Hipmunk on the Make: The First-Birthday Interview
I first covered Hipmunk on August 18, 2010, the day after the online travel startup debuted its innovative time-based flight search interface. One year later, how’s the Y Combinator-backed company faring? Spectacularly well, actually.
In an announcement marking its first birthday last week, the startup said that users are conducting a million combined flight and hotel searches every month, and that the number of searches is growing at 15 percent per month. About 30 percent of flight searches now happen on Hipmunk’s iPhone, iPad, and iPod touch apps, which it introduced in February.
The company now has 10 employees and a groovy office in the old Hamm’s Brewery in San Francisco’s Mission District, and has rounded up $4.2 million in venture funding from Seattle’s Ignition Partners and a long list of other investors, including SV Angel, Real Networks’ Rob Glaser, former Expedia CEO Rich Barton, Gmail inventor Paul Buchheit, WordPress creator Matt Mullenweg, and actor Ashton Kutcher.
Hipmunk’s growth has “wildly exceeded our expectations,” says CEO Adam Goldstein, 23, who co-founded the startup with Reddit founder Steve Huffman. One key to that growth was the addition of a hotel search capability this spring, again drawing on clever visualization techniques to show hotel choices in the context of other important data, such as location.
The actual data and prices Hipmunk displays are pretty much the same as those you’ll find at any other travel search site; the startup’s main pitch is that it can simplify the trip planning process by sorting the results using intuitive measures such as “agony” in the case of flights (the length of the flight, the number of connections, etc.) and “ecstasy” in the case of hotels (proximity to good food, nightlife, convention centers, and other attractions) and providing easy search customization options.
I caught up with Goldstein last week; an edited summary of our conversation follows. Three especially interesting themes: the difficulty of building a business around flight search alone; the growing importance of Hipmunk’s mobile apps; and how the startup is acquiring users without spending a lot of money to do so.
Xconomy: What are you most excited about right now?
Adam Goldstein: Hotels are really big for us. They are interesting for a bunch of reasons. The fact that we didn’t have hotels before meant that we were sending customers to other services not just for hotels but for flights, because people like to shop for both at the same time. The other side of it is that hotels are just a wildly lucrative business. There is a big multiple [in the commission] per hotel book versus flight booking.
X: Why is that?
AG: There are a bunch of reasons you can point to, but they’re all theories. Only a few airlines control most of the market in the U.S., so the airlines have a fair bit of negotiating power by virtue of scale. Whereas if you are searching a website for a hotel and a particular chain is missing, you might not even notice. The big guys know they are going to be losing tons of money if they aren’t showing up in the search results. They’re willing to pay a whole order of magnitude more, as a percentage of the booking price. So the name of the game in online travel, for better or worse, is getting people to buy flights and hoping they end up booking a hotel at the same time.
X: Was it part of your original business plan to offer hotel search, or did you add that after you realized how lucrative it was?
AG: Before we launched, I don’t think we fully understood the business model. We just knew that if flights worked out, it was a natural next step for us. But the path we’re taking is a pretty well-worn one—you start with flights and then hotels, and then cars and packages and local deals.
X: With your flight search, you’re known primarily for the distinctive visual way you present flight options. Your hotel search is also very visual, with an emphasis on maps and a hotel’s proximity to various attractions. How did you decide what the “Hipmunk style” was going to be when it came to hotel search?
AG: The inspiration came from our own way of thinking about searching for hotels. We think in different ways depending on the purpose of the trip. When I’m traveling for business I have a very different way of thinking than when I’m traveling for vacation. Our first version [of hotel search] was very much built around the leisure traveler. It was built to show the interesting parts of a city; it answered the question “just tell me where I should stay.” But as I traveled more for work, I realized that specific geography mattered a lot too. If I have a particular place I need to be in a city, I need to know what hotels are nearby. So we have rolled out a bunch of features to facilitate that kind of searching. We give you a little flag that you can drag around to whatever point you want, and it will sort the options by distance in real time. That’s parallel to the “time sliders” in our flight search, which let you say “I can’t leave before this time or after this time.”
X: Before this interview you mentioned that your mobile apps have been getting a lot of uptake.
AG: This has been one of the other highlights of the year, that we were able to port the Hipmunk experience from the Web and make it more useful on a mobile phone or tablet. There are two ways you might be searching for a flight. One is sitting down in advance and checking the airfares and seeing if there’s a reasonable flight. The other is that you’re at the airport and your flight got canceled and you need to find a new one right now. So the immediate value of using Hipmunk on your iPhone or iPad is solving what’s almost a bigger pain point.
X: Do you think mobile usage of Hipmunk could eventually overtake desktop Web usage?
AG: It’s an interesting question. We are kind of hedging our bets. I do think we are nowhere near the equilibrium point. If it’s getting to the point that 90 percent of our users are mobile, then the only sensible thing to do will be to divert our development resources and say the desktop just isn’t where it’s at anymore. But I think there are certain things you can do on the website that the mobile app experience just can’t compete with, and those tend to be the quick-turnaround features like the flag on the map that are fairly involved if you want to do them in a native app.
X: You mentioned rental car search and vacation package search—is that where you’re going next?
AG: I think if we wanted to roll out cars and vacation packages, we could easily do it this year. But we wanted to take a break after doing a big project [i.e. hotel search] and make sure it’s at the level of usefulness that our brand stands for before we call it done. We can’t do them all together, because we have a really small, nimble development team. The flights website was written by me and Steve. Our hotels project was written by two people. Our iPhone app was written by one guy and the iPad app was written by the same guy. That’s the core of what has allowed us to go so quickly while maintaining design quality—we don’t want to be distracted until we’re sure we can afford it.
X: What is the key to profitability in the travel search business? Is it simply a matter of acquiring as many users as you can? Or do you need to keep branching out into more products with higher commission rates?
AG: You can do it in either of those ways, but it’s easiest if you’ve got scale. It’s extraordinarily difficult to make a profit based just on selling flights. There are companies that have done it, but only by capturing five percent or more of the total U.S. flight market. So if you are hoping to make it without having to raise $100 million and blast out to TV sets across the country, like Kayak, then you have to have some mix of flights and hotels. Cars are nice, but not mandatory. There are fixed costs and variable costs with all of this—the more searches people do, the more money you are paying [for data]. So there is a hump you have to get over to reach profitability, but once you are there you become increasingly profitable with each additional user.
X: So, how do you acquire users without spending $100 million on TV advertising?
AG: That is what I think about all the time, really. It is the question on my mind more than any other one: In this age of the Internet and especially mobile devices, what combination of factors does it take to get broad adoption without having to pay multiple dollars per person to acquire users? Generally speaking, the solution we have worked off of is just word of mouth, but given social media there is a certain speed at which word of mouth is possible. On the one hand have a lot more things vying for people’s attention, and on the other hand you have a lot more sharing. It’s not clear to me which is the more powerful force. But our company is a bet that the power of word of mouth and free products will end up mitigating the need for some of the more expensive techniques.
X: You’ve got the same basic flight and hotel data as everyone else; my sense is that it’s the user experience and the design sensibility that sets Hipmunk apart. Agreed?
AG: All of the companies in online travel right now are essentially glorified marketing companies. The reason is that they are providing, essentially, a commoditized product. You are not going to find meaningfully different prices or experiences on most of them. So the only way they can really grow is by stealing market share from other players, and the way they do that is by yelling louder in more places. If you spend too much time in that mindset, I think you lose creativity. You start to think of our business not in terms of the products you’re building, but the way you’re acquiring users. Our goal, as we grow and start to think about some of these large-scale customer acquisition challenges, is to not become the kind of travel company that most others have become, obsessed with obtaining every last cent and diverting creativity from product into acquisition. That’s the reason we have been able to find success so far.