Reddit CoFounder Dips Back Into YC With Travel Startup Hipmunk (YC S10) | Hacker News Reader

Reddit CoFounder Dips Back Into YC With Travel Startup Hipmunk (YC S10)

techcrunch.com

175 pointsicey78 comments

What I've always wanted is travel booking, not flight booking. I want to know how long, how much and what my options are for getting from $x to $y at $z time.

This still drives me nuts. If I'm at home in Berlin and have a meeting in in central London, I don't want to have to figure out that if I fly from one of the two airports in Berlin to one of the four airports around London that I'll have to use these kinds of busses and trains or taxis or rent a car and have to sort all of that out myself.

I want to know how to get there door to door, and I'll pay a decent premium to have it all put in front of me where I can buy my tickets all in one go. I hate travel booking because it seems to be without fail a minimum 2 hour exercise sorting out the options and picking one.

Back in the day I had a go-to travel agent for such. Unfortunately finding a good travel agent often requires checking out 10-15 of them and I've been too lazy for that as well since recent moves.

Blog post on this a while back: http://scotchi.net/2009/09/what-i-hate-about-booking-travel-...

What kind of premium would you (or the average traveler) be willing to pay for a travel agent replacement webapp? I've been thinking about this as well while working on my own travel based project. Perhaps this is an opportunity for me to pivot in the future, if need be.

10% is my gut feeling. I'd imagine that the plan of attack would be figuring out the top 10 or so most important cities and working out the details for them (airport transportation, transit, car rental, taxi rates) and going from there.

As to loumf's concern, once I trusted the site (say, after getting one booking commission free), I'd have no problem with the details not being revealed to me as to flight numbers / lines, details of public transit, etc. until I'd purchased. But I'm not sure you'd need that if the commissions are reasonable; I'm willing to pay to have my planning simplified.

When I used travel agents, it's more like this -- I have $X and I want to be at a beach next weekend (Island? Florida? Big lake? Fri-Sun, Thu-Sun, Sat-Tues -- what's the cheapest)

If I know exact dates, places, etc -- travel booking isn't hard enough to pay much for. The problem is, you go book it somewhere else when you have the info, so you have to go after affiliate fees and be as good at the booking as you are at the advice.

While working on a similar product (http://tripadvisor.com/flights) I suggested that incorporating transportation to the airport in terms of both time and price would be a great feature. Didn't happen but I'm still convinced it would be really useful in figuring out tradeoffs.

seems daft they'd overlook that... surely they could build it in as a way of upselling bus/train/taxi/car hire tickets if nothing else...

I think its a challenge to collect the information and keep it updated.

Not to knock Hipmunk, but this visualization has been available for years as the "graphical view" of ITA Software's matrix demo app:

http://matrix1.itasoftware.com/

Do a search and then View Flights - Graphical. It's excellent and has been my preferred way to find flights for ages.

It's nice to see that someone who might actually care about marketing is offering the same useful tool!

(I work at ITA on the QPX flight search engine).

The new Matrix (http://matrix.itasoftware.com/) has the color bar functionality, too, but looks a littler friendlier! When you search, click "time bars" on the right below the carrier matrix. It was one of the most-requested features.

Just downloaded the iPhone app, just what I've been trying to find forever!

Comments:

- after doing a YYC to YVR search and getting the result, clicking on a price brings up a blank screen, is that a bug? Clicking on the airline shows results.

- the iPhone version doesn't seem to have the same flexible dates option as the website search, is this going to be added?

Thanks for your feedback! I've passed it on to our iPhone people. If you find anything else, there's a feedback button in the app, or you can email onthefly-comments@itasoftware.com. I wasn't able to get a hold of the development schedule or a developer just now, but I imagine flexible dates would be on the list.

Seems to have been fixed the next day, nice!

Please answer me this:

Why did I not know of this website? You don't even seem to have a front page, on the link another person posted, yet when I tried your link it is easily and by far the best fare search engine I've seen. All the others are bloated, slow and plastered with ads.

Who has been doing your marketing?

Because ITA doesn't care, they power the back-end of a lot of big players (eg. Orbitz) and they use some serious tech (which is why GOOG bought them), they don't really care about the consumer market and don't really want to play there. Makes sense to me.

Oh, I see, I didn't know that, thank you.

I'm curious why ITA doesn't actually make it possible to BUY the tickets. It seems silly to walk people through the entire process, provide them with so much information about available flights and prices and then say... go and find someone else to sell this to you.

Why not just partner with a travel agency and make life easier for people?

Mega-volume travel agencies / search engines are their customers, not you.

Do I have to know the airport codes? It didn't recognize San Diego at all.

Thanks for your feedback! That's a bug - we should have a fix out tomorrow. The issue is with spaces in completion, so if you just type "san", you should see San Diego in the list.

ITA is probably the most underrated travel app on the Net. It's my favorite, cleanest, and also the most powerful tool for searching for good value flights (you can also find flights that will net you maximum miles that sites like Orbitz don't make easy, even with ITA powering the backend).

I think Hipmunk is facing an uphill battle. I don't want to be a naysayer because I want them to do well (and thereby have more choices), but ITA spent years and years to get the kind of buyout they're getting now.

Hipmunk seems to basically just be an attempt to mimic ITA's graphical UI with the ability to purchase tickets, which seems like a dubious business proposition (we'll see). I would think Orbitz could easily just stop doing its affiliate program with them, and then quickly enough make its own identical layout...

And matrix.itasoftware.com is indeed awesome-the query language lets you do incredibly useful stuff you cant to anywhere else. The hard part is then twisting the arm of orbitz/travelocity/etc to get you to book the exact same flights.

Or more likely Orbitz does the affiliate deal with them now. Then if it appears Hipmunk is popular and getting traction then Orbitz tries to buy them, at the very least to keep it out of Google's hands and/or beef up their UI features relative to ITA/Google.

ITA is awesome, but their value isn't their interface so much as they supply the fare/availability data to practically everyone (Kayak, Orbitz, fly.com, and us, through via Orbitz).

Curious, who is "us"?

spez = long-time nick of Steve from reddit (hi Steve!) and now hipmunk

How difficult is it to get a deal with Orbitz or ITS for access to this data? Do you need a bizdev guy to talk them out of a six figure annual commitment?

I've got a very different travel idea, and would love to be able to try it out, and send everyone to Orbitz to book it if they end up finding what they want on my site.

But without the data it is very hard to experiment in making a better travel site.

The main reason flight search is so far behind the rest of the tech world is that the inventory is so difficult to access. You have to either be very well connected and able to persuade one of the gatekeepers to give you access on a trial basis, or have money to purchase data from commercial suppliers like ITA, OAG, ATPCO, each of which is hugely costly and limited in the type of data they provide and the completeness of coverage.

For bootstrapping a travel search product, you're best to get busy building crawlers to scrape airline websites directly.

Look up the affiliate link on orbitz site and start from there. I think you'll find it pretty easy to at least start something out. If you have no special needs above the api they offer their affiliates, then you will be good to go pretty quickly. If not with Orbitz, try booking.com or expediaaffiliate.com

Similar take here. I used to work at Orbitz and I know for a fact about 4 years ago an engineer created a UI extremely close to Hipmunk, in his free time, as a solo prototype project. Not sure if it ever got product-ized however. Travel search visualization and travel completeness/door-to-door still seem like have plenty of opportunities to innovate. So much crapritude in the travel industry.

The visualization, yes, but there's a lot more to what we're doing than that. Better sorting, pruning out dominated flights (80% of results), trains, and time sliders, for example.

True—but Matrix doesn't let you book tickets. :)

An example is here (for the lazy):

http://matrix.itasoftware.com/flights.htm?session=91c4f25b-e...

Sadly the links ITA provides are session-based and expire, which is very sucky.

I really like the visualization and the sorting by "agony", but there's a few features Kayak has that are must-haves in my mind:

* Flexible dates. Sometimes I have a day or 2 of leeway, and it's often a lot cheaper to fly one day different.

* Flexible airports. I live roughly equidistant from 2 airports, and Stanford also has 2 reasonable airports to use.

It's not amazingly obvious, but you can do flexible airports by putting in an area code or city that is between airports in to/from fields. i.e. if you do from:"palo alto" to:"salt lake", it will show flights leaving both SFO and SJC that go to SLC.

Something else interesting I noticed is that it seems to also return results for Amtrak.

What does it take to get the airline price and schedule data?

I imagine a lot of people want to make sites like this, but they don't have the resources to obtain contracts for data feeds.

Much like many people would like to open up streaming sites for Big Label music, but not everybody can afford to enter into contracts with music labels then have hundreds of HDs full of wav files delivered to them.

We need someone to make a pay-per-taste brokered data market.

I gave a decent (well, at least a lengthy) answer to this question here: http://news.ycombinator.com/item?id=1473748

Basically, getting the raw data is quite difficult, but once you have it, you need to expend a really enormous quantity of engineering work into using that data to get useful answers.

For example, ITA invented a language to describe taxes. That's because the tax code changes all the time and is incredibly complex. There are fares which different entities in the airline industry price differently due to tax code ambiguities. The search problem is really very hard once you have the data.

Now, ITA software does offer their flight search engine as a service: you sign a contract with them and issue queries and they send you responses. I don't know how interested they are in cutting deals with really tiny startups, but they might be amenable. You will of course have to pay in some fashion: maybe a fixed fee per query or pay per compute time or pay per successful booking.

That's very illuminating. I have some ideas for displaying / managing flight and travel information that I would really love to implement, but the "Data Problem" has loomed large for experimentation. I didn't in fact realize that it was even as burdensome as you describe.

Just issue query to Orbitz via the affiliate API and they'll pay you for every flight booked. The TOS have some limitation, but if you can work within them you can explore pretty easily.

It's really quite ridiculous. I made the backend to a travel site using data from Galileo reservation system (one of the big ones). At the time, they had a relatively new SOAP-ish API. The data you got back was huge and arranged in ways that are not in any way obvious. Many calls just return what appear to be screen scrapes of 70's era terminal screens, wrapped in XML. It was less than fantastic.

At the time I left the travel agency in question, Galileo was in the process of rolling out a new version of the API, but the few times I tried to get it to work, it seemed completely broken (this was for cruises, not flights, but it was the same general idea).

This is wonderful. I will absolutely be using this the next time I want to search for a flight. The logo makes me smile (although it's a bit indistinct on a white background), and the fact that the default sort method is "agony" makes me very happy.

It seems to be including Amtrak results in there too, which makes for a nice comparison.

The logo was designed by (Reddit cofounder) Alexis Ohanian.

http://twitter.com/kn0thing/status/21425755974

oh hey thats pretty good. I wonder if I could get him to make a dingo for the startup i'm working on :) haha

Thanks. After the reddit alien, then the breadpig, and now the hipmunk, I'm really weary of pushing out a mascot of fail, so I'm pretty selective about work I do :-/

My notes:

- "Hipmunk?" I don't get it. Why not tripmunk?

- I like the visualization and the "agony" sort. Pretty cool

- Is the plan to book the trip directly on the site eventually. Once I got to the end, I thought "Why don't I just go to Orbitz next time if that's where I end up anyway?"

Try finding an available domain name with the word "trip" at the beginning...

tripluckydip, tripyer, tripfingertip, tripunzip, trippirt. Hipmunk is good.

"Hipmunk?" I don't get it. Why not tripmunk

Derived from chipmunk, I presume. Tripmunk would've been better though.

Company site: http://www.hipmunk.com/

Crunchbase profile: http://www.crunchbase.com/company/hipmunk

Launch announcement: http://blog.hipmunk.com/hipmunk-launches

I absolutely love the interface, especially:

-Sorting by "Agony" is a brilliant decision. It seems to take into account time spent traveling and cost, which leaves me with a few easy choices to make (spend 20 bucks more, save an hour) instead of a million tiny choices to make.

-I love that the prices seem to take into account taxes and fees without any trouble. I see $550 on Hipmonk, click "buy," and then see $550 on Orbitz.

-It looks nice and performs excellently.

I wonder if seeing data in this format will cause airlines to differently optimize their fare schedule. Obviously I don't have the knowledge of the data to answer this, but my gut feeling is that more people selecting flights based on travel time might lead airlines to find ways to reduce travel time. Wouldn't that be sweet.

First time I actually got relevant results for a search (a bit obscure search). I don't have any need for tickets now, but when I do I'll certainly use this.

Calendar for input of depart and return was a bit confusing at first.

I've been having to do a lot of airline searches recently and this should definitely alleviate the pain. Being able to see the flight plans visually, especially the layovers, is extremely nice. As is the automatic hiding of flight itineraries that suck.

Being able to adjust the day would be a nice feature, would be cool to be able to drag a time slot around to see if there is anything better a day later or a day earlier. Perhaps being able to compare days side by side would be nice.

Overall though, it's pretty awesome. Only thing I've seen that comes close is Kayak.

Having looked at the interface, it's basically taking orbitz results and slapping a slightly nicer GUI on it. If they were backended into ITA instead of hopping via Orbitz that would be better.

What I would like to see is something that ITA has already, which is see cheapest fares by month. I'd also love to see all the weird fares that FlyerTalk people have. Also, US Centric sites: the pricing information for inter europe or asia flights is dismal! Dismal!

Wow. That is seriously good. I've been planning a trip with my friends and the table-view is absolutely superior to everything I've ever seen. Keep it up Steve!

> When you see the results (see screenshot for example), you’ll never want to see flight results in any other format. It’s one of those that’s so obvious why didn’t I think of that moments.

I disagree. I think there are several different groups of people that search for flights and each of them wants to see their results in a different way. I think the other flight search engines (especially Kayak) meet most of their needs quite well. It's great to see another view into this data, but I don't think it's the only way I will want to view it.

When I search for a flight, the only factor I care about is price. I don't care when it leaves, how long it takes to get there, what airline it is: I want it to be the cheapest it can possibly be. When prices are close, I'll sometimes care about airline (I really on't like to fly AirTran for example).

I think Hipmunk's view into the data is perfect for someone that cares about the time spent in flight, layovers and which part of the day they're flying. These people are usually business-people who have money, so it's probably a perfect market to shoot for :).

You know you can sort by price, right? Right at the top there's a "sort by". And you might see that you only have to pay $5 more to avoid leaving at 6am and not have a ridiculous layover.

My first time through I didn't notice the prices were listed on the page.

A massively large percentage of the travel market is composed business travelers who are generally pretty price insensitive and a lot about things like flight times and layovers. Its probable that you're just not in the target market for a product like this.

>When I search for a flight, the only factor I care about is price. I don't care when it leaves, how long it takes to get there, what airline it is <...>

You may not, but there are others that definitely do

Why go through yc? Or raise a small angel round?

My guess is that it worked the first time, so why not try it again? That and they may feel some loyalty to PG. Well, then there's the "instant post in TechCrunch" one gets when combining YC with a successful prior exit. They are bound to get lots of press and a running start going the YC route, especially when combined with past experience.

I'd like to believe that it's because YC is actually beneficial. After all, a new startup by a founder of Reddit was going to get lots of attention regardless.

What do you think the benefits of YC are to a company started by previously successful founders? (Honest question... seems like you have mostly touted the connections you can broker, which I wouldn't think they need help with now.)

Basically the same as for a first-time founder, which we explain here: http://ycombinator.com/about.html

We've never claimed connections are the most important part. That's probably the most common misconception about YC, and one we try actively to dispel, among other places on the page linked to above.

Maybe to be subject to deadlines?

If you are in the YC program, you have different milestones to achieve by certain dates.

For me, it's like training to be able to run long-distance. Best way to motivate yourself is by actually entering in a marathon race.

Oh, I believe there's certainly something to be gained by going with YC, beyond what any prior success would offer. Sorry if my comment may have not gotten that across.

There probably is some loyalty there, but that's largely due to your prior success in getting Reddit launched. If you hadn't been helpful, someone would likely pursue other options.

Wow, very nicely done! This kind of thing is very important for international travel, as they often give you some crazy travel proposals.

I tried VCE to PDX in October, and it seemed to do pretty well, although I'd appreciate it if it gave me more information on how to get a cheaper flight, if changing the dates would help at all.

It's a bit confusing to figure out where this is better. I don't really grasp the agony part. For flying the real challenges for me are not layovers but planning the stuff out once you get there. Also, finding more local airplane companies internationally..

So how are they driving traffic? They seem to have no SEO focus at all. When you're relying solely on brand equity, you're doing yourself a strong disservice.

Of course, that SEO part is going to be pretty expensive for them to compete in the travel space.

I personally never start travel from search engines -- I go to a Expedia, Orbitz, sometimes the airline directly (Southwest).

Traffic from brand can be better than traffic from SEO in some cases. You have to spend money, but it's under your control.

Travel-related queries of any sort are probably the one query I most shy away from ever entering into a search engine. The results are just so filled with garbage that it's rarely useful--- I'll go straight to a booking engine, to Wikipedia, to Wikitravel, to Lonely Planet, or to some other known site to get info, rather than deal with thousands of content-mill pages and resellers of vacation packages.

Does he need to launch his second startup through yc? yc's most valued service to a startup is right connections. Having been successful once he should already have plenty of them

The name at first struck me as odd, but it's growing on me the more I think about it. As long as the chipmunk in the logo is really cute then it will be memorable.

I think this is the first homepage for a travel site I've seen that doesn't hide the calendar picker. Whoever is designing this site knows what they are doing.

It seems, it searches only for domestic flights.

There needs to be more guided direction on the landing page. Your users shouldn't have to guess what to do.

Is 15K not very less amount. I mean, I've never seen any company receiving such a small amount in funding.

Is this normal?

Really would like to hear how they got access to the data. Does Orbitz do deals like this?

I've got a technology that I think would be applicable. Be happy to hand off any booking to the data provider... but just need to be able to get to the raw data.

This wouldn't compete with hipmunk, it is a different idea in the travel space.

Just trying to find someone to supply data seems impossible.

Reply on news.ycombinator.com