Take the 2-minute tour ×
Travel Stack Exchange is a question and answer site for road warriors and seasoned travelers. It's 100% free, no registration required.

I know that review sites like TripAdvisor have fake reviews (some hotels more than others) - how can I make sure I am looking at the 'real' reviews as using TripAdvisor can be very tricky to filter out the good from the bad.

I am going to Egypt - Sharma El Sheikh using TA to find the better hotels.

share|improve this question
    
Given the advantage of TripAdvisor, etc, is the "wisdom of the masses", I've found the overall rating is generally a better guide than individual reviews. –  dlanod May 7 '14 at 11:34
10  
You don't only have to filter out fake reviews anyway, you also have to filter out poor quality reviews that may or may not be fake. "Best hostel I ever stayed at, 5 stars". "Worst hostel of our entire trip. 0 stars". Also filter out things that are due to one-time problems that are likely to have been addressed, broken bathrooms, etc. Then there are differing preferences: "5 stars, party every night" and "5 stars always quiet and peaceful" are not the same thing depending on what you like. –  hippietrail May 9 '14 at 2:57

13 Answers 13

up vote 64 down vote accepted

Here are some of the main signs I would look out for to decide if a review is fake. (note: updated with some points taken from a study on the matter)

  1. Overly hyped. While it is certainly possible that someone had an amazing time, the more hyped the review looks the more likely it is fake.

  2. "Fake reviewers tend to overdo “self-referencing”, that is, they overuse words such as “I”, “me”, “my”, “mime”, as if they try to underline their existence and credibility." Source (note: I changed this point following feedback in the comments)

  3. Hotel jargon. "They have eight spacious twin suites with amazing views". Unlikely a real guest would write that.

  4. Sudden burst of posts. If there aren't many reviews for the property and then all of a sudden there's a bunch of new ones. It looks like someone decided to go bump up their rating.

  5. No background for the reviewer. If it's the only place they reviewed, it takes away from the trust.

  6. Glaringly different from other reviews. Unless there's been a renovation or change of management, generally speaking reviews are relatively consistent.

  7. Truthful reviewers tend to "focus on spatial details (e.g., bathroom, floor, small, location) of their experience, while deceptive reviewers have difficulties in filling in spatial information. As a result, deceptive reviewers will focus on other types of information, such as why they went to Chicago (e.g., vacation, business), or whom they went with (e.g., family, husband)." Source

  8. "Deceptive reviews demonstrate the characteristics of imaginative writing, i.e., frequent usage of verbs and adverbs, while truthful reviews demonstrate the characteristics of informative writing, i.e., frequent usage of nouns and adjective (except superlatives)" Source

In the end, you just need to look at the balance of evidence and decide for yourself if the person is worth trusting or not.

share|improve this answer
19  
I disagree with #2 - when I've left reviews, I've tried to mention staff who were good by name :/ –  Mark Mayo May 7 '14 at 10:47
4  
I torture myself by reading reviews of $2000+/night resorts. At those prices, everything is amazing and the staff are mentioned by name in every review. These are good rules in general though. –  Kate Gregory May 7 '14 at 11:10
2  
3-5 in particular are really good points. I wouldn't have thought of 3. –  starsplusplus May 7 '14 at 12:20
4  
According to an article, which I am unfortunately unable to retrieve, about fake-review detection tools that outperformed humans by far, some of these points, perhaps counter-intuitively, are actually signs of a genuine review. I recall fake reviews to appear more nuanced to boost their credibility (#1) and less likely to use words describing the location (#3). –  Marcks Thomas May 7 '14 at 14:22
4  
#3 is suspicious not just for the jargon. Most customers aren't going to know how many rooms of a particular type their hotel has. –  David Richerby May 8 '14 at 17:11

First, keep in mind that your goal is not to evaluate the truthfulness of individual reviews, it is to find a good place to stay. There are thousands of reviews across hundreds of places. Analyzing individual reviews is a fools errand, unless you actually know the reviewer or noticed the reviewer seems to agree with you in general.

Here is a small description of what I do to decide on a place:

First pass:

0.a. Price range. Obviously, it's got to fit your budget, but also, don't forget that reviews are relative to the expectation of the visitors. You will find a lot of places with fantastic ratings for a thin bunk bed in a dorm room and a dirt cheap price, while a poorly reviewed high end hotel room would exceed the wildest expectations of the backpacker crowd. Limiting your search to your price range will save you a lot of time and prevent a lot of disappointments.

0.b. Location: check on the map what the distances are to the major attractions and make sure you convert these in hours and minutes. Zoom can sometimes trick you. If proximity to a specific place is important to you, go to Google map and get directions. Google's time estimates are usually pretty good and take into account different types of transportation. You don't want to end up with an hour long commute or worse just because the place "looked close enough". That can really ruin your trip.

0.c. Photos analysis: Does the place look like the kind of place you would want to stay at? Hotel photos provide a upper threshold for quality. If the photos look bad to start with, never mind the reviews, it's rarely going to be better than the pictures.

Next, pay attention to the quality of the bedding and bed accessories. The bed & bed linen are often the single most expensive furniture items in a room and prices vary widely. If the cover looks thick and nice, the bed inviting, it usually means that the hotel spend serious money on guest comfort.

Look at the TV. Is it a flat screen or an old monitor? If a flat screen, is it big? I don't care one wit about the TV, I hardly ever watch it, but in this day and age, reasonably sized flat screen TVs are very cheap. Any place still equipped with cathodic tube TV tells you at once that they haven't done any maintenance/investment in a decade.

If the place "looks nice" and still has an old TV, you can pretty much bet your house on the fact that everything inside will be old and poorly maintained. For higher end accommodations, look at the bathtub. Knowing what furniture cost will give you a pretty good idea of the investment in the room decoration.

Photography techniques and lighting can make a dump look good, but they can't turn an old TV into a flat screen, and they have a hard time disguising the thickness of the bed linen.

  1. Amenities required. If you have a car, you need parking; you might want a gym, etc.

The goal of the first pass is to reduce your choices as fast as possible to a few good candidates based on objective criteria. Then you can dig into reviews. Diggin straight into reviews makes it an endless task.

Second pass:

Now that you have a short list of interesting properties, it's time to have a look at the reviews and ratings.

  1. Sanity check: Quickly look at the rating score on a few different sites. Consider especially the rating score of booking websites where the user must have actually booked, paid and stayed at the place to be eligible for posting a review, like Agoda.

If there is a large discrepancy in scores, you know something is up.

Also, get a feel for the scoring system and how people tend to rate on the specific website. For instance, 6/10 on Agoda is rated "acceptable", but 6/10 really means "dreadful" because there is almost no place reviewed below 5.5/10.

  1. Number of reviews: Ignore score and reviews if there are less than 10 reviews unless the place is brand new. It's easy to override a poor review by writing 5-6 glowing fake reviews, but if the place has thousands of reviews, good luck with that.

  2. Browse through the bad reviews and make note of recurring specifics. Discount generic comments. If reviews say that the place is not clean, that's generic, but if several reviews point out that the walls of the shower stall are filthy, chances are they are. Things like "scorched power outlets", "holes in the bed linen"...

If someone genuinely feels like the place is worth the lowest rating he is typically upset about it and will tend to exaggerate. He wants to get revenge and can be sarcastic but he will be specific about what went wrong whereas a competitor posting a bad will tend to stay vague about it.

Also note the date. Often you will see a bunch of reviews complaining about a temporary problem (no WIFI, power outages, etc.) If these complaints stop appearing in more recent reviews, the problem has probably been solved and the real score should be better.

Pay particularly attention to positive specifics in bad reviews. If several bad reviews admit something is good about the place, that bit's probably true.

  1. Browse through good reviews. Discard overly emphatic adjectives. Don't fully discard jargon, but look at it with suspicion (the reason why you don't want to dismiss it point blank is because frequent travelers tend to be fluent in hotel jargon). As for negative reviews, look for specifics. Pay particularly attention to good reviews admitting a problem that you already identified in the bad reviews. Some people have much lower standards/are less critical/nicer than other and will moderate their critics. "The bathroom was not very clean" in a good review will validate a "Bathroom were filthy" comment in the bad reviews.

Don't try to evaluate individual reviews on their own merits. Besides fake reviews, sometimes the reviewer is just an @ssh*le who treated the staff like slaves and expected a palace for the price of a bunk bed "because it's a cheap country". The review is not fake, but the reviewer is full of it. And similarly, you have guys who will leave positive reviews everywhere no matter how dreadful the place because their own standards are abysmally low. I have had genuine recommendations from Thai friends who marveled about places that can only be described as dumps.

Parsing the reviews and making note of specifics will usually give you a fairer picture of the place than an in-depth analysis of a handful of reviews.

share|improve this answer

Spotting fake reviews is not easy. But there are a few things you can look out for:

  1. 1-star and 5-star reviews: these are much more likely to be fake, for obvious reasons. Be especially cautious of 1-star reviews, which are often either fake or written by someone who has had some kind of axe to grind. Someone who has awarded an establishment a more moderate score, like 2, 3 or 4- stars, is probably going to give a more fairly balanced assessment of its pros and cons than someone who is on the extremes.
  2. Reviewer profiles: in general, people who've posted more reviews and have earned reputation on the sites are better reviewers. Not only are they less likely to be posting fake reviews, but they've also been to more restaurants (or hotels) and have a better basis for comparison.
  3. Review text: there are some articles that list so-called "tell tale signs" of fake reviews. Especially watch out for over the top language. I'd argue, though, that the places who are trying to game the system have become increasingly sophisticated and many of these points don't apply any longer.

I've written an article that explores in more detail why the scores on TripAdvisor are often so wrong: http://www.tripexpert.com/articles/limits-of-user-reviews.

If you're fed up with TripAdvisor, check out TripExpert, an alternative that uses only professional reviews from travel guides, magazines and newspapers.

share|improve this answer

It can be very hard to spot fake reviews for hotels on TripAdvisor. However, I just came across fake (duplicate) reviews and then fake reviewer accounts by accident. While reading many reviews for several different hotels in Plitvice Lakes Croatia, one unassuming positive review caught my eye because a detail (free wi-fi) conflicted with the hotel description and other reviews. It was not until I saw the same review, word for word, posted for a different hotel in the same area, and both reviews were posted by different reviewers, that I did a little investigating.

I then viewed the reviewers' accounts for other reviews, and copy and pasted snippets of the various reviews into google. The search results showed the same reviews (almost word for word, with the name of the hotel or restaurant changed of course) posted for various different hotels or restaurants in different cities and by different reviewers. I can't say I am surprised this is happening, but these were not blatantly obvious fake reviews and some of them included what would be considered helpful details (like distance from a major tourist site) and had been voted "helpful" by others and some also had replies from the hotels themselves.

Some of these fake duplicate reviews were also posted by "senior reviewers" but I suppose the fakers are stealing the wording from legitimate reviews. Very disappointing and I would like to know if the hotels are paying for this, because now I am rethinking my stay at one of these hotels that had one fake review!

share|improve this answer

Spotting fake reviews on Trip Advisor or on any review site can be easy or could be very difficult given that it is becoming widespread. The only bonus is that Trip advisor is becoming well known for this behaviour amongst the travelling community and Google etc and is/has been lambasted for such.

I have recently returned from a two week holiday at an abysmal hotel in Tunisia. My wife booked it primarily based on the reviews and rating on Trip Advisor. I don’t tend to read reviews at all so went on my wife’s say so.

Once at the resort and having experienced the place for myself, room, food, location, staff etc and witnessing the lengths that the hotel staff go to for acquiring good reviews I decided to write one myself on my return and post it to Trip Advisor.

I first read the reviews and could clearly see which ones had been written by the staff. The ones that mention the animation team by name and blow too much smoke. The ones that are posted via mobile phone by someone using a made up name who only seems to have visited this one place. The ones that follow a not so good review with 5 or 6 glowing reviews.

I wrote a no holds barred, honest and helpful review with a reading family spending hard earned money in mind an referred to the misinformation being given about this hotel and the tactics employed by the staff who are clearly writing some of those reviews fraudulently by Trip Advisor's own policy. Trip Advisor didn't like my open, honest and frank review and blocked that review due to the fact that I was ‘too’ honest and frank about the hotel staff activities of offering free WI-FI to guests who wrote a glowing review or who spent their time sat by the pool with the young female guests using mobile phones to dictate glowing reviews for that guest. I witnessed and heard this happening and guests where quite open in discussing it too. Totally unethical and totally against Trip Advisors rules. Trip Advisor’s response to me was that they cannot allow a review referring to activities of fraud to be posted!! WHAT IS TRIP ADVISOR FOR??

I have subsequently directed Trip Advisor to a specific review written by a 15 year old girl who we spent two weeks with, who was approached by a member of the animation team from the resort who proceeded to write that particular review for her and post it via mobile phone from the edge of the hotel pool. The IP address would be very clear! I also suggested that trip advisor block IP address from resorts in an effort to prevent this from happening.

Trip Advisor are aware of this specific incident and have so far done absolutely nothing about that review and it is still on the site.

No credibility whatsoever for Trip Advisor, who appear to be more interested in the money it makes from the listings rather than the travellers and guests it purports to help.

My advice……avoid using Trip Advisor, watch them sink and then perhaps find an alternative is the only solution.

I am more than prepared to divulge the resort in question if requested.

share|improve this answer
1  
I know what you mean I have contacted TrustPilot over blatant fake reviews and they simply ignored me even though my evidence was full proof –  Zabs Aug 4 '14 at 16:48

There is only one way to know for sure, and that would be to track the actual author of the review. Other than that, it is hard sometimes to know the difference between a 'lazy' review that does not provide quality information (just opinions) versus a generic/mass generated review that is used to boost the ratings.

Sometimes it is probably better to look at the negative review information as a whole rather than the overall ratings (but make sure you look at the dates), so this way you can have a better idea of what the worst-case-scenario might be and whether you would accept it. Again this is not bullet-proof if people deliberating post bad reviews to sabotage other businesses.

However, if you have checked multiple sources and did everything you can to verify the information, the truth is unlikely to be very far off.

share|improve this answer

How to find useful TA hotel reviews (as opposed to the guesswork in trying to find "fake" reviews):

Useful reviews concentrate on the objective facts. What do you get for your dollar. My review might say, "Interior corridors. Elevator. Coffeemaker, microwave. 52 TV channels + HBO. Slow free wi-fi. Free breakfast: cereals, breads, yogurt, coffee, juice. No peepholes on doors."

Non-useful reviews meander off into subjectivity-land. "The front desk clerks here are very nice. Everyone was very nice." Or they discuss issues that probably won't recur during your stay: "The printer in the business center was out of ink" or "They were out of quarters for the laundry machine."

share|improve this answer
    
Out of ink might be harder to spot as there might not be an easily visible warning of ink running out and also ink can be expensive. OTOH, being out of quarters for the laundry machines reminds me of "Broken windows theory" -- if you don't even care about something as basic as having a spare roll of quarters ($10 for forty quarters) from your bank what else do you not care about? I, as an individual usually take five rolls from my bank for laundry and then I don't need to care for months. –  chx May 9 '14 at 8:22
    
I've mentioned in a TA review that a property's guest-use ink-jet printer was down for a week. And if more than one ice machine's broken, they will hear about it from me. But not every dirty tray table is a sign of poor jet engine maintenance. –  user35648 May 9 '14 at 8:26

I look at reviews that are “middle of the road” e.g. 2 or 3 star. I then choose a hotel when the reviews complain about things I don’t care about.

E.g. a review saying that “the hotel bar closes at 10 pm and does not have any music” is a positive review for me, even if the reviewer gave the hotel 2 stars because of it.

A fake review posted by the hotel will give it a high rating, a fake review posted by anther hotel will talk about the poor quality of what the hotel does have.

You are looking for the hotel that will be great for you, not the hotel with the most positive reviews, each of us has different requirements.

Another issue is to understand what the reviewer expected, if someone books a hotel at the last minute, paying what they would expect to pay for a basic B&B, they are likely to give the hotel a much better review then someone that paid a lot more and is used to staying in top hotels.

Also some Bed and Breakfasts (B&B) get bad reviews as their description on booking sites makes them sound more like hotels, hence getting the wrong type of customer, but if you like B&B's, they can be great for you.

share|improve this answer
10  
+1 for "I then choose a hotel when the reviews complain about things I don’t care about." –  starsplusplus May 7 '14 at 12:24
1  
This is exactly what I do as well, not just with tripadvisor reviews, but with more or less all reviews on Amazon as well. The reviews with two stars are usually the best for me, they are quite critical but probably not fake. If they complain about stuff that I don't care about (and there are some good reviews as well), then I will probably like the thing. –  Leo May 7 '14 at 14:58
1  
Love how on the flip-side a review that says “the hotel bar closes at 10pm and does not have any music” is a negative review for me. :) I really don't mind loud music, late bar, good music! Really you need to read between the lines, essentially every hotel is good or how you make them. –  Dean Meehan May 7 '14 at 15:35
    
@TankorSmash, the intention is that I can trust the positive things they say due to the negatives, the fact that the review will stop people that like load music going there is even better. –  Ian Ringrose May 7 '14 at 19:39

You need to weed out both good and bad fake reviews.

Some valid points were given above. Check out the reviewer's other reviews too. In addition, timing plays an important role. Unevenly spread reviews stink - it's highly unlikely you can have nothing and then 50 bad or good reviews in a week.

Besides fake reviews, you should watch for genuine but outdated ones. The owners and management change, and so does the service.

share|improve this answer

It is not likely that there is a method that work perfectly, but there are steps that you can take that will help. If I find ANY reviews which are obviously faked I would be very unlikely to stay at the hotel. If they are prepared to lie to create a positive image, or need to do so, they are probably suspect in other areas as well.

When there are a range of hotels available in my price range I tend to look first at those with a substantial number of user reviews.

Fake reviews are usually short and stupidly positive. They are unlikely to contain major negative criticisms. They are unlikely to contain "mini stories" as some reviews do.

If reviews have been posted by people who claim to come from other countries I look for language which shows that the writer does not have language skills or styles which would be expected from a foreign writer. This usually is easy to spot and a clear indication.

Fake reviews often but not always come in batches. If reviews go something like
'OK Ok Poor Ok Bad Ok good good good OK Ok poor ...' I would guess that the run of "Good" assessments was possibly fake. They may not be, but then applying other tests helps check this.

If reports are VERY inconsistent between reviews I am suspicious. If some reviews say "dirty bedding, cockroaches, loud music, drugs sold in lounge, no food,grumpy staff, ..." and others say "clean, tidy, quiet, family atmosphere, polite staff ..." then I'd guess that one of the two sets of reports may be fake. Odds are that it's not the negative ones.


You can use a number of booking sites to check reviews, even if you do not book through them. If you have some specific hotels on a short list it does not take long to book them on other sites. If reviews are not consistent between sites, be suspicious. [I find Agoda is good for finding nearby hotels and checking approximate prices - they have a hover over map which shows hotels in an area and the alleged price shows when you hover over the hotel icon. Other sites can make price hard to check. ANY price is only a guide before all extras have been added.

share|improve this answer
3  
I'm suspicious of some of the negative reviews too-- competitors or unreasonable customers trying to blackmail the hotel.. Especially if they mention an alternative place or the person concentrates on a lack of 'apology' for some slight. –  Spehro Pefhany May 7 '14 at 2:21
    
If reviews go something like 'OK Ok Poor Ok Bad Ok good good good OK Ok poor ...' I would guess that the run of "Good" assessments was possibly fake. I'm not sure that's enough to go on unless those "good" reviews all also sound really similar or something. –  starsplusplus May 7 '14 at 12:25
    
Interestingly, "mini stories" are actually a strong indication of a fake review, according to a study (see my answer). People who write them also try to fake authenticity... –  Michael Borgwardt May 7 '14 at 14:22
    
There are more than one kind of fake review though. Place promotes itself. Place attacks competition. One employee takes it upon themself to promote the place without knowledge of bosses. Disgruntled guests make stuff up to boost negative review. Employee asks guests to write reviews and guests write positively even though that wasn't what was asked. I bet there's other kinds I haven't thought of too. –  hippietrail May 9 '14 at 3:07
    
@MichaelBorgwardt: I would qualify that as at least fake reviews written by freelance writers for the purpose have mini stories, but most real reviews are not written by freelance writers. –  hippietrail May 9 '14 at 3:40

I've personally tried to look at the user making the review and checking his previous reviews (see if they sound all abit similar) - also checking out how many contributions they have made to date and see if they have any titles such as 'Senior Contributor' etc..

It can be very time consuming however - it is a shame that there isn't a site called "Video Trip Advisor" where I could watch homemade videos from actual tourists as it would be harder for hotels to include regular 'faked' videos

share|improve this answer

Here's a story about a study done by researchers who hired freelance writers to write fake positive reviews, asked judges to tell them apart from real ones (they couldn't) and then did a statistical analysis. According to them, typical properties of fake reviews are:

  • A direct mention of the place where somebody stayed
  • Lots of adverbs like 'really' and 'very', exclamation marks
  • Excessive use of superlatives and a lack of detail and description.

The above are what one would expect, but then there are also somewhat counterintuitive ones:

  • The heavy use of 'I' and 'we'.
  • A focus on who the person was with, such as a husband or family
  • A narrative account of a holiday
share|improve this answer
3  
The latter three are probably the result of people trying too hard to sound genuine... and from that point of view I find them quite credible. –  keshlam May 8 '14 at 1:59
5  
To explain the counterintuitive ones: "Only lies have detail" (or more specifically, only lies have 'self-justifying' detail about the why and how rather than just giving the what) –  user568458 May 8 '14 at 13:53
1  
These fake reviews were written by freelance writers hired to write fake reviews. I'm sure this is done to some degree. But usually fake reviews are written by owners, managers, or staff of a place who are not professional writers. Or competitors bagging each others' places in fake negative reviews, also not professional writers. I'd like to see some statistical comparisons of these kinds of fake reviews versus those written by freelance writers. –  hippietrail May 9 '14 at 3:02

In my opinion the best filter for fake reviews is training, by writing some real reviews yourself.

I also make a habit of contradicting previous reviews quite explicitly if I disagree. Either way the method is never foolproof. If I went to a hotel with quite some remarkable reviews, only to regret it afterwards, it wouldn't be the first time. Not because the review was fake, but simply because I don't share everybody's preference. For example, I don't like chatty hosts in the morning, so if a review says: "Frank was a really good host, every morning he gave us detailed instructions on ....", I might go to Frank's competitor.

So after writing some real reviews myself I noticed how difficult it actually is to write a decent review, without resorting to generalities. It takes time to write a decent review. Fake reviewers usually don't spend that much time and they don't have a specific detail to mention. If they do, it eventually backfires. Future reviewers could use it to contradict.

So to answer your question, you could spent an hour on writing some reviews on places you visited in the past, after which you look at the reviews for Sharma El Sheikh again.

share|improve this answer
2  
I think many people who write real reviews don't want to spend much time on it either.... –  Michael Borgwardt May 7 '14 at 14:24
1  
I think this could be the most insightful answer. A bit unexpected compared to the others yet when you read it you know it just has to work. –  hippietrail May 9 '14 at 3:04

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.