Kayak takes on the big dogs

September 27, 2012, 9:00 AM UTC

Kayak founder Paul English

FORTUNE — A successful July IPO drew attention to the surprising story of Kayak (KYAK), an online travel site that’s succeeding where you’d least expect it. Launched in 2005, it faced established competitors like Expedia (EXPE), Orbitz (OWW), Travelocity, and Priceline (PCLN), and Google (GOOG) and Microsoft (MSFT) have since lumbered into the business. Yet Kayak is profitable and growing, using a model that aggregates results from those competitors plus airlines, hotels, and car rental companies to give customers maximum choice. Paul English, 48, is Kayak’s co-founder and chief technology officer as well as an unofficial recruiting machine and customer champion. He had worked at Intuit (INTU) after selling that company a software business. He talked recently with Fortune’s Geoff Colvin about succeeding against giants, hiring people who aren’t looking for jobs, and much else. Edited excerpts:

Q: Google recently bought Frommer’s, the travel guide business. Two years ago it bought ITA, the source of flight data in the U.S. It also offers flight and hotel search. How does Kayak compete with a giant like Google — or for that matter like Microsoft, which is also in this business?

A: Or you could say with a company like Expedia, the No. 1 travel company in the world right now. The main advantage we have over Google is extreme focus. What we do at Kayak is very simple. We help people find flights and hotels and cars. By focusing on that, we’ve been able to create a product that consumers have really loved. Most of our traffic is self-directed. It doesn’t come from Google. People just really prefer the product.

Most companies say they’re customer focused. How does that manifest itself in real day-to-day business at Kayak?

If you were to look at other companies in the travel space, one of the things that’s different about Kayak is that most of our staff are on the tech side. We really invest in product design and development because at the end of the day, that’s the asset we have. We’re building this brand, Kayak, but the brand ultimately comes from the products. If you were to compare Kayak to other companies you mentioned, you would notice first that we have monitors all over the walls that show live data about what’s happening with our customers — the latest feedback coming in, feeds from Twitter, how many searches a day we’re getting in each country. Everything we do is very focused on what’s happening today with the customer. There’s also a sense of urgency in the company. Most of our meetings have maybe three people, are well under an hour, and are incredibly action-driven. We give Kayak employees kudos for getting something done really, really quickly, even if we have no time to perfect it, because ultimately the customers are going to help us refine what’s working and not working.

When you say most of the people are on the tech side, does that mean you don’t have separate customer support people?

Yeah, our product support is actually done by myself and the engineers. I know that sounds ludicrous because we have millions of people coming to the website every day, and in Boston, which is the largest office we have, where all the engineers are, we only have 100 people. But there’s some magic there.

So if a customer calls up with some kind of issue, it’s going to be a tech person giving an answer.

It’s often me. You may have heard about the infamous red phone.

The red phone with a loud ring.

With a loud ring, yeah, because we sit in an open area. No offices. I want to make sure that when a customer is having an issue with our product, it’s visible to the entire team. As to how 100 engineers can support millions of people a day, it’s a bit of a tautology. From day one I made the programmers do the support, and that means when a customer calls and yells at us because we’ve screwed something up on the site, by the second or third time we get that criticism, the programmer is tired of answering the same question. So they stop what they’re doing, they fix the code, and we don’t get that question anymore. It’s a very, very fast cycle. When customers find problems, we try to fix them instantly.

The customers are incredibly loyal. Something like 75% of them come directly to Kayak.com.

We don’t get a ton of traffic by people going to Google and typing “cheap hotels in New York.” We get traffic because people go to Kayak.com or use our mobile app, and they go back because they like it or their friends have told them about it. Because we focus on the things that we do very well, a lot of people get a lot of comfort with the site, and they then come back.

As an e-commerce business you can conduct experiments to test hypotheses. Do you do a lot of that?

Yeah. At any moment we have a few dozen versions of Kayak running. If you and a friend go on Kayak today, there’ll be very, very subtle differences in your user experiences. One might use this color blue and another one might use a slightly different color. On this one we might have the hotel stars shown to the left of the name and on that one we might show them above the name. We’re always refining it to make the display easy to parse for the users, to get the clutter out of the way and make sure we have the information they want, and nothing else. Also, I might have a new idea that I’m excited about, and my co-founder Steve Hafner might have an idea he’s real excited about. We’ll put it up on the website, and if it doesn’t work well, we throw it out.

You’re well known in the company and in the industry as a ferocious recruiter. Where does that come from?

It’s something I’m pretty obsessed with. The joke at Kayak is, if we have a business trip out of San Francisco, when the plane lands, my colleagues will say, “How many people did you hire on this flight?”

I gather you don’t wait for someone to become available.

I think most people hired at Kayak were not looking for jobs. We also don’t look for specific skills. If you go to our job site, we don’t have job descriptions. We say that if you’re known as a team leader and someone who was the most successful in your sales job or product design or whatever you did, we want to talk to you. I frequently create jobs for people. When I meet someone who’s incredibly talented, I’ll try to figure out their talents, their interests, and we try to figure out how can we use that at Kayak. It’s a pretty aggressive process. As soon as I hear someone’s name, I have seven days to make an offer to them. There’s a lot of work in those seven days. I have to convince them to meet me, and I’ll go wherever I need to. Background check. Get them in for the first round with four people at Kayak. Another four people the second day, and then make an offer.

You’ve got a big operation in Concord, Massachusetts, which is a lovely spot but an unlikely corporate site. How did that happen?

It wouldn’t necessarily be my first choice for the tech team. I think my first choice probably would have been Cambridge, right around MIT, where we have a lot of great engineers. The reason we ended up a bit west of that is actually a recruiting story.

When I met Steve Hafner and he gave me the pitch for what he wanted to do in travel, I became pretty excited about it. We probably spent an hour together at Legal Sea Foods in Harvard Square and decided to work on this together. The first thing I did when I left — I was in my car — was call up the two strongest engineers I’d ever met, guys I’d worked with at a few companies in a row. I said, “I’m starting another company. I really want you guys to come work with me again.” These guys were very well paid at a great, great California-based company, and they asked about compensation. I told them what I would pay them, which was substantially lower than what they were making, and I think the third question was, “Where will the company be located?” I knew I needed these guys. They’re incredibly talented, and they’re also great recruiters — people want to work with these guys. So I pretty much said, “You guys can pick. Wherever you want, I’ll put the company.” And they initially chose Maynard, Massachusetts, which is also a nice town but not where I would center a brand-new tech startup.

You talked them back to Concord?

Yeah, and I got them. They’re still with me today, eight years later. And they’re fired up. We’re just having a blast.

The Leadership series This is the latest interview with a top executive by Fortune senior editor-at-large Geoff Colvin. See video excerpts of this interview at fortune.com/leadership — plus find Colvin interviews with GM’s Dan Akerson, Dow CEO Andrew Liveris, Chicago Mayor Rahm Emanuel, Southern Co.’s Thomas Fanning, and many more.

This story is from the October 8, 2012 issue of Fortune.