Tripdelta can't do anything ITA Matrix can't as well -- you just have to know what you're doing with Matrix (input alternate airports via the suggestions, or use advanced routing codes). But Matrix will give you far more flexibility and better results.
Assuming this sets up flights where the legs are separate bookings and maybe even multiple airlines, there's a risk that you're on your own and have to buy new tickets if you miss your connecting flight due to delays/cancellations, no?
Correct, just like with Skiplagged, though their FAQ (https://tripdelta.com/en/faq) indicates they may protect for this in the future.
> Is it safe to book your results?
> Yes. We only work with highly respected partners and choose very carefully who we trust.
> Is it safe to book two separate tickets?
> Yes, it is. Just make sure to have enough time between two different tickets on the same day. To make it even easier, we will soon be offering stopover protection.
This is a weak pitch... Kayak, Hipmunk, any almost any other site have an option to include nearby airports when searching.
I compared the route they screenshotted and apparently the only reason TripDelta come up with a $150 cheaper one-way option ($302 vs $455) is by flying Spirit from OAK to LAX with a 9 hour layover. Kayak probably assumes that layover is excessive and filters it out. Because really, if you're that desperate to save funds, just take Greyhound to LAX.
And if you book round trip, returning a week later, it's $779 on Kayak and over $1086 on TripDelta. Plus Kayak will also give you the probability that prices will drop in the next 7 days.
> The cheapest flights in comparison are more like ~$450.
There is nothing wrong with preferring the bus, but your cost estimate doesn't seem right. Did you try searching with the linked site? Or with Google Flights?
I don't know where you live, but looking a month out, I see lots of sub-$200 fares from Cleveland, Akron, and Columbus, as well as some sub-$300 fares for last minute.
Well, Kayak just failed for me on YVR - SFO. PriceDelta found BLI - OAK for $103 versus $200 on Kayak. That's pretty good! I've used nearby airports on Kayak, it doesn't always do a good job from what I've seen and it's +/- 3 days isn't so great.
I really want them to add support for multi-airport "sfo, smf to hkg" and multi-city "sfo to lax to hkg" to the web search parser.
And the mobile web app is decent but compared to desktop very limited. For example, the dates grid shows just a few days vs. 7x7 for depart and return. I think that's one of the most important features. I find it frustratingly inefficient on mobile.
Searched YYZ-SFO on Dec 22. Google Flights gave me US$206, Tripdelta gave me US$248 with SMF (Sacramento) as the destination, with approximately the same airtime. Tripdelta doesn't even show the $206 flight as an option.
Better off using Google Flights with multiple destinations. Put a little bit of effort into finding which origin/destination airports are viable for you, and search directly for it on another OTA.
I'd like to know the exact name of the problem, but it feels TSP-ish.
I'm borrowing this from Vayant (which I think is a B2B competitor) with seemingly similar challenges:
> As mentioned above, the computational challenge here is to solve a “traveling salesman” style problem, with costs associated with each graph segment and in fact each combination of segments.
Is this some sort of paid-for journalism, posted on HN? What is this more than some ad disguised as legitimate article? It annoys me a lot to see that on the front page.
As others noticed, the connections usually suck and are not guaranteed (looked up a european flight, found a 3h30m layover in London between Heathrow and Stansted - well good luck making that!). That is something I could accepted if I was still a student, but no more (since I did not make a connection one day due to bad weather :)
Personal anecdote: several years ago when Ryanair used to offer ultra cheap flights within Europe (as long as you paid with appropriate credit card type and accepted weird dates and times of the flights) I wrote a Ryanair-only poor man's version of this tool, looking for connections through two particular hubs of interest. The code sucked incredibly, but managed to get me two very cheap bookings (well, had to spend some nights in the airports, but it's nothing when you're young and short on money). Then Ryanair did some changes (and also raised the prices) which made the project obsolete. Old good times :)
Regarding the code, I had to do some reverse engineering to parse the responses from the airline's site. Funny part was that the responses were non-deterministic for a given request, seemed like they used some kind of dictionary of possible transformations of output values of price to prevent easy scraping. The other thing that made me scratch my head for a while was that the response had some weird unicode whitespace in one place instead of "traditional" space and I couldn't figure why the parsing didn't work until I downloaded a response manually and opened it in a hex viewer (when I was dumping the response to console in the script, this unicode space was not there).
Having just watched the most recent South Park episodes Sponsored Content and Truth and Advertising, I'm honestly not sure if this a real story or very well placed native advertising... Oh, Lifehacker.
It did not get me any flights that was cheaper than the site i usually use.
In fact the cheapest flight was in general (over my 3 test searches of flying to New Zealand.) 50-100$ more expensive and had one extra stop. About the same flight times.
And about 40USD more expensive when flying to new york and 20 hours longer
> "The site was built by a team of mathematicians who traveled to more than 100 countries"
Could somebody explain why mathematicians(!) have physically travel to 100+ countries to build a web site? Are these airports so secret they could not be reached with modern means of communications?
Ticket hacking and "Mixing" airlines becomes a problem if you miss a flight. This can be highly risky.
What I would be more interested in is this problem:
Often I get reimbursed for a flight from A to B.
Often I can fly A->C->B and make a nice (paid) vacation in C. Hence I look for the most interesting routes possible for a decent price.
Especially exotic airlines can be a great deal here. Currently I look up the airport to see what airlines are served by this airport and then see if there is an airline that has it's hub in an interesting tourist destination.
38 comments
[ 5.6 ms ] story [ 96.2 ms ] thread> Is it safe to book your results?
> Yes. We only work with highly respected partners and choose very carefully who we trust.
> Is it safe to book two separate tickets?
> Yes, it is. Just make sure to have enough time between two different tickets on the same day. To make it even easier, we will soon be offering stopover protection.
An in fact, those suggestions are more expensive then the actual flights to Austin.
However, in your case if they're still more expensive than the direct flights then seems like a broken calculation.
I compared the route they screenshotted and apparently the only reason TripDelta come up with a $150 cheaper one-way option ($302 vs $455) is by flying Spirit from OAK to LAX with a 9 hour layover. Kayak probably assumes that layover is excessive and filters it out. Because really, if you're that desperate to save funds, just take Greyhound to LAX.
And if you book round trip, returning a week later, it's $779 on Kayak and over $1086 on TripDelta. Plus Kayak will also give you the probability that prices will drop in the next 7 days.
Would be surprised if it gives you the same price, usually direct flights are more expensive.
The cheapest flights in comparison are more like ~$450.
There is nothing wrong with preferring the bus, but your cost estimate doesn't seem right. Did you try searching with the linked site? Or with Google Flights?
I don't know where you live, but looking a month out, I see lots of sub-$200 fares from Cleveland, Akron, and Columbus, as well as some sub-$300 fares for last minute.
Ex. On Google Flights, 2-day RTs from CVG <--> EWR over the next 3 weeks are: $407-621, then $371-520, then $375-593.
1. The Most Expensive Airports to Fly To: Business Travelers, Smaller Cities and Other Factors That Drive Up Prices; Sorry, Cincinnati http://www.wsj.com/articles/SB100014240527023039800045795760...
I really want them to add support for multi-airport "sfo, smf to hkg" and multi-city "sfo to lax to hkg" to the web search parser.
And the mobile web app is decent but compared to desktop very limited. For example, the dates grid shows just a few days vs. 7x7 for depart and return. I think that's one of the most important features. I find it frustratingly inefficient on mobile.
TripDelta found me a $655 flight, except out of Portland (which is like a 3 hour drive away), and a $750 flight that looks like
SEA-EWR / JFK-LHR
which is a hell of a connection. I couldn't deal with the inconvenience, but even so I'm kind of impressed.
Better off using Google Flights with multiple destinations. Put a little bit of effort into finding which origin/destination airports are viable for you, and search directly for it on another OTA.
I'm borrowing this from Vayant (which I think is a B2B competitor) with seemingly similar challenges:
> As mentioned above, the computational challenge here is to solve a “traveling salesman” style problem, with costs associated with each graph segment and in fact each combination of segments.
Source: http://vayant.com/airline-availability-solving-travelling-sa...
Personal anecdote: several years ago when Ryanair used to offer ultra cheap flights within Europe (as long as you paid with appropriate credit card type and accepted weird dates and times of the flights) I wrote a Ryanair-only poor man's version of this tool, looking for connections through two particular hubs of interest. The code sucked incredibly, but managed to get me two very cheap bookings (well, had to spend some nights in the airports, but it's nothing when you're young and short on money). Then Ryanair did some changes (and also raised the prices) which made the project obsolete. Old good times :)
Regarding the code, I had to do some reverse engineering to parse the responses from the airline's site. Funny part was that the responses were non-deterministic for a given request, seemed like they used some kind of dictionary of possible transformations of output values of price to prevent easy scraping. The other thing that made me scratch my head for a while was that the response had some weird unicode whitespace in one place instead of "traditional" space and I couldn't figure why the parsing didn't work until I downloaded a response manually and opened it in a hex viewer (when I was dumping the response to console in the script, this unicode space was not there).
I want that job. Maybe I should do a startup that rates the world's vegetarian restaurants. VCs, are you listening?
In fact the cheapest flight was in general (over my 3 test searches of flying to New Zealand.) 50-100$ more expensive and had one extra stop. About the same flight times.
And about 40USD more expensive when flying to new york and 20 hours longer
Could somebody explain why mathematicians(!) have physically travel to 100+ countries to build a web site? Are these airports so secret they could not be reached with modern means of communications?
I suspect it is just a bad marketing pitch...
What I would be more interested in is this problem: Often I get reimbursed for a flight from A to B.
Often I can fly A->C->B and make a nice (paid) vacation in C. Hence I look for the most interesting routes possible for a decent price.
Especially exotic airlines can be a great deal here. Currently I look up the airport to see what airlines are served by this airport and then see if there is an airline that has it's hub in an interesting tourist destination.
Better solution?