi, L. (2023). Meta’s oversight board: A review and critical assessment. Minds and Machines, 33(2), 261-284. http://dx.doi.org/10.1007/s11023-022-09613-x
Yildiz, E., Güngör Şen, C., & Işık, E. (2023). A hyper-personalized product recommendation system focused on customer segmentation: An application in the fashion retail industry. Journal of Theoretical and Applied Electronic Commerce Research, 18(1), 571-596. http://dx.doi.org/10.3390/jtaer18010029
Zywiolek, J., Trigo, A., Rosak-Szyrocka, J., & Khan, M. (2022). Security and Privacy of Customer Data as an Element Creating the Image of the Company. Management Systems in Production Engineering, 30(2), 1-22. https://doi.org/10.2478/mspe-2022-0019