Hipmunk, Conceived by David Pogue’s Teenage Co-Author, Embarks On Mission to Make Travel Search Easier

8/18/10Follow @wroush

(Page 3 of 3)

eat and breathe flight search algorithms, and have made their company the dominant provider of flight information to other companies like Orbitz. (Which is probably why Google is in the process of acquiring the outfit for $700 million.) What makes Goldstein think that he and Huffman have the key to the next generation of travel search?

“It’s all about the user experience,” Goldstein says. “The back end problems, you’re right, are big, big problems, and there are people a lot smarter than me working on them. I think the opportunity is to make the presentation of the information more easily understandable and more useful. That’s why we decided to start there.”

Goldstein argues that the competition is moribund. Orbitz, Kayak, Expedia, and other companies have been offering the same graphical views of flight options, often with ITA’s software under the hood, for years. Last year, TripAdvisor added a new flight meta-search service to its traditional hotel and restaurant guide—and in an odd coincidence, one of the people who built it, Brian Krausz, is a classmate of Goldstein and Huffman’s at Y Combinator this summer, working on his own startup, GazeHawk. But while Goldstein says that while he appreciates “the thought they seem to be putting into flights,” he doesn’t see the TripAdvisor offering as a qualitative improvement over other flight search options.

Goldstein and Huffman—who have split up the coding work on Hipmunk about 40/60, Goldstein says—have clearly put a lot of work into their own interface. Flight data, which currently comes exclusively from Orbitz (which means it actually comes from ITA), shows up extremely quickly, and is arrayed using time bars that make it easy to visualize flight departure and arrival times. When there’s a cluster of similar flights available, the interface shows only the one with an optimal combination of price, duration, and layovers, and hides the others, though they can be easily revealed. A row of buttons along the top of the screen lets the user instantly re-sort flights by price, duration, number of stops, departure time, or arrival time. Draggable sliders let the user adjust the acceptable departure or arrival times, and there’s even a tab system that allows users to keep multiple searches open in different tabs.

ITA Software's flight search matrixIt’s an impressive effort that obviously involved a lot of Python and JavaScript coding. But overall, Hipmunk’s user interface is very similar to a search results matrix developed by ITA Software (see graphic at left). ITA hasn’t ever seriously marketed this interface, and as far as I can tell, it’s a secret to all but the geekiest travelers. But the point is that Hipmunk is going up against a 500-employee company that’s about to become part of a 20,000-employee company.

Goldstein’s reply comes in two parts. First, he says, Hipmunk has thought harder than most other flight-search providers about usability. “We think [ITA's matrix] is a cool interface, but we think we have a lot of stuff that differentiates us, like the draggable sliders for the time of day, hiding the flights that you’re not interested in,” he says.

Second, he says it would be premature to judge Hipmunk based just on the product it launched Tuesday, which, if you wanted to boil it down, could be described as merely an improved interface for flight search. Which gets to an interesting paradox that afflicts almost all Y Combinator companies: Graham and the other mentors involved in the startup incubator relentlessly push YC companies to launch products early and “iterate,” that is, come out with serial improvements based on user feedback. That’s a perfectly good way for a software startup to get through its early stages, but it does put the incubatees in the position of having to plead for customers’ (and journalists’) patience while they iterate toward something weighty and worthwhile.

Goldstein insists that there’s much more to the Hipmunk vision than meets the eye. “The broadest frame here would be that we want to help people find what they are looking for,” says Goldstein. “A little more narrowly, it would be helping people understand what the tradeoffs are in the world of transportation and tourism. And then, narrowing down even more, what we have right now is a tool that helps people understand the tradeoffs in flights.”

The trick for early users—not to mention potential investors—is to look at Hipmunk’s current tool and then zoom back out to the bigger frame, imagining what might become possible if Goldstein and Huffman are given the time and resources to iterate on their vision of rationalizing the travel search business. If they can conquer flight search, after all, there’s a lot more data that needs to be organized and displayed better: hotel rooms, rental cars, restaurant reservations, and basically any commodity where people need to weigh prices, schedules, availability, and other data all at once.

Time will tell whether it’s Goldstein’s destiny to solve such problems. But he’s going to try. “Everything in my life has been leading up to working on a startup,” he says. “Sometimes people ask why I didn’t drop out of college to start a company, and I say that I never really found an idea that was good enough to drop out for. If I’d found this idea three years ago, I might have.”

Wade Roush is a contributing editor at Xconomy. Follow @wroush

Single Page Currently on Page: 1 2 3 previous page

By posting a comment, you agree to our terms and conditions.

  • http://danweinreb.org/blog Daniel L Weinreb

    I’d like to comment on the general concept that it’s best to get out to customers, get feedback, and iterate as quickly as possible. There are a lot of good things to be said about that approach, and they’be been said in many places, so I won’t repeat them here.

    There’s the drawback that Wade mentions. But also: “You only get one chance to make a first impression.” If someone comes and sees your newly released “0.2.1 beta”, which has fewer features and more bugs than you would like, it’s possible that she will say “This site is no good”, and never come back, even when you’ve fixed the problems.

    It seems to me that to make the “get out there early and iterate” strategy work, it’s a lot safer to get out there early with only a small number of users, chosen to be friendly enough to not bad-mouth you, but honest enough to give you uncensored feedback. During this period, you should not publicize yourself. For example, you should tell Xconomy that you’re not ready yet. You should avoid buzz and hype. Save that up for when the product is “ready for prime time”, i.e. ready for lots of users who you did not hand-pick.

    This is entirely my personal opinion about the strategy in general, not about Hipmunk in particular. It has nothing to do with (full disclosure) my being an ITA Software employee nor an Xcomomy investor.

  • http://www.roomatlas.com Roger Middleton

    Love it. A true “why hasn’t anyone done this?” moment for me. If they can expand the range of providers to be competitive with Kayak and Skyscanner I’d make this my first and only stop.

    Speaking of travel visualizations… a shameless plug for my own “Hotels.com hotels on Google Maps” site: http://www.roomatlas.com. Live prices in the markers, Street View and Wikipedia placemarks.

  • http://grumomedia.com Miguel Hernandez

    Hipmunk’s interface has improved considerably after this post.
    I created a little video showing this new and improved interface and also explaining what Hipmunk is all about.
    The video is here -> http://youtube.com/watch?v=6teBPUgz4Y8
    Regards, Miguel.