The team from Cornell analysed thousands of online reviews of Moscow hotels to try to determine patterns and correlations that could help hoteliers in determining what guests were really saying with their ratings and where to focus their attention. They compared various factors of the reviews, such as length, tone, star rating, number of topics covered and what kinds of topics the reviewers spoke about to create a foundation for the study.
In the study, semantic trends were found in both good and bad reviews. Examining these trends and what they mean can provide hoteliers with insights that will help them to achieve consistently higher ratings in their online reviews.
The Cornell team's analysis of Moscow hotel reviews showed that the quality of the guest's experience, especially in comparison with their expectations, is what will influence the overall review rating of a property one way or another. We set out to see if reviews of London hotels followed the same trends.
We looked at TripAdvisor reviews of six London hotels (two of the top rated, two from among the worst rated, and two medium rated), and found that the reviews of these hotels did indeed follow the same patterns.*
Hoteliers should be focussing on ensuring that the experiential qualities of a guest's stay at their property outshine the transactional aspects.
The fact that guests who have had negative experiences tend to mention price more often, and that when guests who have enjoyed their stays mention price it is in the context of it being “worth it” shows that guests are not so much sensitive to price as to perceived value.
A hotel that provides good perceived value for money by giving guests good experiences can charge more than a hotel with apparently similar features and amenities that does not provide a pleasant experience, and still be perceived as being better value for money.
The results of another Cornell study support these findings, stating that if a hotel can increase its rating by one point on a five-point rating scale such as TripAdvisor’s, it can increase its price by more than 10 percent without losing any occupancy or market share.
While looking at overall ratings is helpful, they do not always tell the full story, and hoteliers need to actually read the content of reviews in order to learn from their valuable insights.
The Cornell team found that negative sentiments tended to have a stronger impact on a guest’s overall rating of the hotel than did positive sentiments, and we also noticed some disconnect between review content and review score. One reviewer gave a hotel only one star, while still stating “This hotel is very clean, lovely comfy pillows & beds” but focussing on the noisy nearby road as being the reason for her poor rating.
“information from the text can potentially yield insights not indicated in the ratings for how hotels can improve their operations and better meet customer expectations” – Cornell University
The fact that negative experiences have more of an influence on guests ratings than do positive ones also suggests that hoteliers should put more effort into ensuring that guests have a consistently acceptable experience, rather than one that is exceptional in many ways, but falls short of the guest's desires in one or two key ways.
Hoteliers can use the length of reviews to quickly scan for negative or positive sentiments.
To find out what you (or your competition) are doing wrong, pay close attention to longer reviews, as these are likely to give you insights into where a hotel is falling short, and are likely to be more detailed and provide more specific information. If you want to know what you are getting right, scan through the shorter reviews.
When analysing your reviews, and those of your competitors, bear in mind what kind of hotel experience you are trying to provide, and focus on the aspects that matter most to your business.
The Cornell study found that guests at different tiers of hotels generally tend to care about different things:
However, unhappy guests tend to focus on the same things, regardless of the type of hotel.
Expectations may matter more than actual experience. One of the London hotels we looked at received a 1-bubble TripAdvisor review, even though the reviewer could find little fault with the staff, service or rooms, simply because he was expecting a bigger hotel.
In our collection of reviews of London hotels, many of the negative reviews mentioned the reviewer being “disappointed”, while the positive reviews, of budget and luxury hotels alike, often mentioned the hotel exceeding expectations.
Hoteliers should put effort into ensuring that guests expectations are, at the very least, met, and should strive to exceed them. This is not only a case of ensuring that guests have a positive experience at your hotel, but also making sure that their expectations are reasonable to start out with.
*We excluded reviews written by people who visited the hotel, but weren’t planning on staying at the hotel (visiting for lunch or tea, for example), and reviews written in difficult-to-understand English.