It is interesting to observe that when it comes to online reviews, we use the terms "hotels" and "hospitality" in a generic way when we know they include a wide range of properties, from economy to luxury hotels. It is as if we consider that consumers' expectations and behaviors are similar across industry segments when it comes to online reviews.

This generic approach prevents us from identifying the key aspects of the relationship between online reviews and hotel sales. In our study, "The differential effects of the quality and quantity of online reviews on hotel room sales" published in the November issue of the Cornell Hospitality Quarterly, we examine the effects of quality (i.e. online rating) and quantity (i.e. number of reviews) of online reviews on RevPar across segments. We investigated the relationships between user-generated online reviews and hotel sales performance on a sample of 319 hotels in the London market and uncovered the following four main insights:

1. Higher online ratings are associated with higher RevPar

In line with the general perception and sentiment, our results show a positive correlation between quality of online ratings (i.e. the rating left by a reviewer) and RevPar. A more developed model also confirms that rating is a significant variable when it comes to predicting RevPar. Simply, when examining all 319 hotels without distinguishing the scale segment, the higher the quality of the review (i.e. the rate given by a guest) the higher the RevPar in the hotels we examined.

2. More reviews do not necessarily come with higher RevPar

Our analysis disconfirms the general belief that the higher the number of reviews the better. We were surprised to observe that there is no correlation between quantity (number of reviews) and RevPar. It turns out that the effects of quantity and quality are a little more complicated. While quantity and quality are highly correlated (i.e. the higher the number of reviews, the greater the ratings) when we examined all 319 hotels as one type, their effect differs when it comes RevPar. The quality of the review enhances sales performance, whereas the number has no effect when we examined all 319 hotels without distinction of industry scale segment.

These were the findings when we considered all hotels together as one product. The results are different for the second stage of the study, when we distinguished between economy, midscale, upscale, and luxury properties.

The effects of online reviews on hotel sales are not the same across all categories of hotels.

In particular:

3. Online ratings have a minimal impact on RevPar for economy hotels

Higher quality of ratings is critical for luxury hotels, for which any increase in the online rating left by a guest can have an important impact on its RevPar. This statement also holds true for upper scale properties, but the effect is weaker in this segment. Our results reveal that the effect of online ratings in luxury hotels is more than double the effect of upscale hotels. By contrast, when it comes to economy and midscale hotels, the effect of an increase in the ratings on sales performance is very weak, almost minimal.

4. More reviews do not necessarily come with higher RevPar… in the luxury segment

The relationship that we first observed between quantity and RevPar (insight #2) changed as we distinguished between industry segments. Our findings show that for midscale and economy hotels, the relationship between quantity and RevPar existed and is positive. It is however negative for upscale and luxury hotels. The effect on luxury hotels is over ten times the effect for upscale hotels. Simply, more reviews are a booster of RevPar for economy and midscale hotels, but have a negative effect on RevPar for upscale and luxury ones in particular.

Michael C. Sturman (Ph.D., Cornell University) is a Professor of Management, the Kenneth and Marjorie Blanchard Professor of Human Resources, and the Associate Dean for Faculty Development at Cornell University’s School of Hotel Administration. His current research focuses on the prediction of individual job performance over time, the influence of compensation systems, and research methods.