What should be our focus? Investigating restaurant reviews using zero-shot and text analytics-based important performance analysis

Purpose One of the fundamental challenges in hospitality marketing is decoding the unstructured voice of the customer to predict revisit intention across diverse economic segments. Therefore, understanding how dining attributes shift in importance across price tiers is critical for practitioners to optimize service delivery. This study investigates the determinants of restaurant service quality and their impact on customer satisfaction by analyzing large-scale online reviews. Design/methodology/approach his study analyzed a massive corpus of 300,784 online reviews from top-ranked restaurants across six major Asian culinary destinations employing a high-recall zero-shot classification calibration to capture sparse linguistic signals despite known precision trade-offs. By operating at an unprecedented empirical scale and bypassing the biases of manual surveys, this big-data approach fundamentally shifts the research paradigm beyond traditional DINESERV constraints, empirically validating the chef as an orthogonal loyalty driver. Findings The results show that while food quality remains the universal baseline for satisfaction, the role of service quality is asymmetric, thus acting as a critical failure point for dissatisfied guests, but a hygiene factor for satisfied ones. Furthermore, price and value emerged as the primary “Concentrate Here” priorities for Midrange and Fine Dining segments, thereby indicating a tension between cost and perceived authenticity. Originality/value The findings also validated the chef as a distinct dimension that is separate from general food quality. The results provide a data-driven framework for operators to tailor their resource allocation based on their specific price positioning to enhance customer retention.