No abstract

No abstract

Purpose – The aim of this paper is to examine the impact of three main technologies on converting browsers into customers: impact of review rating (location rating and service rating), recommendation and search listings. Design/methodology/approach – This paper estimates conversion rate model parameters using a quasi-likelihood method with the Bernoulli log-likelihood function and parametric regression model based on the beta distribution. Findings – The results show that a high rank in search listings, a high number of recommendations and location rating have a significant and positive impact on conversion rates. However, service rating and star rating do not have a significant effect on conversion rate. Furthermore, room price and hotel size are negatively associated with conversion rate. It was also found that a high rank in search listings, a high number of recommendations and location rating increase online hotel bookings. Furthermore, it was found that a high number of recommendations increase the conversion rate of hotels with low ranks. Practical implications – The findings show that hotels’ location ratings are more important than both star and service ratings for the conversion of visitors into customers. Thus, hotels that are located in convenient locations can charge higher prices. The results may also help entrepreneurs who are planning to open new hotels to forecast the conversion rates and demand for specific locations. It was found that a high number of recommendations help to increase the conversion rate of hotels with low ranks. This result suggests that a high numbers of recommendations mitigate the adverse effect of a low rank in search listings on the conversion rate. Originality/value – This paper contributes to the understanding of the drivers of conversion rates in online channels for the successful implementation of hotel marketing.

The correlated nature of security breach risks, the imperfect ability to prove loss from a breach to an insurer, and the inability of insurers and external agents to observe firms' self-protection efforts have posed significant challenges to cyber security risk management. Our analysis finds that a firm invests less than the social optimal levels in self-protection and in insurance when risks are correlated and the ability to prove loss is imperfect. We find that the appropriate social intervention policy to induce a firm to invest at socially optimal levels depends on whether insurers can verify a firm's self-protection levels. If self-protection of a firm is observable to an insurer so that it can design a contract that is contingent on the self-protection level, then self-protection and insurance behave as complements. In this case, a social planner can induce a firm to choose the socially optimal self-protection and insurance levels by offering a subsidy on self-protection. We also find that providing a subsidy on insurance does not provide a similar inducement to a firm. If self-protection of a firm is not observable to an insurer, then self-protection and insurance behave as substitutes. In this case, a social planner should tax the insurance premium to achieve socially optimal results. The results of our analysis hold regardless of whether the insurance market is perfectly competitive or not, implying that solely reforming the currently imperfect insurance market is insufficient to achieve the efficient outcome in cyber security risk management.

No abstract