The rapid advancement of artificial intelligence has significantly transformed educational landscapes, yet many AI-powered educational platforms struggle with converting free users to paid subscribers. This study investigates the expectation-experience gap affecting AIEdu, an AI educational tool by PT. EDTH in Indonesia, which demonstrates high user acquisition (27.99% monthly growth) but low conversion rates (0.41% average). Using a qualitative approach based on the Unified Theory of Acceptance and Use of Technology (UTAUT) framework extended with Perceived Value, Satisfaction with Free Trial, and Price Sensitivity constructs, this research analyzes user expectations and experiences through semi-structured interviews with 12 participants. The study employs thematic analysis supported by NVivo software to identify patterns between market expectations and actual user experiences. Findings reveal significant gaps in five key areas: performance expectancy (surface-level content vs. deep adaptive learning), effort expectancy (onboarding friction vs. seamless access), social influence (peer validation vs. promotional exposure), perceived value (quota limitations vs. academic enhancement), and price sensitivity (energy-based vs. time-based models). The research proposes a strategic implementation roadmap using the Kano Model, prioritizing foundational improvements (content accuracy, technical stability) before enhancement features (premium value proposition, flexible pricing). Results indicate that addressing these expectation-experience gaps through curriculum alignment, simplified onboarding, enhanced premium features, and student-friendly pricing models can significantly improve conversion rates and user engagement in the Indonesian educational technology market.