Branding and brand management have been top management priorities in the hotel industry. Some researchers have concluded that strong branding would be an efficient way for hotels and hotel chains to differentiate themselves from each other. Recent studies have focused on the establishment of a brand equity model and the relevant causal relationships of the model. Most of these studies have used types of desirability scales examining the importance of individual factors in measuring brand equity. However, they ignore the trade-offs that affect and characterize choice. Particularly, the personal decision process implied by the hierarchical brand equity model is absent. This study proposed two alternative measures of brand equity, analytic hierarchy process (AHP) and conjoint analysis (CA), to address these limitations. The AHP and the CA were compared using several validity measures to aid in selecting efficient methods. This study examined the validity of AHP and CA under two data collection methods applied to hotel branding: paper-based survey and online survey. Result showed that the AHP data collection methods were easier, as well as with respect to saving time and costs. Results also indicated that the AHP is equivalent to the CA with respect to predictive accuracy. Practical differences for hotel branding in attribute preferences were clearly observed between the AHP and the CA. The AHP results were consistent with previous studies by awarding high importance to perceived quality and brand loyalty and lower importance to brand awareness and brand image. Managerial implications were provided for results. In terms of practicality in data collection, the study results revealed that the data gathered online leads to a slightly lower internal and predictive validity. A limitation of this study was that the two methods were not perfectly comparable. Nevertheless, the validity of both AHP and CA seems satisfactory for both methods. The study results also offer useful perspectives to consider when choosing between the two methods, as well as between AHP and CA.