Research brief
AI Personalization, Trust and Video-Streaming Experience
Research in Malaysia’s Klang Valley examines how viewers experience AI-based recommendations. Personalization, trust and perceived interactivity help explain satisfaction, with PLS-SEM and Necessary Condition Analysis offering complementary views of the evidence.
- Context
- Klang Valley, Malaysia
- Method
- PLS-SEM and NCA
- Focus
- Personalization and trust
- Online publication
- 11 September 2024
What does the study examine?
The study investigates customer perceptions of AI-based suggestions in video-streaming services. It considers perceived interactivity, service quality, commitment, trust and personalization, together with the role of expectancy confirmation.
The research question concerns the customer experience around recommendations. It does not benchmark the technical accuracy of a particular recommendation algorithm or compare the performance of competing streaming platforms.
Why combine PLS-SEM with NCA?
Partial Least Squares Structural Equation Modelling examines relationships among the constructs and their association with customer satisfaction. Necessary Condition Analysis asks a complementary question about conditions needed to achieve a desired outcome.
A factor that contributes to an outcome is not automatically a necessary condition for that outcome. Using both approaches helps distinguish these ideas when interpreting customer experience.
What did the study find?
Personalization had a significant relationship with satisfaction, with a reported path coefficient of 0.34. The paper also reports that trust strengthens the role of perceived interactivity and that expectancy confirmation moderates relationships between service attributes and satisfaction.
The coefficients describe the statistical model used in the study. They should not be interpreted as percentage increases in revenue, retention or satisfaction, or as guaranteed effects for an individual platform.
What does this mean for AI-enabled services?
The findings suggest that tailored recommendations should be considered alongside the conditions that make people trust and engage with a service. Personalization is part of an experience that also involves service quality, interaction and users’ expectations.
For service teams, the practical question is how users perceive the whole experience around AI suggestions. Decisions about recommendation design should be tested with the relevant audience rather than relying on a single factor in isolation.
How should the evidence be interpreted?
The study draws on customer data from Klang Valley, Malaysia. Interpretation needs to respect that setting and the research design, rather than assuming the same relationships hold for all platforms or markets.
The publication list uses the final citation year of 2025. The publisher’s online publication date is 11 September 2024. The original article contains the detailed measures, analysis and limitations.
Original publication
Impact of AI on Customer Experience in Video Streaming Services: A Focus on Personalization and Trust
Ahmed, S., & Abd Aziz, N. (2025). Impact of AI on customer experience in video streaming services: A focus on personalization and trust. International Journal of Human–Computer Interaction. First published online 11 September 2024.
DOI: 10.1080/10447318.2024.2400395
This page is an original research summary. Access to the full article is governed by the publisher. No full-text paper is hosted here.