Nirav Tolia Cofounder, Epinions

Jeffrey Rayport: How does somebody at one of the hottest companies of the new economy, otherwise known as Yahoo, leave Yahoo to go and start something that you heard about on a Friday afternoon, and you were gone on a Monday morning?

Nirav Tolia: It's actually pretty difficult to do that, and sometimes I wonder if I made the right decision. No, seriously, I actually had a great time at Yahoo. I was there relatively early; I was one of the first 80 employees, joined just after they went public, and had some amazing times there. As you mentioned, it's an incredibly hot company, it's a fun place to work, they're great people, and it continues to be successful. I was actually so inspired by my experience at Yahoo that, frankly, I wanted to try to do it myself. So it actually wasn't as flippant as you make it out to be. I didn't hear the idea on a Friday and suddenly quit on a Monday. I had been thinking about being an entrepreneur and trying to branch out and do my own thing. So when I heard the right idea, you're right — it was like a light switch; it just got flipped.

Rayport: But you were waiting to find something that you could light onto as an entrepreneur in your own right.

Tolia: I wasn't sort of desperately looking for the right reason to exit. However, I was open to being inspired, if you will. I find that a lot of entrepreneurs in this economy are just inspired. They're passionate about what they're doing. So I wasn't thinking to myself, I need a better job, I need to make more money. It was nothing like that. I was thinking, hmm, I've seen David and Jerry, the founders of Yahoo, create this incredible company, it now has over 2,000 employees. Could I actually try to do the same sort of thing with a group of my friends the way they did? So when I heard this idea for Epinions, it immediately resonated with me. It sort of hit all the things I was looking for in an idea, so I took the plunge.

Rayport: Where were you when you heard about it? And what was it that flipped all those switches and you said, "This is it"?

Tolia: Yeah, it's actually a little bit of a longer story. When I was at Yahoo, I was lucky enough to cofound a nonprofit in Silicon Valley. The nonprofit is called Round Zero, and Round Zero actually refers to what entrepreneurs are before they get round-one financing — basically entrepreneurs without any money. It brings together entrepreneurs once a month to talk about the industry, to talk about how to get financing, to talk about the things that are hot. And I had met all of these people at Round Zero over the three years that I'd founded it. Many of them had come to me and said, "Let's start a company together. Let's actually go build something ourselves." And I always said, "No, I'm in Yahoo. It's a great place to be. I'm learning so much; I want to stay."

One particular guy I met at Round Zero, a friend of mine, Naval Ravikant, who's actually a cofounder of Epinions, had told me at a particular Round Zero event, "I have a great idea for a company. We have to do something together." And I sort of looked at him like, "OK, I've heard this pitch before." I actually said, "Before I even hear the idea, I just want to spend some time together. Because if we start a company together, we're going to be spending 20 hours a day together for months and years. So for the next couple of weeks, let's just hang out, have a good time."

Rayport: You mean, forget the idea for now, let's just get to know one another.

Tolia: That's right. Let's go out for pizza, let's talk about who our favorite football team is, etc. Let's just see if we can coexist as friends and colleagues. So I did that with Naval for a couple of weeks, and we actually didn't kill each other. We actually found a lot of common ground. Then one night, he shared the idea with me. It was actually a Saturday night, and he put this thing up on the whiteboard — it was the concept for Epinions. And I was looking for something that had a lot of potential; that was definitely on my list of things to look for, I was looking for something that was very consumer-facing. Being at Yahoo, one of the great benefits is, you walk down the street and say, "I work at Yahoo," and people smile. They light up, they say, "I love Yahoo, I use it all the time." What a great name. I was looking for that same sort of emotional connection with consumers. And I felt like Epinions had that potential.

And then, finally, I was looking for a real business. When you start a company, it's not just a hobby. It's a real business. You have to have revenue; you have to be held accountable to your investors; you have to be profitable. So those were the three things I was looking for. Something really big, something that was very, very specifically consumer-facing, and something that was a real company.

So Naval shared this idea with me, and I immediately resonated with it. I started thinking in my own mind, wow, this is something I could really do, leave Yahoo, it's a pretty radical thing to do. So I stayed up all night, pulled an all-nighter, called my father in the morning. Said, "Dad, I have some good news and I have some bad news. The good news is, I've actually found something that I'm incredible passionate about, I want to do. The bad news is I'm actually going to leave Yahoo." Of course, my dad being a conservative physician in Odessa, Texas, where he lives, said, "Well, gosh, you're vesting, could you maybe wait a couple of months or maybe a year or two, and then after you're done vesting, you could leave?" And I said, "I've got to do it now." So I went into work one day later, on Monday, and started sharing the news with people that I was leaving.

Rayport: People must have thought you were crazy walking away from all these invested options sitting on the table. Again, such a successful company.

Tolia: Well, I left during a time when it was becoming more and more accepted to take risks. So I left Yahoo in April of 1999, and that was sort of the height of the funding cycle that we saw. And people were actually encouraged to leave secured jobs, and leave these nice old-economy type positions, and go take chances. So yes, many people looked at me like, "Do you really know what you're doing? Is this really the right opportunity? Gosh, you have such a great gig." But others were actually pretty supportive. And without my parents' worry, I think I launched into this feeling pretty good about myself.

Jeffrey Rayport: Epinions is famous for having gone zero to 60 very, very fast. Measured in weeks, not months. How did you do that? How do you put a company together in four, five weeks to the point where you had funding, office space, and people? That you had developed the concept, and all of a sudden, here it was.

Nirav Tolia: Well, we worked really hard, but I mean to be serious. I don't know if speed is the one thing I would concentrate on if I were starting Epinions again. But once again, moving back to April, May of 1999 when we founded the company, we felt that first-mover advantage, the concept of being the first person in the market to offer a specific benefit, we felt that was very important as Epinions. So we started, we worked around the clock, we were obsessed with speed, this concept of instant company, this concept of hiring as many people as possible and then putting something up on the market, actually having our website up and live in 12 or 20 weeks after we founded.

So it actually ended up being a great deal of support from our financiers. So we actually were funded by Benchmark Capital and August Capital, two of the top VC firms on Sand Hill Road in Silicon Valley. They helped us quite a bit, because they already have a lot of infrastructure built out. They know where to go to find office space. They know where to go to get office furniture. They know what legal things you need to do. So there's sort of an operational part of it, which is just setting up the company, and then there's a product piece of it. Believe it or not, the various founders had been thinking about Epinions in different ways for the past couple of months. Even though we hadn't quit our jobs, we had been thinking, "OK, if we did this thing, Epinions, what would it look like?" So it wasn't as if we founded the company, and then we had to start thinking about what we wanted to create. It was really just, "OK, we've got our funding, the money's in the bank, let's just run." And we ran as hard as we could, we worked incredibly hard, and as I mentioned, we were obsessed with speed.

Rayport: People talk about speed as one of the hallmarks of what's new about the new economy. What were some of the downsides to speed?

Tolia: Well, I think the fundamental observation is, it's not just OK to be first mover in any market. It's fundamental to be first mover, done right. So you can actually learn this if you study the history of the Internet. Yahoo was not the first portal. It was not the first website directory.

Rayport: Nor was Amazon.com the first bookstore.

Tolia: Amazon was not the first book site; eBay was not the first auction site. Over and over and over again, you see these examples. And you learn that it's not about just being first, it's about being first, right. So you asked what were some of my regrets, or what I would do differently. I would probably get a few parts of the service that weren't completely right when we launched, I would actually get those perfect before launching. So there are a few things that you need to do, clearly better than anyone else, to get consumers on your side. Because on the Internet, the first time a consumer comes to a website, that's the only experience they've had. And if they have a bad experience, they never have to come back. It's so easy to just click away to another website; that first impression is very important.

So when you're obsessed with speed, you put that first impression up, and if it's not as valuable as your real value proposition, you could run into problems. So I think I would have taken a little more time, I would have taken more time in hiring, because we actually hired about 60 people in the space of three or four months and got really big really fast without knowing, do we need this many people? Is this the right model? And I would have just probably crossed the t's and dotted the i's a little more carefully.

Rayport: What feature on the site, or what aspect of the user experience, did you regret most?

Tolia: Well, the biggest thing is creating a site so quickly, we couldn't do hours and hours of user testing. So we were essentially creating a website that we felt would be useful for ourselves. We didn't know, "OK, if we like it, does that mean that everyone else is going to like it as well?" So the one thing that I would do differently is, I would literally spend weeks having users come in, and look at a beta test or prototype of the site, and tell us what they liked, and didn't like, and then based on that feedback, I would then iterate on the website, and then release it once I was sure that the market was actually going to respond positively to it.

Jeffrey Rayport: Could you have done what you did in the spring of '99 in today's environment?

Nirav Tolia: Well, they say that, you know, hindsight is 20/20. So I don't really even remember what the summer of '99 was about, because we were working so hard and it was just sort of this big blur. I think the biggest difference between the market then and the market now is the amount of money you can raise. So for example, when we started Epinions, we didn't have a real business plan. We had a great concept, we had what we felt was a very strong team, and we had 16 slides on a PowerPoint presentation. We presented that to the VCs, and were able to raise $8 million. We weren't exactly sure where the revenue was going to come from, we weren't exactly sure what the value proposition or endgame for Epinions was. We didn't even really know how big the market was. Today, I think you need to have answers to all those questions. So whether or not you can complete a website as quickly as we did, I think that you can probably do. But getting the funding, and maybe even hiring people, because in general people are a little more skeptical to just jump from their established jobs to dot-coms. I think those two things would be very difficult to do today.

Rayport: Everyone today is worried about B2C businesses, business-to-consumer businesses, which is, of course, what excited you about this idea.

Tolia: Right.

Rayport: It's consumer-facing, it could become a household word, it could change the way people live, the way they buy stuff. The big rap on B2C businesses is, where's the economic model?

Tolia: Right.

Rayport: So when you discovered the business model, what did it turn out to be?

Tolia: It's interesting. I think the first note is that I find people fall prey to hype far too easily, whether it's positive hype or negative hype. Speaking very frankly, we actually benefited from a lot of the positive hype. We had lots of press around the launch of Epinions. As I mentioned, we were able to raise the $8 million. But people were a little unrealistic in terms of the potential for B2C. Now I think they're unrealistic about the lack of potential for B2C. So the hype thing actually works both ways. You can actually fall prey to positive hype; you can fall prey to negative hype.

I think the bottom line on the B2C side, in terms of generating revenue, and it turns out we have a B2B revenue stream as well, and whether that's because of the climate or because that's where our value is, we're not sure yet. The real value on the B2C side is scale. So you look at large B2C companies — Yahoo, AOL — they have many, many users, so when they monetize those users, they're not selling them anything. It's actually free to use Yahoo. If you go to AOL.com, it's free, although it's not free to use their dial-up service. You need millions and millions of users coming into your site to actually monetize them correctly, using advertising.

Rayport: Meaning that if you're selling advertising at a cost-per-thousand — and the cost-per-thousands are in the single-digit dollars for most consumer-facing sites — you'd better have a lot of users.

Tolia: That's exactly right. To illustrate by analogy, if you're a TV channel — TV is another media that relies on advertising — or a TV show, and you only have five or 10 viewers, you won't make money. Because what advertisers pay for is the access to many, many consumers. So think the fundamental observation I've made with B2C sites is you really need scale in numbers. You need lots of users coming to the site to actually capture money. We have 2.5 million users coming to the site every single month. I'm not sure that's enough to actually get profitable. You need to get to 5 million or 10 million. Yahoo's at 150 million users. So all of the websites out there that assume that they're able to get to 10 million users and 20 million users, those ones are going to have a hard time.

What we've found is that it's easy to get to the first million users; it's a little more difficult to get to the next million. So when you're 2 million you're feeling pretty good and you're thinking, yeah, I can get there. Getting from 2 million to 3 million, incredibly hard. From 3 million to 5 million, even harder. So as you gain more audience, it's more and more difficult, but you need to do so in order to monetize the business to consumer side.

Jeffrey Rayport: Nirav, you're describing a dynamic which on the one hand you've seen, it's empirical data. On the other hand, doesn't it fly in the face of all the early promise of increasing returns models on the Web? Where actually getting to the first million was tough and expensive, but then the next million started to look like a higher-margin business, and in fact because of viral effects and all the rest of it, network dynamics, you'd build that curve very fast.

Nirav Tolia: I think increasing returns is still a principle that applies to the Internet. The problem is that people don't know where that inflection point is going to occur. So you have to keep on working, keep on working. Maybe it's a million users, maybe it's 2 million, maybe it's 5. Getting there is very difficult. The example I'll give is eBay, which is considered an increasing-returns type model. In four years, eBay went from zero users to 4 million users. That's fairly respectable, but what we call organic growth. It took time. It has actually been quicker for us to get to 2.5 million users than it did for eBay. However, in Year 4, eBay went from 4 million users to 15 million, because they hit an inflection point. When they hit an inflection point, they went into increasing returns, and they experienced what's called non-linear growth.

So I think increasing returns and being out there and having the viral growth and all the benefits of scale, those are all true things, but don't get caught into the negative or positive hype that it will take a million users to get there or 2 million or 3 million. We just don't know when these systems really explode. Microsoft didn't know when it sold its 1 millionth copy of Windows that it would suddenly have a lock on the market. You're just sort of doing as much as you can, building your business. Then, when increasing-returns models hit, you just aren't even prepared for it. All of a sudden you hit an inflection point, and things explode. So I still believe in that concept. We just don't know where those inflection points are.

Rayport: You mentioned the need for speed. Of course there's been a lot of attention because of the book by Malcolm Gladwell, The Tipping Point, another term for the inflection point you're describing. The only reason that your or any business might not get to that inflection or tipping point is that somebody else captures those returns.

Tolia: Or you run out of money, in this market.

Rayport: Right, you can't walk far enough to actually hit that wonderful end of the rainbow. There are other companies: Deja.com, PlanetFeedback.com. How do you think about positioning Epinions and continuing this, as you said, laborious growth as you march toward the inflection point, vis-a-vis competition.

Tolia: Well, I learned a wonderful lesson at Yahoo. That is, you can't influence the external part of your company. You can't concentrate on your competitors. You can't do anything about market conditions. What you can do is build a great product, hire incredible people, and continue to deliver to your users, over and over and over again. I think if you study Yahoo, you'll find that in the early days, when it was Yahoo vs. Excite vs. Infoseek vs. Lycos vs. everyone else, Yahoo didn't really get into this sniping: How are we better than Excite, how are we better than Infoseek and Lycos. They just continued to deliver value to the consumer.

For Epinions, I think we worry a lot more about just creating the best experience at our website. To get specifically to your point, we believe that there are fundamental parts of our model that are different than Deja or PlanetFeedback or anyone else that's really in the space. Those actually create what we call a marketplace for content. What we've tried to create is as close to eBay as we can. EBay is a marketplace for commerce, where there are buyers and sellers of goods. Epinions is a marketplace for content. There are buyers of content — those are the readers who come to read the reviews on our site. There are sellers of content — the review writers, the ones who provide content on our site.

None of the people you mentioned, or really anyone in the space, is really trying to build a marketplace. Many people are looking at message boards, and other people are looking at feedback mechanisms. But no one has really created that true marketplace for content, the way eBay has created the marketplace for commerce. As far as we're concerned, we're just going to continue to try to build that marketplace because we know that if we're successful, we'll hit that inflection point, that tipping point, and we'll have the same sort of returns that eBay has.

Jeffrey Rayport: Let me understand how that market actually works. Because I know that people can go there, read reviews. So the buyers of content are, in effect, getting freeware, which is a classic Web strategy.

Nirav Tolia: Not really freeware, because you have to pay with your time, but we can get to that in a second.

Rayport: Fair enough. But the same would be true of a free PC model or the Netscape Navigator giveaway and so forth. So we need to talk about that. But on the supplier side, people are writing reviews. You're actually paying them with what you call royalties, of a sort, in order to make their time worthwhile to contribute to this common good.

Tolia: Yeah. We fundamentally believe that everybody is an expert at something, right? There are some things that you know about better than anyone else. Now, if we can take that expertise out of your head, put it on our website, and share it with everyone else who wants that information, you should be compensated for that. You should be compensated by people coming up and saying, "We respect your opinion, well-done, you're a featured reviewer, you're an expert reviewer, you're an adviser." You should also be paid in cash. It's just like the real world. Siskel and Ebert always got paid when they did their movie reviews, the same way when you come to Epinions, you should expect if you're providing a service and sharing expertise, to be paid.

On the economic side, paying reviewers is very important because in building a marketplace, you need to have three elements. Liquidity, which just means lots of buyers and sellers, and in our market, that means lots of readers and reviewers — that just means traffic, lots of users. The second thing you need is currency. In any marketplace, money has to change hands; transactions have to occur. If transactions don't occur, it's not a marketplace. So in our system, when a reviewer comes up and creates content and hands it to us, we need a transaction to occur where we pay that reviewer in money. And so that actually works on our system. The third part of the marketplace is transparency. You need to see who you're dealing with so you feel trust when you're actually engaging in these transactions. We have a lot of transparencies on our site. In fact, we feel that the content that's provided by reviewers is secondary to the context of the reviewer himself.

The example I'll give is that when you want advice on a cellphone, you don't just run out in the middle of the street, stop a car, and say, "Excuse me, can you tell me what cellphone to buy?" You go to the person that you believe knows a lot about cellphones. The same way on our site, when you're reading reviews, you're not reading reviews from random people. You're reading reviews from people who've said, "I'm a cellphone expert, whether it's because I sell them, own them, or study them." That's a really important thing.

Getting back to paying reviewers, a lot of people say, "Well, how much do you pay? Because it's one thing to say that you really pay reviewers, but is it really a lot of money?" It turns out that several of our reviewers have quit their full-time jobs to write full time on Epinions, because they're making thousands of dollars a month helping other consumers make buying decisions. On the economic side of the model once again, we don't pay more for our reviews than we generate in revenue. So unlike a lot of these zero-margin models, where they're giving away a dollar and asking 85 cents for it, we don't pay more to our reviewers than we generate as a company. So the model is working.

Rayport: How does it work? I come to the site, and I find that cellphone expert, and I find that she's written a review of exactly the phone that I want, I read the review, Epinions knows of course that I've consumed that review, and now compensates her for having written it. How do you actually create value out of me, other than on a CPM basis, cost-per-thousand basis because I'm one of the people spending time at the site, so you can sell advertising. But as you said before, the scale may never hit, or it may never be a robust enough revenue stream as a lot of people think about consumer-facing Web businesses today, to sustain Epinions.

Tolia: Yeah, let's talk about two things. One is, how exactly do the reviewers make money? Because I think that's interesting as well. But let me answer your question first, which was, how do we capture value from the readers or consumers of reviews? Now, using the example you talked about, you come in, you read a review on a cellphone, we may monetize your time by flashing a banner ad or a sponsorship. But that's not very interesting to us. The way we monetize you is, after you've read the review, there's a little link, and it says, "Where to Buy." So you've decided, "Hey, I've read this review on a cellphone, I actually want to purchase this or purchase service. Where do I go now?"

We will lead you to the right place to actually consummate what we call a decision or a transaction. We lead you to someone else to fulfill that, because we're not like Amazon. We don't have warehouses; we don't actually sell you anything ourselves. We lead you to people who do sell you things. Once we lead you to those people, we get paid every time one of those leads is generated. This is actually a very similar business model to CNet's business model in computers. CNet's business is all about giving you the right information so that you make the right computer purchase. And every time you decide to make one of those purchases, CNet will capture some of the value. Our site works the same way.

Rayport: An affiliate marketing model.

Tolia: Exactly. It can be called affiliate marketing. Now, what's interesting about affiliate marketing, however, is that it's usually tied to a transaction. So with basic affiliate marketing, anyone can create an affiliate site for Amazon, where you come to my personal website, you buy books that I've got listed from Amazon, and every time you buy one, I get paid something. Our model's a little bit different because it's not tied to an actual transaction.

Rayport: Meaning all I need to do is click through to the place where I'm going to buy the stuff, and whether I buy the stuff or not, Epinions gets paid.

Tolia: Yeah, that's right. Because, frankly, we can't determine whether or not you will buy. You may go to a retail site and have a bad experience, or you may actually not want to buy at that point in time. There may be a whole host of other things. The other thing is that we may actually have products and services on the site that aren't available in the online world, right? So, for example, it's still pretty difficult to buy cellphone service in the online world. What isn't difficult, however, is us introducing you to a place in your locality where you can actually buy cellphone service. So we need to capture value on those sorts of transactions as well. The way we do that is on a lead basis rather than a transaction basis.

Rayport: You mentioned earlier that there was a business-to-business aspect of what you do. Is that it?

Tolia: That isn't it. That's still what we call consumer facing. And, really, that's monetizing our portal. When I think about our business, I think, there's a way we can monetize the portal: That's through advertising, generating leads, transactions, all the things we're talking about. And then there are non-portal ways that we can monetize the value we created. There are actually two pieces of value that we're monetizing away from the portal. The first is, in the last year, since our site's been up, we've actually generated over a million pieces of content. So we now have over a million reviews that have been rated and ranked and sorted. Average review length is 300 words, so these aren't just one-liners saying you should buy this or you shouldn't buy this. These are very well thought out, almost professional in quality, user-generated pieces of content. We can take that content and syndicate it to other sites, in much the same way that Reuters syndicates its content.

So, for example, if you're a retailer, and you want user reviews on your site, but you don't want to send them to Epinions, we can send you a feed. It's a very easy process for us, and we essentially send you the content, which you can then put on your website. So anyone who wants user-generated content can come to us. Because now, frankly, we're the No. 1 provider of this sort of content. We get 3,000 to 5,000 new reviews every single day. Just to give you some perspective, we had more restaurant reviews than Zagat. We have more automobile reviews created every day than every other automobile resource combined. We have amazing reviews — we have cellphone reviews, we have cellphone-service reviews. Pretty much anything you can think of, we have reviews on. And people actually want to buy that content. So that's the first B2B component.

The second component is, you may ask, how you get all of this content? We've created what we feel is actually a pretty robust engine for user-generated content. And the engine isn't just a place where you come and you write a comment and bang, it's posted. It's an engine where you write content, other consumers rate the quality of that content, you get a quality rating, it's posted in real time. And, actually, of our million pieces of content, 99 percent have been rated five-plus times for their quality, not by anyone at Epinions, but by our community at large, our users. So even though we get 3,000 to 5,000 new reviews a day, we don't have one in-house editor. It's real time posting, it automatically gets put on the website, and the users come in and determine that this is a good piece of content or a bad piece of content. Now that functionality — forget the content that's generated — is something that a lot of other websites want.

Rayport: So a technology-licensing business.

Tolia: That's exactly right. It's more like an ASP, what's called an application service provider business. The analogy I'll draw is Inktomi provides shopping services for many portals. We can now layer onto those shopping services and provide user-generated content for reviews, ratings, advice, you name it, for a variety of sites out there from retailers to manufacturing sites, to any affinity site. So, for example, if the American Medical Association wanted a way for users to come up and comment on drugs or treatments or diseases and conditions, they wouldn't want just open message boards, because you know message boards don't work. It's just all noise; there's no structure, there's no feedback. They would actually want to take Epinions' system — not any of the content, because they don't want book reviews or car reviews. They would take the system, layer it on top of their website, and all of a sudden they would have a robust way for their community of users to create content on their site.

That's actually what we believe is a huge opportunity. Not only can you generate more revenue per deal — we're not talking about the tens of thousands of dollars you make on advertising; we're talking millions of dollars per deal. There's a huge benefit being an infrastructure player, compared to a B2C player.

Jeffrey Rayport: Absolutely. Final question. In 10 words or less, what is the next big thing?

Nirav Tolia: Hmmm, that's a very, very interesting question. I think the next big thing, from a market perspective, is consolidation. I think for a long time people have believed that there would be millions, and if not millions, thousands and hundreds of thousands of websites out there. I think that's going to be true, but in the business world, I truly believe that many rollouts are going to occur. So the Internet is a place where being No. 1 is way better than being No. 2. And if you're anything less than No. 2, you probably don't survive. So all of these different spaces, whether it's the portal or auction space, the online advice space with Epinions, all of these different spaces are not going to have multiple competitors. The companies that you hear about that survive, are going to actually own their markets. I think there are going to be very few sites out there that are capturing all the value. If you look at the top-rated sites on the Internet today, I think the top 20 sites capture 90 percent of the total traffic. I think next year you'll see those top 20 sites starting to capture 99 percent of all of the traffic. I think it's a natural thing that happens in most mediums.