Interview: Ryan North, author of Dinosaur Comics, and co-founder of Project Wonderful, a new ad auction service

by Josh Fruhlinger

May 24, 2007 —

Josh Fruhlinger recently spoke with Ryan North, author of the popular Web comic, Dinosaur Comics and perhaps a little more relevant for this audience, he's also the person behind a new ad auction service that's called Project Wonderful.

Listen to the original interview here, or visit our Podcast Center for more audio interviews.

Josh Fruhlinger: Let's start with a little bit about your professional background as a developer. I know probably most people who have heard of you in your Internet fame have heard of your comics, but what's the technical background behind the comics?

Ryan North: I actually trained and worked as a computer scientist, as a developer, and started the comics in my spare time and they became really popular. So I ended up doing the comics full-time and then computer science is kind of an itch that you have to scratch and so I started thinking, oh, I should do some other stuff, what can I do? What can I enjoy that's also computer science? And that's where Project Wonderful came from. I'm coming from a computer science background today, graduate degree in computational linguistics, which is basically figuring out how to get computers to talk to us and to each other in natural language, which is fun, but hard.

Fruhlinger: I can imagine. You never tried to get the computers to write your comics for you?

North: If I could, I would have a very easy job.

Fruhlinger: So when you started doing the comics fulltime, the idea of doing the ad service wasn't the motivator behind that? You didn't like leave a job to say, I'm going to change the way that ads are sold on Webcomics or something like that?

North: No, no, I graduated and thought I could get a job, or I could work as a cartoonist and one of these seems like a lot more fun.

Fruhlinger: Once you did start to do Project Wonderful, what was it that you wanted to do differently? Why didn't you just put Google ads all over your site or something like that?

North: A friend of mine, Tim, and I, sat down and were talking about how advertising online is really kind of terrible and we thought, well, if we were going to start fresh - like if we invented the Internet yesterday and want to put ads on it today, how would we do it? And so we tossed around a few ideas and came up with what eventually became Project Wonderful. That was about eight months later. But the main idea behind it is that a lot of advertising is sold based on clicks or displays and unfortunately, the Internet isn't really designed to keep track of who clicked where, when, and who viewed what page when, it's just too easy to game it and there's a huge cottage industry, basically, of click fraud and display fraud, things like that. And so Project Wonderful says, well, you can hack displays and you can hack clicks, but unless you have a time machine, you can't hack time. And so it sells advertising based on time. You say I'm willing to pay a dollar a day for an ad on ITWorld.com and that's what you pay as long as your ad's the high bidder and it's displayed.

Fruhlinger: It's kind of interesting because it's almost like that's going back to the genesis of advertising, like pre-Web advertising. It's almost like the story of advertising on the Web has been further and further refinement of how are we going to exactly measure how many people see this ad or how many people click on this ad and it almost is like you're saying, you know what, that's not working.

North: That's pretty much it, and I think that the problem is that the Web gives you this chance where you think, yes, I could keep track of who clicks there and you can approximate it because the information is there, but it's too easy to fake. But as soon as you start selling things based on that information, there's this huge motivation to just - to game it. And we thought, well, if Google hasn't been able to solve this, there's still this huge problem, so probably we won't be able to. Why don't we just sidestep it and do something that doesn't have this weakness.

Fruhlinger: Now I have noticed -- and I should say, just to put this out here that I have on my humble site - I have some Project Wonderful ads, so I get to see a little bit of the background behind how it works. And I've noticed that you have been adding more information to what the people, what the publishers can see. So now I seem to remember at the beginning, I didn't necessarily see how many clicks I've gotten on my ads and now I do. But that's still not determining exactly how much money people are paying for it.

North: That's true. The idea is that this information is there and we'll give it to you, but we're not going to sell based off of it. So the hope is, and what we've seen, is that there's less of a pull there to start messing with it and trying to make it do stuff, show stuff that's not really there.

So one of the things Project Wonderful tries to do differently is make it a more transparent process. We looked at what advertising was doing right now on the Web and the whole industry is kind of based around not sharing the information that you've got. We saw that, as a publisher, a Website was there, you made $20 this month, here's the money, be happy you have it, without really saying what their take was, where the people were advertising from, things like that. And as an advertiser we saw that you would pay - they said, alright, it's two cents a click here, but you wouldn't know what other people paid in the past. Is that a good deal? The information wasn't really there. And so what Project Wonderful does is give you the information up front, to both the advertiser and the publisher, and say, here's what we've collected, here's what you can see, make your own decisions.

Fruhlinger: When you are thinking in those terms, you're coming at it - in some ways from someone who, like almost more from the perspective of the people buying and selling ad space rather than the middle man, which you sort of have become. Do you see what I'm saying? Have you sold advertising on your site before and were sort of dissatisfied?

North: I tried a couple different ways. I did AdSense for a bit and that didn't work and then I said to people, for $15 a week, I'll put your button here, and that worked, but it was a huge lot of overhead to manage it, to email people, do all the communication that's needed to do that sort of thing.

Fruhlinger: That actually brings us to another question, which was the sort of technical infrastructure behind Project Wonderful. You may not want to talk about all the details, but it was interesting to hear you say that since you're the one who ultimately has to run Project Wonderful and so how much overhead is that for you to run for all these hundreds of advertisers and hundreds of sites?

North: I built it to pretty much run itself and that's a necessity, I guess. There's so many people using it that I can't spend all day fixing things that break or making sure it's still up. It needs to do that on its own. And so it's basically built on a bunch of Open Source software. We have a coupleof MySQL databases and an Apache front-end. And just a lot of coding to make sure that it can play by itself in the corner and not need me to be babysit it all the time.

Fruhlinger: Are you philosophically inclined to use Open Source? Does that sort of go with your open information aesthetic, or was that also partly just sort of necessity as in how can we sort of ramp up quickly to do this?

North: I suppose it's both. It's the right tool for the job is the main motivation for it. It's solid good software, but also the idea of having the transparency behind it is another plus.

Fruhlinger: How much time a week do you feel like you spend fiddling with it and did it turn out to be more than you expected or less than you expected?

North: I'm not sure if fiddling is the right word.

Fruhlinger: I'm just sort of visualizing there's like - it's a machine in the corner of your bedroom or something and every once in awhile a lot of steam comes out and you have to kick it or something like that.

North: I wish it was. That would be kind of cool. My day is the morning doing comics and the afternoon working on Project Wonderful. And a lot of that is adding new features that people have requested, new designs people have been asking for. Can it do this? I need to do this. Why can't I do this sort of thing, which has really been interesting, because it took about nine months to build and then we launched and thought all right, this is the perfect advertising network. Nothing will ever go wrong. And the next day, there's 15 emails saying like, oh, I wish I could do this and I wish I could do this. And so in the past six months developing it, it's been real interesting to see how it's developed and how people respond to that. And it turns out if someone asks for a feature and you give them the feature, they love it.

Fruhlinger: And then you get a warm fuzzy feeling, and a customer who is probably going to stick around a little longer.

North: It's pretty amazing how bad customer service must be on the Web. I find just doing it - someone asks for something that they reasonably shouldn't actually have to ask for, I should have done that already and they get that feature, and they're amazed that you would do this for them. And it's sort of funny how bad we've gone in customer service where doing what really should be expected is seen as a favor.

Fruhlinger: On that note, you mentioned a friend of yours who you sort of hashed this out with at the beginning. Are you and he doing this together or are you the main Project Wonderful janitor and CEO and customer service rep and all these things?

North: Right, right. Well, we're partners, but it's mainly me who does the development in day-to-day and the rest of it. He's sort of the idea man.

Fruhlinger: Are you surprised with how many people that you've gotten that you have to field future requests from? Are you pleasantly surprised? Do you wish there were more? Are you about where you thought you would be?

North: I'm pleasantly surprised, actually. When we launched, there's a lot of new ideas in Project Wonderful that hadn't been tested before and so we weren't even sure if it would work. Like, we're basically a free market for advertising with almost a perfect free market in the economic sense. And we weren't at all certain that the people would like this, would it work, would they say why are you doing this? And so it's been great. We've been ramping up as fast as we can, adding new servers, making the code more and more efficient to take on more and more people. And right now I think there's a waiting list of about 300 people waiting to get on there. We're just adding as fast as we can, which is a great position to be in. An ideal position is to have a waiting list of zero because you can handle all those people.

Fruhlinger: When I see Project Wonderful ads online, I mostly notice them within a fairly specific niche, within the Webcomics world and in some ways it's an interesting angle of selling it, that you're already sort of a celebrity within that world and so people within that niche already read your site and were instantly aware of Project Wonderful without having to do a lot of promotion for it.

North: That's true.

Fruhlinger: Do you have goals to ultimately promote it, actively, elsewhere on the Web? Are you sort of letting it spread out kind of virally?

North: Oh, well, I absolutely would want it to be elsewhere beyond just Webcomics, but the nice thing is that whenever there is a Project Wonderful ad box, or advertising area that people bid on, that links you back to Project Wonderful and explains how the system works and stuff. So in a sense, every advertising area is also kind of a pitch for Project Wonderful, so it kind of sells itself. But you're absolutely right in that my notoriety in the online comic circle made it really easy to launch the system because I could put it up and then just start using it and there's an instant audience of 70,000 people looking at it, which is great. I never really put much stock in celebrity endorsements before, but being able to have this sort of built-in audience that can say, if you like my comic, you might also like this advertising network I invented, it really helped launch - and that's, I think, the main reason why right now we are in this Webcomics niche is that that's where we started and that was the logical, easiest place to start with. But there's nothing in the system that makes it, I think, specialized for Webcomics. I mean it works well for them, but it's not limited, I guess, I would say.

Fruhlinger: Well, one of the things that struck me about it is not that this makes it specifically specialized for Webcomics, but that there's kind of a social dynamic to it. Especially because you have it set up so that if people want to, some of the money can stay within the system, that you can use the money that you've earned for ads on your own site and you can use that money to pay for ads on other people's site without it ever actually coming out of your Project Wonderful account and into your bank. Were you sort of visualizing that social component as being important, or has that sort of happened organically?

North: It kind of took me by surprise. The idea was that we'd make every seller, also a buyer. You're in the system, why not advertise your own site? Make it easy for them to do that if they want. But the social aspect that's built around it has been really surprising, and a lot of fun too. There's one guy's site who had space for I think four button ads and people starting bidding with these ads that were just saying ridiculous things about how, this person likes to kiss horses, and then they'd link to their own comics. And over the course of the day, it just got bid up higher and higher, with more and more ridiculous things. And of course, as a publisher, you can control what shows up on your site, but he was allowing it and when sort of word got out that he was allowing these ridiculous joke ads, people just filled it up with them.

Fruhlinger: Do you worry that if larger advertisers come in, or larger publishers come in, that that sort of sense of whimsy might get lost a little bit, or will there always be sort of small-time operators who can sort of play around like that?

North: The social aspect is basically like any other social site. I mean it's a lot less advanced, it's not designed to be a networking site, but if you think of a place like FaceBook or MySpace, there's big companies there, but you can still have this little community with your friends or people you know and you don't really lose that sense of these are my friends and oh yeah, they're all on this site. Or these are the people I work with, I advertise on, I have this relationship and it's still all on Project Wonderful.

Fruhlinger: And the fact that Philip Morris is also buying ads doesn't affect that?

North: Yeah.

Fruhlinger: I did see in your announcement recently that you've set up a mass bidding system. Is that something that you hope maybe larger companies will be using or do you think that some of your current advertisers are also interested in that sort of thing?

North: Well, it was designed to appeal to people who want to bid or place ads on our member sites that don't to really handle the day-to-day aspect of it. Because right now - or before it was up, you would say, I want to bid on this one site and I'm willing to pay $2 a day from Wednesday 'til next Wednesday with this ad to make it happen. But now you can say, you know what, I just want to bid on any site that has these - like this level of hits, this bidding level in the past, use these ads. Don't bother me with the details. Just sort of make it happen. And it has been designed to appeal to larger companies who don't really want to get into the system that deep, but still want to advertise on our members.

Fruhlinger: For bigger advertisers like that, are you going to take active steps to pitch to them Project Wonderful? Are you still sort of in a position where you can wait for them to come to you?

North: The going out -- coming to them is what I've been calling Sinister Phase 4. With a nice happy name like Project Wonderful, all the new features have a Sinister names since the Phase 2. And so yeah, that's the next step, now that that campaign system is in place to bid across members, the next step is to say, we've got this system, here are people who are willing to put your ad in their site, you should really be giving us money.

Fruhlinger: Once you do that, you are starting to bump up against some of the largest publicly traded corporations in the world like Google and all those other fun people. And so how do you feel about that? The Internet is a little more of a level playing field. But does that make you sort of nervous or excited or do you just not think that it really affects you?

North: Well, it's exciting. It's sort of exciting like going on a first date. It could be a disaster, but it could be nice. I think that if you look at what AdSense is doing and what we're doing, at the core we're both selling advertising on Websites, connecting advertisers to publishers, but we're going about it in very different ways. And I'm not entirely certain that the people at AdSense would see Project Wonderful as a threat right now.

Fruhlinger: There's no panic going on in Google headquarters somewhere in the ...

North: Yeah, I don't so.

Fruhlinger: There's like a siren going off and they're like, my God, have North eliminated immediately.

North: What's Ryan up to this time? But yeah, I don't think it will cause too much problems right now. I hope not anyway. I'd hate to make an enemy out of some of the largest publicly traded corporations in the world. I think most people would say that.

Fruhlinger: Yeah, that's probably fair. Do you and your partner have a long-term trajectory that you're hoping to look for? Do you hope to have more employees at some point or do you just want to see where it goes?

North: I've always said as soon as it starts taking up more time than I'm able to give to it, then it's time to hire someone. So the plan is to eventually start hiring people and actually get maybe an office, someplace nice. But I think it's amazing that - I mean compared to 10 years ago - that I could, or like 2 people can sit down and spend 8 months, 9 months developing this software, launch it and run it and it's almost a one-man show. And it's doing so much for all these different people, that it's exciting and kind of scary, because if I ever collapsed, I would have a lot of work to do very quickly. But there's a whole bunch of safeguards and backups and redundancy to keep everything running to avoid that situation.

Fruhlinger: Do you have anything else you would like to add about Project Wonderful, about Dinosaur Comics, about being an Internet sort of notorious person in general?

North: The thing about Internet fame is that you wake up in the morning, you check your email and there's like six emails from someone, people you've never met telling you that you're great, which is amazing for the ego. Then you go outside and nobody knows who you are. So if you're going to be famous, it's a good kind of famous because you can turn it off and as soon as your head gets big, you realize you're just a guy sitting at a computer refreshing his emails. There are more things you could be doing.

Fruhlinger: All right, well, thank you very much for talking to us.

North: It's my pleasure.

Fruhlinger: Once again this has been Josh Fruhlinger and Ryan North talking. And if you want to learn more about Project Wonderful, you can just go to ProjectWonderful.com and if you want to read Ryan's awesome Dinosaur Comics, you can just go to Qwantz.com, that's Q-W-A-N-T-Z .com, or the more pedestrian DinosaurComics.com.

ITworld.com