Show HN: TripExpert – Professional Hotel Reviews(tripexpert.com)
tripexpert.com
Show HN: TripExpert – Professional Hotel Reviews
http://tripexpert.com/
44 comments
Thanks for the feedback! Very glad to hear that you like the site so much. I agree with a lot of your suggestions, and some of them (like the filters automatically updating) we're going to be implementing soon. To answer your questions, it took about 9 months to build the site (RoR), although the majority of that time was spent developing a back-end that we use to manage all of the hotel and review data.
Awesome. So what do the stars represent? Is that a different sort of rating for each hotel?
It is actually just the "standard" 5-star hotel rating scale. (I put standard in quotation marks because there isn't actually a worldwide common standard for what constitutes e.g. a 4-star hotel.) We've been wondering about whether it is confusing for people to display this alongside our own TripExpert Score. Sounds like you think it is?
Yeah I really had no idea what it was. I think the TripExpert number score is enough. That's the number that matters. At least that's what you're telling me, so why muddle that with another number.
The main reason to include the 5 star rating would be if your score was relative to that. For example, an 80 at a 5 star hotel was much better than an 80 at a 3 star hotel. But I don't think that's the case, is it?
The main reason to include the 5 star rating would be if your score was relative to that. For example, an 80 at a 5 star hotel was much better than an 80 at a 3 star hotel. But I don't think that's the case, is it?
I agree. We're probably going to delete it. Thanks for the feedback!
Very cool. As someone who travels a lot, this is a welcomed resource. I find it very hard to get trustworthy reviews of hotels from Yelp. Generally people have posting bias (negative experience) and often it's specific to a certain suite type or pre-construction of a certain wing of the hotel. Professional in depth reviews will go a long way to help fix this.
Valuable things in a review would also be a listing of some travel essentials like availability of cost and speed of internet, microwave, fridge, room service hours, free shuttle availability, quality of concierge during night shifts, etc.
Valuable things in a review would also be a listing of some travel essentials like availability of cost and speed of internet, microwave, fridge, room service hours, free shuttle availability, quality of concierge during night shifts, etc.
Out of curiosity how do these websites work? (e.g. like booking.com) They build an API and the HOTEL connects it's system this API and then info about rooms/scheduling/prices/pictures are available?
The financial goes by % I guess. You get a % for every booking through your website. But what happens on the technical side?!
The financial goes by % I guess. You get a % for every booking through your website. But what happens on the technical side?!
TripExpert is using expedia's affiliate program. You can signup as an affiliate in a few minutes. They give you a link that takes your users to expedia's site to complete the booking. You get a small percentage of the bookings made via your link.
Hotel.com which is owned by expedia connects to hotels via a Global Distribution System (GDS) like Sabre or Pegasus. All reasonable sized hotel's connect to one of the GDS's (there are 3-4) and travel agents and OTA's like expedia and priceline also connect to them. Getting access to one of the GDS systems takes time and costs a lot. The GDS's speak XML using the Open Travel Alliance's spec. Travel agents typically make a 10% commission on books they make. Hotels pay the commission to the TA and usually a couple of dollars per booking to the GDS. Expedia and the other big OTA's do it differently, making deals with each hotel company and charge much higher commissions.
Hotel.com which is owned by expedia connects to hotels via a Global Distribution System (GDS) like Sabre or Pegasus. All reasonable sized hotel's connect to one of the GDS's (there are 3-4) and travel agents and OTA's like expedia and priceline also connect to them. Getting access to one of the GDS systems takes time and costs a lot. The GDS's speak XML using the Open Travel Alliance's spec. Travel agents typically make a 10% commission on books they make. Hotels pay the commission to the TA and usually a couple of dollars per booking to the GDS. Expedia and the other big OTA's do it differently, making deals with each hotel company and charge much higher commissions.
I like this.
The biggest problem with TripAdvisor hotel reviews (besides fake reviews) is that varied background of the individual reviewer.
Witness how a Holiday Inn can wind up rated higher than a Park Hyatt, simply because of the background of the majority of TA reviewers.
I've largely given up on TA for reviews. For chain hotels, I find FlyerTalk's convention of a dedicated thread per hotel or locality far more valuable, especially because those reviews are specifically from the perspective of frequent travelers rather than once-a-year flyers heading to Disneyland.
For example, 2100+ posts on every possible minute detail of the Conrad Hong Kong: http://www.flyertalk.com/forum/hilton-hilton-hhonors/124932-...
The biggest problem with TripAdvisor hotel reviews (besides fake reviews) is that varied background of the individual reviewer.
Witness how a Holiday Inn can wind up rated higher than a Park Hyatt, simply because of the background of the majority of TA reviewers.
I've largely given up on TA for reviews. For chain hotels, I find FlyerTalk's convention of a dedicated thread per hotel or locality far more valuable, especially because those reviews are specifically from the perspective of frequent travelers rather than once-a-year flyers heading to Disneyland.
For example, 2100+ posts on every possible minute detail of the Conrad Hong Kong: http://www.flyertalk.com/forum/hilton-hilton-hhonors/124932-...
Yeah, I agree; in my experience, high-end hotels often get unfairly rated on TA, in large part because they often get a lot of negative reviews from people who thought that they were too expensive or offered poor value for money. They often end up ranking far too low.
FlyerTalk is a great resource, although you do have to spend a lot of time reading a lot of reviews. The advantage of an aggregator like TripExpert, with snippets from various sources all on one page, is that you can get a much better sense much more quickly of what a place is like.
FlyerTalk is a great resource, although you do have to spend a lot of time reading a lot of reviews. The advantage of an aggregator like TripExpert, with snippets from various sources all on one page, is that you can get a much better sense much more quickly of what a place is like.
It's like Rotten Tomatoes for Hotels.
The main problem with a site like this is that the reviews in most of the guidebooks write something based on TripAdvisor or the like.
Most guidebooks these days are written by compilers who gather information from the internet.
The main problem with a site like this is that the reviews in most of the guidebooks write something based on TripAdvisor or the like.
Most guidebooks these days are written by compilers who gather information from the internet.
This looks great: it's a well-designed site and solves a lot of the problems in TA. The idea that people will rate hotels higher based on "marginal experiences" is very well put, and probably true. Will definitely use in the future.
Nice site. A few thoughts:
- The city search doesn't work if you specify the state (San Antonio, TX for example). I didn't wait for the autocomplete to load and just pressed enter and it returned no results.
- It would be nice if the results automatically updated when changing filters, instead of having to click an "Update Results" button. Also, usually filters are above or to the left of results, not on the right.
- I expected clicking on the small map on the list view to take me to the map view, instead of having to click on the link below. The map is too small in the list view to really be usable as an interactive map.
- The city search doesn't work if you specify the state (San Antonio, TX for example). I didn't wait for the autocomplete to load and just pressed enter and it returned no results.
- It would be nice if the results automatically updated when changing filters, instead of having to click an "Update Results" button. Also, usually filters are above or to the left of results, not on the right.
- I expected clicking on the small map on the list view to take me to the map view, instead of having to click on the link below. The map is too small in the list view to really be usable as an interactive map.
Cool site.
One nitpick: It looks like the automated scraper is bad at "de-duplicating" syndicated (or stolen?) content.
Example: http://www.tripexpert.com/saint-paul/hotels/le-meridien-cham...
Note the identical review text. That would immediately turn me off to the site as "obvious marketing spam" if it wasn't first posted here.
Pretty neat though, I can see how this would be useful in companion with some of the user generated review sites.
One nitpick: It looks like the automated scraper is bad at "de-duplicating" syndicated (or stolen?) content.
Example: http://www.tripexpert.com/saint-paul/hotels/le-meridien-cham...
Note the identical review text. That would immediately turn me off to the site as "obvious marketing spam" if it wasn't first posted here.
Pretty neat though, I can see how this would be useful in companion with some of the user generated review sites.
This is pretty cool. It seems like the search doesn't work that well for me, though. I first typed in "San Diego, CA" and got "no results matched." Only then did I realize that if I type "San Diego" and wait a few seconds (it took about 5 seconds), the autocomplete would populate. But it seems like the matching should be a little bit looser. I tried to find results for Maui, Hawaii, but couldn't figure out how (I entered "Maui" and waited and nothing popped up in the autocomplete, and then searched simply on that term and got no results).
Thanks for the feedback. We're definitely still working on perfecting the search, so this is helpful.
This is a great site and idea; much needed. Do you have an API? I work at a "sort of" OTA "start-up" (200+ mm visitors a year) and it would be nice to explore including these reviews on our site.
Yup, we've got an API. Shoot me an email at [email protected]!
Interesting... my corporate firewall flags your site with an adult-and-pornography filter. Did you recently purchase the domain from someone who may have been running a different type of site? :)
Yeah, that's so weird! We've heard that from a few people. We're currently looking into it. Neither of us can find any history of the domain being used for anything adult or porn related. So strange.
It looks like the domain had three eras. It was parked for a while, then it was taken over by domainauctionhouse.com, and now you guys.
http://web.archive.org/web/20040601000000*/http://tripexpert...
The domainauctionhouse page had a link to domains for sale, which had an Adult section. Maybe the links to domains with bad words in them caused it to end up on filters.
http://web.archive.org/web/20050205024618/http://www.domaina...
http://web.archive.org/web/20040601000000*/http://tripexpert...
The domainauctionhouse page had a link to domains for sale, which had an Adult section. Maybe the links to domains with bad words in them caused it to end up on filters.
http://web.archive.org/web/20050205024618/http://www.domaina...
Interesting to note that this is where TripAdvisor initially started. It turned out its really hard to scale this 1) globally and 2) efficiently and thats where UGC came into play.
Question on the trust factor: Why should a user trust "professional" reviews over the collective opinion of the crowds? Professionals can never experience everything, so will be working from a smaller knowledge base to start and are probably more corruptible than 150 MM people or wherever TripAdvisor is these days.
Question on the trust factor: Why should a user trust "professional" reviews over the collective opinion of the crowds? Professionals can never experience everything, so will be working from a smaller knowledge base to start and are probably more corruptible than 150 MM people or wherever TripAdvisor is these days.
I think they give a pretty good answer to this question in their 'about' section:
http://www.tripexpert.com/about
1) up to 40% of user hotel reviews are fake, 2) users rarely visit more than 1 hotel per city so have no basis for comparison, whereas experts (presumably) visit multiple hotels in each city, 3) users tend to post reviews after having a 'marginal experience' that isn't really all that applicable to other users.
I think they should consider putting these 3 points on the main page as it was the first question I had when I visited.
1) up to 40% of user hotel reviews are fake, 2) users rarely visit more than 1 hotel per city so have no basis for comparison, whereas experts (presumably) visit multiple hotels in each city, 3) users tend to post reviews after having a 'marginal experience' that isn't really all that applicable to other users.
I think they should consider putting these 3 points on the main page as it was the first question I had when I visited.
Great site!
So how do you come up with the TripExpert score? Do you read every review, map it manually to a number, then get the average? Or is it more like Rotten Tomatoes where each review only gets a shot at a thumps up or thumps down, and then you average those?
Btw: Slight coding error: On city pages, hotel names are overlaid over the preview icon, which makes them very hard to read on Safari 5.1.
So how do you come up with the TripExpert score? Do you read every review, map it manually to a number, then get the average? Or is it more like Rotten Tomatoes where each review only gets a shot at a thumps up or thumps down, and then you average those?
Btw: Slight coding error: On city pages, hotel names are overlaid over the preview icon, which makes them very hard to read on Safari 5.1.
Thanks for the bug report.
We have an algorithm that looks at a large number of factors (number of reviews, each reviewer's unique denotation system, number of hotels in a destination, sentiment, etc) and calculates the number.
We have an algorithm that looks at a large number of factors (number of reviews, each reviewer's unique denotation system, number of hotels in a destination, sentiment, etc) and calculates the number.
This is a great looking site but for me it illustrates the problem with "expert" sites versus "crowd" sites. Yeah, you have a ton of hotels in NYC, LA, etc. but I'm going on vacation to Connecticut and Maine this summer and there's nothing for either state in your search. I'd love to use it but there's going to need to be a lot more experts.
We'll be growing, don't worry. Today is day two. :)
Outstanding execution could be smoother in terms of getting to the booking.
have you considered an I'm feeling lucky function?
Do you have an API that other developers or designers can extend?
Have you considered other verticals such as cruises and tours?
Hope it goes well guys love to hear from you what the conversion rate from users.
hutch
have you considered an I'm feeling lucky function?
Do you have an API that other developers or designers can extend?
Have you considered other verticals such as cruises and tours?
Hope it goes well guys love to hear from you what the conversion rate from users.
hutch
Thanks for the positive feedback! Yes, we're building an API that will let other travel sites pull our scores, rankings and review extracts. If you or anyone you know might want to make use of it, tell them to get in touch! And, yes, we're considering other verticals.
I can't see a way to search for a hotel by name once a city has been selected. This is the first thing I tried to do.
The rank is supposed to be how well it does in its class. But I looked a few cities and the best hotel was always expensive (over $300). Never a good for the value midrange hotel. So it seems best isn't best in class.
The rank is supposed to be how well it does in its class. But I looked a few cities and the best hotel was always expensive (over $300). Never a good for the value midrange hotel. So it seems best isn't best in class.
Beautifully designed and very useful. I hope this is a big success! What's your background, how did you start this site?
My co-founder and I actually met at a hackathon a few years ago and worked on an early version of the site. It's very exciting to finally launch!
My co-founder Andrew is a lawyer-turned-developer. I'm a former tech journalist (Laptop Mag, The Verge).
My co-founder Andrew is a lawyer-turned-developer. I'm a former tech journalist (Laptop Mag, The Verge).
Who did the design? Also, a little usability tip: When a user clicks on "Price" to sort, I think most users expect to see the lowest price first. You have it sorted by highest price first. I would reverse that.
Also, I noticed the lowest price hotels often don't have a great score (around 60-70). Do you only list hotels that hit a certain score threshold (meaning, as a user, can I trust that no matter what hotel shows up in your listings, it means I can expect a good stay?) - or do I still need to be wary of the scores and avoid ones with low points?
Also, I noticed the lowest price hotels often don't have a great score (around 60-70). Do you only list hotels that hit a certain score threshold (meaning, as a user, can I trust that no matter what hotel shows up in your listings, it means I can expect a good stay?) - or do I still need to be wary of the scores and avoid ones with low points?
Our designer was someone that my co-founder had worked with before. And thanks for the tip about price sorting.
We don't list any hotels with a score lower than 60, which means every hotel is a quality hotel.
We don't list any hotels with a score lower than 60, which means every hotel is a quality hotel.
this is very nice. im literally in the middle of booking a few weeks in greece with my girlfriend. trying to figure out the hotels on the different islands is a pain in the ass. will give you more feedback after using it a bit more.
It looks good. My suggestion for the "Explore"-Page would be a map or at least a way to sort alphabeticly. I guess the list is now sorted by the ammount of recommended hotels.
Have you considered a 'hint' feature where people can kind of recommend hotels that are not yet in your database and could be considered personal tips?
"Finally, hotel reviews you can trust."
I thought the beauty of TripADvisor and the like is that it does away with the old travel guides. Guides are written by people with even more questionable motive and are quickly outdated. Not sure how many times I have been saved from bad experience by reading a recent reviews saying "BEWARE Hotel is under construction for next month..." or something current like that.
Maybe there value in both... not for me.
I thought the beauty of TripADvisor and the like is that it does away with the old travel guides. Guides are written by people with even more questionable motive and are quickly outdated. Not sure how many times I have been saved from bad experience by reading a recent reviews saying "BEWARE Hotel is under construction for next month..." or something current like that.
Maybe there value in both... not for me.
I like the irony of AirBnB styling for the hotel industry.
A couple of things I noticed:
1. After updating a search, the back button doesn't work the way I would expect it to. If I choose to look at a property and then back out it takes me to the pre-updated search whereas I would expect it to take me back to my search with filters in tact. If I don't look at a property and press the back button it takes me out of search altogether where I would expect it to take me to the pre-filtered search.
2. On the search results page, the stars confuse me. I don't know what they represent.
3. It would be nice to be able to see hotels that score below 60 (perhaps as a filter option). Budget conscious shoppers might find they are lacking options otherwise in certain cities. Sometimes you're just looking for the best of the cheapest, even if it's not a 60+ hotel.
4. From the blog page, I think the TripExpert logo should take you to the main TripExpert home page, not the TripExpert blog home page. I see the Visit TripExpert link but considering that is (probably) the primary action you're probably hoping for, I think it shouldn't be hidden in the nav like that.
5. I was also initially confused when I added a filter and the search results didn't automatically update. I think that's become such a standard UX that you might want to adopt it too.
Overall though, this is really impressive. Great job. I think you have a wonderfully bright future ahead of you.
EDIT: Added 4th suggestion
EDIT 2: Added 5th suggestion