Rohin Dhar uses blogging to hit more than 2 million monthly visits and make money at Hustle Con

So I'm Rovan, I'm the founder and CEO of Priceonomics, and today I'm just gonna tell you guys a little bit about, you know, how we grew to be 2 million visitors a month and how we started to figure out how to make money from basically just blogging. So what is Priceonomics? If you talk to almost anyone, what we're known for is being a really big, interesting blog. Um, we're a team of half writers, half engineers, and we had an observation in the market. Like, newspapers are dying, blogs are like being able to like print out hundreds of millions of page views a month, but all of them just make money from ads. So they have all this traffic, but they don't really know how to make money. So that's like one half of the world, just content sites that are all dying. And then the other half of the world are Silicon Valley startups. And these people already know how to sell software for a lot of money. But their problem is they don't really know how to get traffic. When you start your company, that's going to be the hardest thing. How do you get users? How do you get distribution? How do you get traffic? And so our observation, what made the difference for us, is we said, well, what if we combined the two? What if we became a really, really great content site that basically used that content to occasionally promote our own services? And that's what we did at Pressinomics. Our main product is we crawl data for hedge funds and tech companies, and they pay us between $2,000 to $10,000 a month to get them a feed of that data every month. And the one, the only way we get customers is by writing about data and attracting people that way. So this talk today is gonna be about basically how do you create great content like that that can power your whole business. But I think this will be a pretty honest conversation. We'll try to be honest. I think a lot of speakers will be honest with you about how do you get there. And a lot of times when you read TechCrunch or you read VentureBeat, it sounds like it's this great linear path. We came up with this idea and then we were successful. It took us a long time to get there. So let's start about the history of Priceonomics. Part of the reason our name is Priceonomics, even though it probably doesn't have to be, was I had this idea. I sort of like love getting great deals on things. And I found a bike I really liked on Craigslist. I bought it for $300. And it turned out I didn't really like it after like a week. So I put it on Craigslist again, and I sold it for $1,000 a week later. And I was like, this is not how a market should really work. There should be really great price transparency about how much a bike is. This is ridiculous. So I told some friends. We built a team together. And we're like, well, let's just crawl through Craigslist and eBay and see how much these bikes are selling for, or iPhones are selling for, or anything are selling for. So we built these crawlers. We built this website. And it was going to be like Kelly Blue Book for everything. And it was called "Pricenomics." And we built this thing and it was sort of like a side project. All of us were doing other things. I had started a different company years before. I was halfway running that. And we were like, "Well, this sounds like a cool idea. Why don't we apply to Y Combinator?" And then, you know, to our surprise, we got in. And it was sort of like, "Oh crap, now we have to turn this into a company." It wasn't like we were like so passionate about being a price guide. But now we were like, stop doing this. And so we got into Y Combinator and we're like, well, the one thing they tell you there is you have to figure out a way to grow. And early on, basically the only way we were getting users was from Google. People would Google, like, how much is my Trek 5000 bike worth? And there weren't a lot of results for that because, as my example from arbitraging bikes tells you, there was no data about that. So we were showing up and it was going pretty well. And we're like, well, how can we grow faster? Well, we need more links. We need people to cover us. We need, basically, to do well in Google, you need a lot of content, you need people to link to you so you have a lot of authority. So we're like, all right, well, let's start a blog. And we can write about our data. The OKCupid blog was really huge then. And we thought, well, let's just do what they did and write about data. And I'll sort of like talk you through a little bit about like how we started off and how it was sort of advantageous, some of the things we did in the beginning that accidentally led us to go down the path of mostly blogging. And like we did that and it worked pretty well. And we started like going up the ranks in Google and we were in YC. And we managed to raise like, you know, over $1 million in funding afterwards from great angel investors. From the outside world, it looked like this sort of perfect startup. You're getting to live the dream. But internally, like, every day we'd look at the numbers, and it was all driven by Google. Like, other than our blog, all the users were just coming in from Google. And then one day, like, it would go down 20%. And then it might go up 10%. Or it might go down 50%. Or it might go up 100%. But it really, like, eventually we figured out that, like, These aren't our users. These are Google's users. And eventually, are we going to be able to build a real business just off the back of Google? We have to think about something else. And so while we were doing this, through being a Y Combinator and being in Silicon Valley, you just get to meet a lot of other people that are doing startups. People would always come to us and ask, hey, we're trying to crawl these enormous websites. That's all you guys do is crawling through these sites that have so much data, how do you get them? Can you help us with this? We were always like, all right, yeah, we'll have coffee with you, tell you about this. Eventually someone was like, well, can we pay you for this data you're getting? And we're like, no, we don't want that. Crazy, we're a price guy. And then eventually we're like, well, I guess, would you pay us $2,000 a month? We just threw out a really absurdly high number. Say like $200 a month. And they said, "Yeah, okay." And we're like, "Oh, okay, well, that's like way more money than we were making from this like maintaining this site, and that's just one customer." So that's like when everything clicked. We started to have this business model in place, so we're like, we'd write great content that sometimes involved crawling data from the web, people would find out about us, a very, very small percentage of them would like contact us and say, "Hey, we need this sort of data." And those people we would charge anywhere between $2,000 to $10,000 a month. And then once we like figured out that model, then it was a little bit more repeatable and we could write more awesome stuff, we could sell more stuff. And so that's sort of like the story about how we got here. And like, there's so many times in that little story I told you where we just totally could have died, you know. But the one thing that always like stuck with people was that they really liked our blog. And that was like the one thing that we always stoked. We said, "We gotta write really interesting content. We gotta find cool data. We gotta share it. We gotta like stoke this thing that people love." And that ended up saving us and letting us like figure out and try new things that we could like sell to our audience. And that's where we are today. So this talk is sort of aimed at you who's just trying to start a company and who wants to make great content so you can get like in the newspaper or TechCrunch or wherever, so that, like, people can find out about you, so you can market your product, so you can test out new things, and so you can get distribution. And, like, as you're thinking about how do you market your startup or how do you write a blog post that, like, promotes yourself, just remember that no one cares. Like, and don't feel bad about that. Like, when you email someone, Like, they're busy doing their job. They have like a million other things. They don't care about like what you just emailed them. They probably don't have the mental overhead to even process it. And like, don't take it personally. Just like, no one cares. So you're gonna have to just really, really stand out and make other— make people like demand to care. Like that you are so special that like you help them do their job or you're just so interesting that they care. And that's really the only way to get through, in my opinion. And so we have one philosophy about marketing, PR, anything. And that is information. Information for us is currency. We're in the business of just giving people information. And why do we do that? Yeah, it helps us sell things, because sometimes we sell that information to people. But mostly, think about you want to get coverage somewhere online. And this person, this reporter, like, they're doing their job, and that's a lot of work. It takes a lot of time to, like, research a company, to write a blog post, to do anything. And how can you convince that person to spend the time to write about you? Like, would you, like, write a book report about some random person that talked to you on the street? Like, that would, like, kill your week. So, like, why don't you do something that can make their lives just way, way easier? Give them a report that if they write about it, they will look incredible. One of the first things we wrote about that really had a lot of currency, or did really well, was we crawled through every iPhone and figured out how much they were depreciating over time. And then we also crawled through every Android phone and found out how much they were depreciating over time. And we also, this was like 2 years ago, so BlackBerry, 2 and a half years ago, so BlackBerry was even like, a thing. And we figured out how much BlackBerrys were depreciating over time. We came up with this index, it was like, "Oh, it looks like iPhones depreciate the less— the least." They're like the best resale value compared to Androids or BlackBerrys. And we wrote that, and like all of a sudden, like nothing really happened in the beginning. Like we just wrote it, we put it in our blog. Then we like emailed like a few places. We emailed like Hugo, we emailed TechCrunch, we emailed VentureBeat. And eventually, like, one of them saw it and wrote about it, because like that was a story for them. Like, iPhones had the lowest depreciation. And like, that's not like he didn't have to like put in a ton of work to like do the calculations himself or write about it. And that made him look like really smart. It helped him do his job. It brought new information to the world. And then that led to like a torrential, like, you know, like everyone then covered this thing. Like, iPhones are number one in depreciation according to Priceonomics. Which was just like a couple of dudes. But that was in the New York Times, that was in like the Wall Street Journal, that was everywhere. And so, like, just give people information because like otherwise you're gonna be like, "Can you write about me?" and they'll say, "Why?" You won't have a good answer other than, "I would like you to." So how did we get started? Like, and what do I define as information? It could be like awesome data that your company puts together or is a byproduct of your company, but it can also be like an anecdote that only you or your company authentically know. And like, what do I mean by that? When we first launched, it was right before Christmas, and we had arranged for like— because we're in Y Combinator, they like helped us make sure like TechCrunch would write about Pressnomics, Kelley Blue Book, for everything. And we were like, why don't you write a blog post too about like our company with the launch. And I wrote this thing up that was like— because it was Christmas time and like we're a price guide— it was like 5 used things you can buy your wife for Christmas or something. Like, it was something very lame. But I just thought like, well, this is what a blog is supposed to do. Like, you talk about your product, you like shamelessly plug it. Like, it's not supposed to be that popular. It's just like a blog. And so, but the thing was like TechCrunch was gonna write about us before Christmas, but they ended up writing about us on like the 26th. So I had written this blog post all ready to go, like that I was gonna like post afterwards, after we launched, about like, you know, Christmas presents, but it was after Christmas. So I couldn't do it. I couldn't write about this thing. So by accident I was like, well, I gotta produce something today. And what do I know about? So I thought, like, well, one interesting thing we did last week was we emailed all our friends in Y Combinator and we're like, hey, we were testing out our data and we're buying Aeron chairs. And we'll, like, buy them on Craigslist and then we'll resell them to you at a markup. And so we would, like, sell them for $400 and then source them online using our data for, like, you know, $300, $350. And that to me was like, okay, well, I can write about that. Like, it's not like that much about our product really, but it's more about how we did this. How would we like go out there and find these chairs and try to get people and like try to rip off our classmates by taking their money? And that ended up being like such a better story. Like, people actually liked that because it was authentic. It was stuff that we knew, even though it was just an anecdote about us. And then like that ended up on Hacker News, it was the number one article there, it was in like AllThingsD, it was in like all these places that like if we'd just written like the boring thing about our company nothing would have happened. So in the beginning every time we wrote a blog post it would take us at least 40 hours. So when you, like, read it, it might even not even be that many words, and it might look kind of breezy, the tone, like someone just dashed it off, like, um, but that was never the case. Like, it would take so much time for us to assemble the data, for us to figure out how to visualize it, and for us to tell the story. Um, we wrote things like the Fix— the Hipster Fixie Bike Index, where we figured out, like, where in America a fixie which year bikes were selling the most. And like, doing that is a lot of work. Like, you have to crawl a lot of data to figure that out. And we posted that, that like went gangbusters. We did something about San Francisco rent prices over a year ago before like people really were able to quantify how fast the prices were increasing in the city. And that got 30,000 Facebook likes. But it took us like almost a month to pull that data together and visualize it. We've done stuff where we compared Airbnb prices to hotel prices by crawling through all these massive websites. And like, those ones that use a lot of data take a lot of time. But even the ones that don't involve a lot of data, that we just write about because we want to bring new information to the world, they— we put a ton, a ton, a ton of effort into it. Sometimes we don't put that much effort into it, though. This was like our most popular blog post ever. It was called "How Much Does It Cost to Book Your Favorite Band?" We thought it was like kind of a crappy blog post. Someone had sent us this data about like how much does it cost to book like Madonna versus like Justin Timberlake versus like everyone. We just like threw it on our blog. It took us like no time. Because we've been doing this so long and people were sort of expecting interesting information from us, we sort of started to have like all these advantages now. We put something up there, And like this one blog post got 1.2 million views and over 100,000 Facebook likes. And we just like threw it on there. We thought like, we're not even going to send this to our email list. It's not good enough. No one will care. And so, and like that didn't really drive us any— I mean, we got like a million visitors in a day. It didn't really like drive any business to us. But at the same time, we got a lot of business that day. And how that happened. We had also written a piece earlier about RetailMeNot, and we had crawled through all their Google search results to show how awesome they were at SEO and how that 71% of the time they were the number one result for coupon-related Google searches. And then that same day we had that massive Madonna spike, RetailMeNot Panda— Google's Panda search update hit. RetailMeNot got creamed, and their stock price dropped like 30% in a day because they were getting all their traffic from Google. And then all these hedge fund analysts started Googling like, RetailMeNot SEO data, and we just showed up. And like that day, like almost every big hedge fund on Wall Street emailed us. And it was the same day that like Madonna was going viral on our site, but something else that we had written unrelated was then really taking off and driving business. So as you do this, you'll have your own rules about like making interesting, authentic information. Ours is like, bring information, be right, be interesting. Like, we want to basically make content that's like the most interesting articles that are in the business section in the New York Times or the Wall Street Journal. And you don't have to like, you know, you don't have to like reinvent the wheel every time. Like now as we're doing this, we have like themes. Like, you can tap into like a really popular subculture. We wrote about like the business of fish sometime. Like, we— you can like just write stuff that like people feel good when they share it. Some information for us that's like really surprising, or we like saying like something is bullshit, and like, or the economics of food trucks, or economics of, you know, anything. You can tell a story about, like, how you guys started, about how one of your friends started a company, you know, the hero cycle as an entrepreneur. You can just post a chart that tells an interesting story. And you can come up with various themes that, like, you keep hitting over and over. And then in the beginning, you have to basically do some work to make it spread. And after a while, like, I mean, in the beginning, like, you'll write something, it'll be really cool, but nothing will happen unless you like sort of push it. And the good news is that like if you push it a little bit, it can really, really spread because of Facebook, because like one blog reads another blog, whatever. You just need to get mentioned like once or twice, or you need to like get on like some subreddit, or you need to get like somewhere on Hacker News, and then eventually it'll just avalanche. And that's like the beauty of— the beauty of like content today is that you don't have that many disadvantages versus just being like the New York Times. You know, like, they rely on people sharing it on Facebook just like you rely on people sharing it on Facebook. If you make something as good as what's in the New York Times, people don't care if it's on your little blog or the New York Times. Like, they're just sharing it on Twitter, they're just sharing it on Facebook. They're just a blogger needing to have some data to write about, and they'll write about you just as quickly as they'll write about something that's in The Economist or New York Times. So like put in a little effort up front to like, you know, email a journalist and say, "Hey, here's this quick— here's a study we wrote. You might like it." You know, like do those things. They'll really pay off. And eventually, like, eventually it will like become easier, you know. Like that article we wrote that had a million visitors, that wouldn't have happened the first day that we get a million visitors when we wrote something. Like you have this cumulative advantage over time, so stick with it. It's not It's not easy. And then finally, like, we were so like, we loved our content so much that we're like, not all of us needs to necessarily even promote our business. Like, 10% of it needs to. 90% of it just needs to be good. And we found it was so good, we like loved it so much, so authentically, we even put it in a book called Everything is Bullshit, and we self-published it. And a month ago, we just like put it on our blog. We didn't do anything. And that month it sold $20,000 worth of like profit for us. And it was the number one econ book on Amazon when we launched it, above like Freakonomics. And that was just like the level of intensity that we put into making our blog really, really, really good. That like, we're like, we think it's so good that people would buy these recycled blog posts. And we added new content to it, but it was really just a testament to, like, our commitment to making good things and that they would spread. So, thank you. That's Price Mountains. Woo! Thank you, very good. If you guys have a question, please come down. Unique opportunity to speak to one of the best content experts I've seen, actually. Questions? Ah, yes, I'm here.

Awesome presentation. Enjoyed it. Definitely have to check you guys out after this. So I just have a question about how do you get inspiration, I guess, for your content? It seems like you take a lot of time to develop it, so what do you do and what stands out? I got I guess, about writing it?

I guess we just get struck by things that we find are interesting, and we start with that. The other day, we were talking about the Witness Protection Program over lunch. We're like, "Wait a minute, is that a thing? Do people actually go in a Witness Protection Program? What happens to their Facebook profiles?" And then we have a lunch conversation. And then we're like, "Well, why don't we just— it's not that much work to dig into it." And so like now, right now, someone's working on a piece about like how the Witness Protection Program works. And like it's just sort of like an ephemeral decision, like this seems interesting to us. And like sometimes we like, you know, well, for every like good idea we have, we have like 10 bad ideas. But like we really just focus on like something we find that strikes us as interesting.

Interesting.

Cool, thanks. Hey, how you doing?

So I'm Michael. My company is Lots of Stops, which I call air traffic control for delivery. But I'm curious, did Madonna bookings increase after that blog post?

I would guess not. I'm just kidding.

No, my question is, and maybe I wasn't paying attention in the beginning, is So is your revenue model you being hired to do data research for other companies?

Our revenue model is basically, um, we will crawl the web, like, in a customized way for people. Um, and so, like, if you ever, like, start a company that involves, like, a lot of data, like, you're, like, crawling through, like, a retailer site all the time, or you're crawling through, um, a marketplace or whatever, Like, it requires at least one full-time person, possibly like a team of full-time people. And these sites are always changing. They're a pain in the butt. You might get blocked. Like, all these— there are all these like very idiosyncratic things about like getting data. And so what our business model is, we basically will do that for a company because we just specialize in that. So we can just get them the feeds of the data. And while for them it would take like a full-time team, at least one person, possibly like one or two, and that person would hate doing it, For us, we love doing it, and we focus on that. So that is 80% of our revenue.

So if you get backlogged on requests from other companies, what do you do there? Do you outsource it?

No. We're sort of building the platform to make it easier every time. The first time we did it, it took us a couple months to really get the data in the right way. Then the next time we did it, it took us a month. Now it takes us like 2 weeks to set up a customer on our platform. And eventually the goal is to make it so that a customer could run the data on our platform and monitor like, oh, this site changed and now my crawler's not working, or this thing happened. But for now, it's like every time we're building this plane as it's flying in terms of making it easier for companies to get data. Cool. Well, I enjoyed your talk. Thank you. Thank you.

Hi, I'm Amy. I come from the editorial content side of things, so I was wondering if you could speak a little bit more like data crawling for dummies, the sort of the technical side of things.

The data that you're finding, like old-fashioned ways, all the research that editors have to do to optimize that and make it quicker, but to understand the tech side, I think would Yeah, so there's like all this stuff about data journalism out there today, like, you know, FiveThirtyEight or Vox or things like that, and people are like, well, um, you know, when people apply to our jobs for writers sometimes they're like, hey, you know, I know how to crawl data, like I did this big project, um, but like for our writers we don't have them do any of the crawling, um, because the one— the data that we want is so like hard to get We want like real, real professionals to get it. And almost nothing we do could be done by someone that was just like doing it as a hobby or like as a side thing. And so, you know, like other people out there might like have writers or like get the data too, but we like split it up. So we have like the professionals do the professional data crawling, we have the writers like do the professional writing. And so, yeah, our philosophy is that it's like too difficult. To do it unless you really specialize in it.

And crawling, it's— I assume there's a technical side of how the crawling happens, and then it's probably synthesizing the data as well. Is that what the crawlers do?

The crawler— I mean, typically I end up synthesizing it, or someone on our team does, but it's typically me, because I want to put my stamp of approval that every time we put something out there, it has to be right. And if it's not right, we lose all our credibility. If we're going to say that Airbnb prices this much versus hotels, it has to really, really be right. So ultimately, like, nothing gets out the door without me either doing it or approving it.

So it's best if you're an editor to go to somebody like you?

It's best to not mess up. So like, if you feel uncomfortable with it, like, anytime you put something on the internet, like, someone will know it better than you. And so you have to like keep that in mind, that like you have to really, really be right because it it will get out if you make a mistake. Or it has to be done in a way, like, don't give an opinion on something that you're not 100% sure about. Like, have the scope of the argument very, very narrow so that you are the expert in that at least narrow question.

So how do you source that crawl data if you are an editor? Like, who is an expert? Like, who could someone go to?

I mean, basically— Minimum. Yeah, I mean, you'd have to, like, sort of work with some engineer, I think, to get it. Sir.

Hey there, my name is Munjal Shah. So my question is, I have a couple questions about Y Combinator. So my buddy got in and chose not to join. So my question is kind of two-part. Number one, do you know what the percentage of people who get in choose not to go? And the second is what the opportunity cost to not joining is.

Like, I would never not do it. Like, people always like come up to me and are like, "Hey, how do you raise money?" and like, "How did you do this?" I'm like, "I have no idea how I did it. I just got into YC and it was so much easier." So it was like the best thing ever. It's sort of like, if you got in and decided not to do it, I don't know, that would be weird.

That's what I always tell him, so I figured maybe it could be a future blog post or something just about like the economics of rejecting Y Combinator, what it really means?

It's like, I mean, economics-wise it makes so much sense. Like, this— I don't even know how we would have raised money. Like, the key thing about— I think, like, when people, like, people email me all the time, like, hey, can you help me raise money? Like, how did you do it? Can you introduce me to this person? And like, no one wants to give money to someone that wants their money, you know? Like, that's a real problem. Like, if you just get a cold email, like, hey, Like, "I'm doing this startup, I cold email you, like, can you give me money?" They will always say no, because it just signals like, this person, I don't know, signals they want my money, which means there must be something deficient about them. Whereas like, being in an incubator, at least like Y Combinator, they extract that away. The investors are forced to talk to people. And instead it looks like, "Oh, well these people don't really necessarily need my money, there must be something great about them, I have to give them my money." So yeah, I'm with you.

I just figured that way I can take something back and show him something, you know, like a blog post saying that he's wrong.

We've got time for 2 more questions, but if we can do it within the next 2 minutes, that'd be really great. So, next question.

I'm Ben Harvey. I'm interning in San Francisco, and I also started a company. But you said you only use blogs, so I'm kind of wondering, like, what's your advice, or what do you think about other social media platforms as an opportunity Yeah, I mean, when I said we only use our blog, like, eventually where most of the traffic comes in through is Facebook.

Um, and like a bit of it comes through Twitter too, and like journalists hang out on Twitter and it's great like for tracking what's going on. But I mean, today, like, part of the cool thing is that so much traffic comes from Facebook or Twitter or other places that it really doesn't matter where you wrote it. Like, you could write it on your website, you can write it on Medium, You can write it on the New York Times. You can write it on TechCrunch. Travis is coming from Facebook, so you have a— it's sort of democratizing where you can put content.

That answered my question. I just thought you meant only blogs, so appreciate it.

Really quick, do you have any tips on how you structure the sharing links? Do you like to put them at the top, at the bottom, on the sides? Is there a particular service to add them in? Did you custom write that plugin? So I think we've never known a really good answer. We put them at the top. We put them at the bottom. And our blog posts are so long that I think we can kind of tell that some people just share it right away without possibly reading 3,000 words. And then some of them, they get to the end, and then they will share them. But I think a lot of the data shows— this isn't necessarily our data, but it's sort of my gut— is that people will just share it on Facebook naturally, or they'll email it naturally. They might not even use your buttons. They're just like, be like, "Oh, this thing was so awesome. I have to like go to my Facebook feed and paste the link in." And so a lot of it is just like people, if it's really great, people will share it no matter how crappy your buttons are. So, yeah. All right. Thank you, Roman.