Cornell study analyzes TripAdvisor reviews to help hotel managers

Online reviews of hotels don’t usually make for scintillating reading. But with the power to make or break a hotel’s profitability, they are read closely.

So hotel managers may be curious to know about how university researchers are shedding new light on how guest reviews should be interpreted.

A new study, published by Cornell University’s School of Hotel Administration’s Center for Hospitality Research, used a bespoke version of a software technique called text-mining to pore over details of how guests evaluate a hotel.

The researchers offer fresh perspectives into how managers should interpret, and make use of, reviews. They also suggest specific practical actions managers could take to boost their property’s online reputation.

Their method helped divine subtle meanings in the phrases of reviews not previously available.

One key conclusion: The most informative reviews to watch out for are the lengthy ones that “focus tightly on just a few issues.” The details in these reviews are generally more accurate than the numerical ratings the reviewer gives. These reviews also tend to be the most useful ones to act on.

“The median isn’t the message”

The researchers looked at a sample of 5,830 English-language reviews from TripAdvisor — which provided data for the study — covering 57 hotels in Moscow, Russia.

They separated out many different strands of cause and effect to the relationships of 18,106 distinct terms relating to five specific attributes: amenities, experience, location, transactions, and value.

The researchers conclude that “the reviews’ content can vary substantially (in sentiment, quality of writing, and themes) from the numerical satisfaction ratings assigned by the review writers.”

In other words, a review may have more damning information than the rating suggests — or vice versa — because consumers sometimes have trouble assigning a number or star-rating that fully captures their sentiment. Hotels need to look for insights not indicated in the ratings to learn how to improve their operations.

Another challenge for hotel managers is that not all reviews can be interpreted in the same way. The authors say that reviews for different hotel tiers gave varying weights to those attributes.

“For instance, the guest’s experience was mentioned more commonly in reviews of high-tier hotels, while amenities and location came up more frequently for motels in the middle tier compared to hotels in other tiers….”

“One particularly noticeable feature of the reviews is that ratings sank when guests wrote lengthy reviews that focused tightly on a limited number of hotel attributes, while relatively briefer reviews that took a wider view of the hotel generally had higher ratings….”

The study also found that guests write more about value and transactions when they are dissatisfied.

“For the highest-rated hotels, 70% of reviews discussed the guest’s experience. In contrast, experience was mentioned in only 32% of the reviews in the lower-ranked hotels and in 45 % of reviews in the middle tier….”

“We also show that, on average, reviews with higher numerical ratings tend to be shorter and discuss topics more broadly, whereas reviews with lower ratings tend to be longer and focus on a smaller number of major issues….”

“Negative comments carry more weight in a guest’s rating than positive ones. That unevenness means a simple average of positive and negative scores may not provide a clear view of a guest’s opinion of the hotel….”

So what to make of this research?

The authors say that, as an overall strategy, “it is better for hotels to provide guests with a moderately good overall experience than a hotel stay that is extremely good in some regards and terrible in others.”

When looking at reviews, it’s especially useful to note the types of customers who stay at their hotel and note what each wants from a hotel stay, the study’s authors say.

Managers should elicit reviews from a larger portion of customers to get a more balanced perspective. They should, to state the obvious, act on the negative feedback to improve.

But just as importantly, they should take advantage of positive feedback to better market themselves. If they receive a positive review from a type of traveler who particularly tends to gravitate to their hotel, managers might consider prominently marketing that positive feedback so that similar types of travelers also see it.

Some private companies already offer software to help hoteliers extract information from the nuanced language of such reviews. These companies increasingly try to offer “sentiment analysis” to boil down the wordy reviews into themes that managers can act on.

The researchers were optimistic about the value of these companies — having interviewed executives at several of them (ReviewPro, Fishbowl, and Brandify, etc.) for their project.

The researchers endorse the use of software to analyze reviews. They write:

“Given that it is now possible to easily partition the positive and negative comments in a review, hotel managers should be appropriately prepared to make operational and strategic changes in response to both positive and negative content.”

The study, “What Guests Really Think of Your Hotel: Text Analytics of Online Customer Reviews,” was written by Hyun Jeong “Spring” Han of the National University Higher School of Economics in Moscow; Mankad; Nagesh Gavirneni, associate professor of operations, technology and information management at Johnson; Shawn Mankad, assistant professor of operations, technology and information management at Cornell’s Samuel Curtis Johnson Graduate School of Management; and Rohit Verma, the Singapore Tourism Board Distinguished Professor in Asian Hospitality Management at Cornell’s School of Hotel Administration (SHA).

Share on FacebookTweet about this on TwitterShare on LinkedInEmail to someone
Sean O'Neill

About the Writer :: Sean O'Neill

Sean O’Neill had roles as a reporter and editor-in-chief at Tnooz between July 2012 and January 2017.



Your email address will not be published. Required fields are marked *

  1. Štěpán Chalupa

    Sean, great article and an even better study (again from Cornell University). Reviews contain huge volumes of important insights and data hotels should use. In my point of view only text-mining approach is not the best. For getting more valuable and precise information it is much better to connect the outputs with other data in reviews (country, age, etc.) and focuse more on data-mining. Reviews can be great communication tool for every hotel becasue everybody can react. Based on your reaction you can influence your customers a lot according to PhoCusWright studies for TripAdvisor! We are curretly working on similar project, that is focused on data-mining approach with usage of text-mining methods.


Newsletter Subscription

Please subscribe now to Tnooz’s FREE daily newsletter.

This lively package of news and information from Tnooz’s web site provides a convenient digest of what’s happening in technology that drives the global travel, tourism and hospitality market.

  • Cancel