Over 90 million travel reviews tested – 60 percent found to be positive

User review giant TripAdvisor has always said the majority of reviews on the site are positive – but a massive piece of data mining has also found independent proof.

green tick

Spain-based ReviewPro decided to look at every one of the 90 million reviews it has on its system to establish whether travel-related reviews left all over the web are generally positive, negative or somewhere down the middle.

The reviews, including those in ten different languages from around the world, came in as follows:

  • Positive – 60%
  • Neutral – 28%
  • Negative – 12%

The data analysed came from 65 review sites from across the web, including the omnipresent TripAdvisor and a number of large online travel agency sites.

The results will no doubt please TripAdvisor, which since its launch has continually faced questions from hoteliers and critics that it is just mouthpiece for users with poor experiences of travel accommodation.

ReviewPro says the data illustrates that online reputation management “is not only about protecting against negative feedback”.

“The need to monitor the social web for damaging content remains, but ReviewPro’s research proves most people go online to share positive experiences. An opportunity in reputation monitoring involves collecting positive feedback and using it identify areas of competitive advantage that can be emphasized in advertising and marketing communications.”

The company says the past few years there has been a general move towards using review sites to simply share experiences rather than places to rant when something goes wrong.

The motivation for hoteliers now, says ReviewPro, is to ensure they identify and encourage those that leave positive reviews about them to become brand advocates in their own right.

Share on FacebookTweet about this on TwitterShare on LinkedInEmail to someone
Kevin May

About the Writer :: Kevin May

Kevin May was a co-founder and member of the editorial team from September 2009 to June 2017.



Your email address will not be published.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

  1. Over 90 million travel reviews tested – 60 percent found to be … | Travels in VietNam

    […] the rest here: Over 90 million travel reviews tested – 60 percent found to be … Author: admin Time: Friday, July 15th, 2011 at 12:43 pm Category: wiki travel vietnam […]

  2. Robert Gilmour

    I just read that Trip Advisor has reached 50 million reviews, and boasts its ‘community of 20 million travellers’ worldwide. So I really think this notion of needing to dispel the perception that most reviews are negative, is really a bit of a red herring, and to be positive and constructive in this sphere we need to ignore it and move on.

    My problem, and hotels’ problem with Trip Advisor is not bad reviews, it is inaccurate and fake reviews, and the fact that the reviews can be very demographically unreliable, and review independently multi-graded properties on one level playing field, which I think is conceptually wrong – it is much easier to get a good review for a low grade budget hotel than for a 5 star spa hotel, as the variables – and hence the things that can go BOTH RIGHT AND WRONG are much greater.

    I have 125 hotel clients and THEY ALL buy the idea of reputation management online and accept constructive reviews whether good or bad, not inaccurate or rogue ones. I’m afraid i can’t count on one hand reviews published, which are clearly from people who have never stayed at the hotel, or staff who have left an rubbished them &c &c

    I reiterate that you can’t use the slogan – ‘get the truth then go’ to apply to a source which can’t validate that all reviews are genuine. this will always be the Achilles Heel for TA and will not, and should not – go away until they can devise a validation process for all reviews published that the guest actually stayed at the hotel.

    Many of my clients are using on line guest surveys, where they get much more specific, and usually more reliable, assessments of all the key services and amenities at the property, and naturally – eliminate the rogue review completely.

    • Kevin May

      Kevin May

      @robert – i agree.

      But for all the very valid criticism from the industry and digerati about TripAdvisor, the company’s growth shows that consumers either do not care or, at least, do not feel it has an Achilles Heel at all.

      They trust the wisdom of the crowds aspect of it – a handful of rogue reviews in the midst 500 other reviews stands out, especially when aggregated scores are taken into account.

      • Robert Gilmour


        ‘or they don’t know of/aren’t aware of its weaknesses – ‘get the truth then go’ – hmmmm – where’s the IAB here? I’ve been pulled up in the past for total trivialities compared to this.

        Look they are a multi billion pound company – is it inconceivable, or unreasonable to ask – that they have/develop some system which validates the reviews are actually stayers – or PLEASE GIVE A CAVEAT at least

        Remember too that my ear on the ground tells me that actually most rogue reviews are GOOD reviews trying to boost the profile of the property hence rather contradicting the perception of negative, but still rogue nontheless

  3. Lee R

    Re Giuliano – on our OTA site, a guest has to have stayed to leave a guest review. We used slightly different thresholds than those Joshia used when we conducted similar research, but the outcome was very similar.

    It’s good that hoteliers can see that so few are negative. We get comments from owners who prefer that we don’t show negative reviews at all, and don’t understand the benefits of transparency. We seem to have a long way to go still in terms of changing opinions on this, a very emotive subject within the industry.

    • Robert Gilmour

      The accuracy/reliability of a single specific review is as important, in fact, far more important than the generality. It is of no/little interest to me as a travel shopper whether 60% of reviews are good, what interests me is the legitimacy and accuracy of the review/s for the property/ies i’m checking out.

      What would also interest me as a travel shopper, is a transparent caveat by Trip Advisor that not all reviews may be legitimate, rather than just be reailroaded with the misleading statements, such as get the TRUTH, then go!

      Joe Public needs to know the limitations of Trip Advisor reviews, just as he needs to know that all booking.com reviews can only come from people who have stayed at the property. The latter needs to be the norm, the minimum standard, and needs to be clearly stated as such. Any who don’t conform to that norm need to give the potential user a caveat. I for one trust Booking.com reviews far far more than I do Trip Advisor ones, and not surprisingly, so do most hoteliers.

      Trip Advisor is just an Expedia product > leopards don’t change their spots – so frankly I dislike them because I utterly detest Expedia and many of its practices! Love it or hate it, I have used it, I have to admit!!

  4. Best of Tnooz This Week - Language, Intelligence, Flash, Positivity | Tnooz

    […] Over 90 million travel reviews tested – 60 percent found to be positive […]

  5. Robert Gilmour

    I agree with David. Also its amazing how so many of these posts and articles end up with a semantics discussion. Perhaps it would be useful if the terms of reference/definitions of terms were made clearer at the outset. One of my concerns with TA is the real lack of supporting QA and demographics info to really put the review into reliable context/. for example if a 5 star hotel gets a score of 80%, surely without further info that can’t necessarily be said to be positive. Likewise the 2 star hotel that gets 90%. By the same token, if a C or D demographic customer gives a 5 star hotel 60%, how reliable is the perception of quality of that customer? Furthermore as we don’t really have reliable data on the number of so called ‘fake reviews’ (good or bad) to factor in, where is all this discussion actually taking us?

  6. Alex Kremer

    @Josiah: Did you primarily use sentiment analysis or depend on things like star ratings for indications on whether the review was pos/neg/neutral?

  7. Professor Sabena

    Knowing that there is a universal definition of the term “positive or negative” doesn’t make the reviews conform to a common standard. One has to wonder if the “wisdom of the crowd” applies here. I am sure that the team did their best to do this. Thus we have to take these results at something of face value.

    I would like to make the point about the amount of time a negative review exists. For many larger (or more focused) properties who are actively managing their brand reputation via tools or services – the amount of time that a negative review is out there is going to be shorter than those who either through ignorance or lack of capability do not actively manage their reviews. This would tend to provide a bias more positive than negative as the negatives have a shorter lifespan.

    @Giuliano I think the answer is that the authenticity works both ends of the spectrum good and bad. as such it will tend to cancel themselves out in aggregate but clearly that doesnt help an individual site/property.

    Sadly we know (just like SEO) that all the systems of reviewing have both been gamed and need to be gamed by properties and sites to enable a level playing field – or at least some semblance of that.

    In my humble opinion this will favour the larger or actively managed properties which in turn will likely raise the proportion of positive reviews and lower the number of negative reviews in a purist metric sense.

    Overall – I think the old adage applies.

    Garbage in, Garbage out.


  8. Giuliano

    How many of these reviews are authentic?
    As far as i know there is no verification.
    Reviews have become so important that hoteliers invest in false bookings to write there own “positive” reviews.

  9. Hans van Pruissen

    I am a travel agent, and i use TA with my clients and write my own reviews there as well.
    TA shows me that there are a LOT of hotels that need serious improvement of what they offer.
    I will never advise a hotel with a rating below 3.
    WQhat i do not understand is why so little hotels use TA to monitor their quality. Free customer research, not to be found elsewhere.
    OK, i admit, sometimes responses are over the top.
    Good example is the Carter hotel in New York; each year voted as the most disgusting hotel.
    And still, touroperators offer this hotel (because of its rates) and still customers go for the low rate and expect clean rooms.
    I mean, wake up people. Wake up, manager of the Carter hotel.
    We live in a digital world with easy ways to find out about these things.
    I applaud to every research project that monitors what happens on websites like TA because it shows the industry that there are not only complainers there, it’s the other way around, they are the minority.

  10. David Wood

    This survey shows a great potential for improvement – a Sigma score of 2.68 which is well below the all industry norm.
    It would be interesting to divide the results into the areas of the physical and service reviews. Finding out if there is a difference between positive, nutural and negative reviews on the basis of facilities and also service as two different areas of investigation.

  11. Josiah Mackenzie

    I agree with you, Guillaume, and this data clearly shows the hotel industry has a lot of room for improvement.

    Neutral satisfaction scores don’t necessarily mean the guest doesn’t care – it could just mean they rated one part of their stay 5/5 and another part 1/5….so it averages each part out.

    Cases like this are where I see a big opportunity to move beyond quantitative analysis of reviews and get into qualitative analysis. Semantic analysis is playing a big role in helping hotel managers get more specific in their efforts to improve quality.

    • Martin Soler

      While there is a lot to improve let’s take into account that these are online reviews published for others to see, not surveys done randomly to guests for internal use. Which changes the picture quite a bit. Other industries monitor this with a survey for internal use which is anonymous.
      When someone posts a review on TA or other site they are trying to tell others something. And in that regard I think 60% is not bad at all. Considering that unhappy customers are more likely to want to complain than happy customers want to praise.
      The happy customers that praise are the VERY happy ones. Those that are just pleased wont go through the trouble of writing online reviews.
      What’s interesting is that from a recent study we did TA is the 3rd most efficient source of bookings showing that they do help.

  12. Guillaume

    Great insights but I would +1 Robert Gilmour. Only 60% are satisfied which means the industry has a lot to improve. Having neutral satisfaction is not good enough. By the way what’s neutral satisfaction? Is that we consumers don’t really care?

    You either are satisfied or not. You can’t be in between. Unless you really don’t care when you stay.

  13. Josiah Mackenzie

    Robert – for this research, negative reviews were defined as having a satisfaction rating of 59% or less, neutral reviews were between 60-79%, and positive reviews included scores 80% and above.

    This data includes virtually every online hotel review worldwide in 10 languages. There is always room for improvement for the industry in general, but it’s interesting to note only 12% are truly negative reviews.

  14. Robert Gilmour

    60% positive, 40% neutral (whatever that means!!) or negative – that’s NOT GOOD

    • Kevin May

      Kevin May

      @robert – I don’t believe we are saying 60% is good. We were simply highlighting the figure because there is a perception that reviews are mostly negative.


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