Definitely Doug 3/18/22: Put Your Online Reviews to Work!

For better or worse, every hotel is affected by its online reputation. Summary ratings posted on metasearch sites like Tripadvisor and Google often determine which hotels even get consideration at the start of the planning phase. Further down the sales funnel, reviews themselves become important as travelers look to see what others like them enjoyed or disliked about specific hotels. Online reviews are an anomaly in consumer marketing: unlike your internal data, they are available for anybody to see, both for your hotel and for your competitors. You can be certain sure that if you don’t manage your reviews well and your competitors do, it will hurt you.

This week’s column will delve into technology tools that can help hotels manage online reputation more effectively. To be sure, Online Reputation Management (ORM) tools work best in tandem with Guest Satisfaction Surveys (GSS) and key customer systems such as a Customer Relationship Management (CRM) system or Property Management System (PMS). Taken together, these tools can be used proactively over time to deliver a better-rated product, which can attract new guests. Feedback collection can also be enhanced through integration of chatbots, call-center surveys, and other tools, which I will not cover today but will come back to in subsequent weeks.

The goal is to obtain, analyze, and act on guest feedback. The “act” part may be as simple as acknowledging past experiences on a future stay by the same guest or choosing the best email messaging for a particular market segment. Or it can be as significant as deciding whether guest satisfaction scores might improve more if you renovate the pool vs. investing in staff service training. Hotel managers need to focus limited resources on things that will “move the needle,” but that means having the right metrics and understanding how they affect guest satisfaction.

ORM, GSS, and CRM/PMS systems each play often-overlapping roles in this process. This week’s column features a deep dive just on ORM; future installments will cover the role of the others. The main objective of ORM tools is to help maximize the attractiveness of your hotel to prospective customers, especially first-timers. As such, it is a primary marketing tool. The feedback you get in the form of ratings and reviews is not very scientific and probably doesn’t represent your average customer. It is mostly qualitative and will not really tell you where you will get the biggest return on investments in improvements. To be sure, it can help you identify strengths and weaknesses – especially ones you never suspected. But most important, it is the first and possibly the last thing many prospective guests will see about your hotel.

To quantify the value of different potential actions a hotel might take to improve ratings, Guest Satisfaction Surveys are much more useful: they are more representative, you can target them at specific issues, and depending on when you collect them, they may enable service recovery while the guest is still in-house, avoiding a bad review. And to act on specific customer concerns voiced in online reviews or surveys, or to market to customers in a more personalized way, you will need to bring at least some of the data from ORM and GSS tools back into your CRM or PMS. A full solution should involve these three sets of tools.

In researching this topic, I spoke with several companies that offer ORM, GSS, and/or CRM tools (some focus on just one, others have two or all three). There were many commonalities but also many differences in the solutions. ORM is part science and part art, so differences are to be expected. But I learned a great deal in conversations with several providers across this spectrum, including dailypoint, GuestRevu, Medallia, RepUp (part of Xperium), ReviewPro (part of Shiji Group), Revinate, and Sentinel (part of D-EDGE); I thank them for their time and thoughtful contributions. TrustYou also plays in the ORM space but did not respond to multiple requests for an interview.

About Online Reviews

First, for readers whose familiarity with ORM may be limited, a few basics may help to set the stage. The experts I spoke with agreed that online reviews are mostly about acquiring customers; surveys are about keeping them and tuning your operation based on feedback.

The concern is often raised that many online reviews can be fake. This is true to a degree, notably on Google and Tripadvisor, but not really the point. While the reviewer and supposed stay may sometimes be fake, the review itself is always real – it has been posted online where your potential customers can see it. It is therefore critically important to take all online reviews seriously and respond to them as if they were legitimate (of course, you can also report fake reviews to most sites and hope they will be taken down). Also, more and more reviews today represent verifiable stays, posted either on an OTA or brand site where they were booked, or as the by-product of a post-stay survey sent by the hotel.

The consensus of the experts is that responding to reviews is critical; this is especially true for negative reviews. Several of the companies cited a Forrester/Tripadvisor study that found that 79% of readers were reassured by a management response on a poor review, and 68% said it would sway them from one property to another. While this study was done ten years ago, there is little doubt that it is still directionally accurate.

ORM tools are designed to let hotels respond to reviews as quickly and effectively as possible. They can also help you identify which aspects of your hotel are working perfectly and should be emphasized in marketing, vs. which areas need improvement. This can help you decide where to spend your next investment dollars: renovating specific facilities, providing better training to certain staff, even just stocking more towels at the pool. They can also provide powerful data that can be used in marketing to corporate and group accounts, showing how much better your hotel performs on things that matter to them vs. your competition.

Why pay attention to online reviews? A study that STR and Cornell conducted, based on the ReviewPro Global Review Index (an overall review score from one large vendor) found that a one-point improvement in a hotel’s index score (on a scale of 100) correlated to a 1.42% increase in RevPAR. And key techniques supported by ORM tools have been shown to raise such scores by as much as 3-4% - even without any upgrades to the hotel’s facility, staff training, or hard product. Add the right investments in those and good execution in a previously underperforming hotel, and review score improvements of even 10-20% can be achieved, which could mean RevPAR increases of 20% or more.

What to Look for in an ORM Solution

Avoid the temptation to do a Google search for Reputation Management Software. Most of what you will find will be generic tools, appropriate for a dentist’s office or nail salon, but not for hotels or restaurants. While most businesses can be effectively rated by choosing some number of stars and leaving a brief comment, hotel review sites understand that why and with whom you are traveling with can be important to readers. They also ask users to rate key aspects of their stay that are relevant only to overnight accommodations. Tripadvisor, for example, asks reviewers to rate location, cleanliness, service, and value; Booking.com asks for ratings for these as well as for staff, comfort, and Wi-Fi. You need software that can handle these aspects. Rather than Google, your first stop should be a site like HotelTechReport or HotelHero, which can point you to solutions that are relevant for hotels.

Coverage: The first thing to consider is which review sites you need to cover. Most hotels in North America will want to cover at least Google, Tripadvisor, Facebook, Booking.com, and Expedia; hotels elsewhere will often want more regional sites as well. Many of the available packages support dozens of sites, but you should consider which ones matter in the source markets that feed your hotel. Many smaller sites and even some larger ones like Ctrip, Agoda, HolidayCheck, or Wotif are heavily used in specific countries or languages, and if these are important to you, you will want to make sure they are included.

Analytics: Most ORMs work by collecting whatever demographic information and summary ratings the sites collect, as well as written reviews. The reviews are fed into proprietary Natural Language Processing (NLP) algorithms that look for keywords associated with hotel experiences, such as “room,” “staff,” “cleanliness,” “breakfast,” or “bed bugs.” They also search for adjectives and adverbs that indicate like or dislike, such as “friendly,” “delicious,” “filthy,” or “gross.” Each review is then tagged based on the keywords and associated sentiment, which can be positive, negative, or neutral. Some ORMs also try to detect the importance of each detected sentiment: there is no reason to get too excited about a very good or very bad rating if it relates to something that the guest found unimportant. Many ORMs look for “emotion” words like “loved” or “hated” to help assess importance.

NLP is far from perfect, and not all NLP algorithms are equal. One solution provider told me that 85% accuracy in scoring was about the best to hope for, meaning that 15% of reviews may be improperly classified. A great test is to identify some actual reviews that have significant commentary, score them yourself manually, and then see how each of the solutions you are considering scores them using its NLP algorithms. While most of them will do a fair job with reviews written in flawless English, reviews are often written more casually and may use slang, poor grammar, have spelling or punctuation errors, or other issues that can confuse NLP. An important feature is the ability for your staff to fix misclassifications, so they do not distort your results.

Languages: One differentiator was how the different products handled languages. Some handled multiple (5-20) languages natively, meaning that they did NLP analysis in that language, using the words and grammatical constructs of the language directly to score the review. Others translate some or all languages to English using a tool like Google Translate, and then apply the NLP analysis to the English. The first option will generally work better because as we all know, much is in any translation (and especially machine translation). In particular, the “tone” of a review is often buried in nuance and may not come across in a machine translation. However, the tools that analyze reviews natively in multiple languages are likely to be more expensive. This may be worth it if your hotel gets a lot of reviews in different languages, or overkill if it gets only a few.

Whenever possible, you should respond to a review in the language in which it was written. If your hotel regularly gets reviews in languages in which only a few of your staff are fluent, you may assign specific colleagues to responding to reviews in their languages. If this is the case, you will want to look for a solution that can automatically route reviews in certain languages to the appropriate person. Most solutions also support machine translation for cases where no one speaks the reviewer’s language; the review can be translated to the responder’s language, and their response translated back to the reviewer’s.

Responding: You want a tool that can make it as time efficient as possible to respond to every review quickly and meaningfully. The level of integration a product has with different review sites will make this easier or harder. “Direct Reply” is the ability to enter your reply to a review on the ORM dashboard and have it posted to the site automatically. Short of that, you may be able to click to link to the site to post the response, which can take longer and may require a login. In still others, it may be necessary to open the review site, navigate manually to the review, and then enter a response. These differences can add up to a lot of staff time for a hotel with many reviews.

Most of the products I saw offered templates for responding, but none of them recommended using the template without some manual personalization. Multiple templates can be used for specific common issues or as starting points for responding to positive reviews, but in each case best practice is to personalize the response at least a little, and more if specific negative issues are raised. Depending on your clientele, you may need templates in multiple languages. In any case, it should be simple to select a template, personalize it, and post it. If your ORM is connected to your PMS or CRM and can identify the specific guest, then template should ensure that only the reviewer’s username on the review site, and not their real name, is inserted into the response – it’s can be an easy mistake to make to use the guest’s real name if it’s in front of you, but it’s a serious privacy faux pas.

The better systems have capabilities like alerting the hotel when there is a review awaiting response (particularly useful for designated responders for specific languages, who may not have reason to look every day), for measuring the average response time, and for escalating reviews that have not been responded to within a certain amount of time.

Depending on how the response dashboard is set up and how your hotel manages responses, it may be useful to filter reviews that require responses (omitting those that have already been responded to or where no response is permitted), or to filter on overall or specific ratings, for example to prioritize responses to bad reviews when there is not enough time to respond to them all, or to enable similar templated responses to reviews that all raise the same basic issue.

Part of the response process should be forwarding specific issues to the appropriate department, either for response, for corrective action, or to pass along praise for a particular colleague. Depending on the size of your operation and how it is organized, features such as the ability to request a colleague to take ownership of a response, or just to be notified of it, can be helpful. Some systems will also support opening tickets for maintenance and similar items, either through a process within the software or via an interface to a work-order management system.

Competitors: Most ORM systems allow you to select some number of competitors and see the same information as for your own hotel. This can help you understand your relative strengths and weaknesses, see trends of improvement or deterioration over time, and provide ammunition for sales to use with volume accounts. Common reports will show how you and your competitors rank on different factors, where your hotel ranks within all of the hotels in your region or brand, or even (at the brand level) how your brand ranks against other brands.

Analytics. Knowing how you are performing on different metrics is one thing, but in order to make the best use of ORM, you need to be able to ask detailed questions, both about your own performance and that of your competitors. If, for example, a significant portion of your business consists of German tourists, you should be able to filter and see how they rate different aspects of their experience at your hotel, and how that compares to what they say about your competitors. You may want to filter not only based on the reviewer’s country, but on language, the review site, the sentiment (positive, negative, neutral), the trip purpose, the size and nature of the traveling party, the dates, reviews that mention specific keywords, and other factors. It can also be useful to filter based on whether a review was organic (posted directly on a site by a logged-in user) or the result of a review collection process (often embedded in a post-stay survey), because the former will generally be unverified. If your ORM can be integrated with your PMS or CRM, it may be possible to filter results down to the level of specific corporate accounts or groups. This information, particularly when combined with similar ratings for competitors, can be a powerful aid to sales staff trying to close the next contract.

Images. An emerging phenomenon is the capture of user-generated photos that have been posted with reviews. The pictures guests snap and post online are probably what got their attention, whether a feature like the spa or a drawback like mold in the bathroom. Used in conjunction with sentiment analysis and NLP, they can help identify what the hotel should be marketing more heavily (and to whom), and what deficiencies need attention. One vendor I spoke to expected to have photos from Tripadvisor integrated soon.

Online Training. ORM is not particularly complex, but there are a lot of “dos” and “don’ts” that have been identified over the years as best practices. For all but the smallest hotels, you will want to get multiple people in your operation looking at and responding to guest feedback, but staff turnover can create challenges in managing these processes consistently. Look carefully at the training and reference materials the vendor offers. Can someone dive in and be productive with the toolset in an hour or so, and is there good guidance for responding to reviews that they can read before they get started and to help them improve over time?

A key goal should be to teach the staff how to turn guests into promoters. While this isn’t difficult, it does require training. When a guest compliments the hotel to a staff member, the staff member should know what actions to take to maximize the likelihood the guest will post a positive review online.

Licensing. The more of your staff that has access to an ORM system, the easier it will be to close the feedback loop from guests to management to front line. Some providers offer only limited licenses, while others provide access to as many staff as you like. Make sure you have enough seat licenses to use the product effectively. Positive guest feedback is a great motivator for staff, but only if they see it!

GSS, CRM, and PMS connectivity. I’ll cover this in more depth in subsequent columns, but connectivity matters. The GSS should capture survey data from current or recent guests but should also give guests the option to leave reviews online; more reviews alone will boost your review scores and rankings. This means knowing who checked out when, who completed the survey, what it said, what the narrative text said, and what the ORM’s sentiment analysis said about it. Feedback should be captured in the CRM or PMS for future use, and “hooks” established to make sure that it is in fact used in future marketing campaigns or subsequent guest visits.

Reporting. Different hotels have different reporting needs, so consider how easy it is to create custom or semi-custom reports, and whether the vendor will do custom work for you if needed (for a fee). Whatever reports you use, look for the ability to retrieve them on demand but also to schedule regular delivery by email to individual people or groups. Each user and department should also be able to save any reports they create for future use.

Operational Metrics. Key performance indicators that an ORM system should be able to track include review volume, the percentage of reviews during a given period that have management responses, and the average time to response. Review volume can be heavily impacted by the hotel’s practices, especially around guest surveys (which I will cover in the next installment). Volume matters for ranking (independent of star ratings), so it’s important to measure your effectiveness in generating more of them.

Vendor Considerations. Aside from the product, there are other aspects of the vendor company that merit close review. Many of these are no different than for other products, but several of the companies I spoke with talked about the speed of change in the online reputation space as consumers in different parts of the world move from one preferred review site to another, and as the review sites themselves change. Does the vendor give you an account manager and periodic consultations that can help you identify and react to these changes? The more customers, markets, and review sources a vendor works with, the better intelligence they have access to, but it’s not enough for them to have it, they need to have a process in place to transfer it to you.

Conclusion

ORM is just one aspect of collecting and analyzing guest feedback. It has its limitations but is very important in its ability to help attract new customers and to compare your performance to competitors. My next column will focus on other methods of feedback collection that can give more actionable insights for using guest feedback to drive investment and operational decisions.

‍

‍

‍

‍