Blogging for Dollars

In mid-July of this year, the new weblog service Typepad launched, and after a couple months of beta-testing I was hooked on the features. Rather than try and move my highly-customized blog over to the service, I decided starting a new site would be much easier, so I decided my next site would be housed at Typepad. I've long been a fan of TiVo, having owned one for the past three years. Last year I finally started hacking them and found it to be much easier than I thought. I have always followed TiVo news, I'm often writing long emails evangelizing the product to those that ask, and I've wanted to write easier how-tos than the ones I'd read on TiVo hacking. I realized then that a blog focused on the digital recorder space would be perfect for Typepad, and PVRblog was born (actually I originally wanted to call it DVRblog, but accidentally bought pvrblog.com before I realized my mistake, so pvrblog it is). When I started planning and designing the site, I realized that since it was at Typepad, it was going to cost me some money to run it each month. Currently I already have one commercial unix account, a co-located debian server, and a co-located windows server. I wasn't looking forward to another monthly bill, so I looked at my options and realized I could throw Google's Adsense textads into the template and perhaps defray the costs. I designed the site templates with this in mind, making sure there was room for it.

I've run Adsense textads on Blogroots since Google launched the service, but they weren't very lucrative. I figured it was probably because "blog" wasn't too highly sought after of a keyword, so consequently Blogroots usually make only a couple bucks a week, tops. My dream with PVRblog was to make five dollars a week, in order to easily cover any hosting costs. I've been blogging for four years and paying for it throughout, but if I could write a blog I enjoy doing and have it pay for itself, well, that certainly would be something. I launched PVRblog publicly on July 16th with half a dozen posts, then announced it on my personal blog. In a matter of hours, dozens of other sites linked to it, the site ascended Daypop and Blogdex's lists, and all told the debut was a big success. Late that night I remembered the ads and logged into my Adsense account to see how the day went. I clicked over to reports to see the activity. From approximately 3,000 visits (not too shabby at all), enough people clicked through that I made $40 in the first 24 hours. The first thought that came to mind was this:

Great googly-moogly, holy crap. Crap, crap, crap. What the hell just happened? What did I do? What does this mean for weblogs? Would the world be covered in textads when I tell people about this? Shit!

To say the least, I was a bit freaked out. I was measuring everything in increments of $20, hoping to make my monthly hosting and in one day I had enough to pay for two months of hosting. The next day brought another month of paid hosting, and this continued until a few days later I was a Yahoo pick for new site of the day and it resulted in twice the traffic I'd seen so far and over $100 in click-thrus came in during a 24 hour period. Once again, I freaked out.

Micropayments suck. No, really.

I've been a big fan of Scott McCloud's comic works for years and I have optimistically followed his essays and speeches about someday paying writers and artists online through micropayments. I've been a user of Paypal since May of 2000 and have bought many a CD and shirt from local musicians selling online. I've followed the recent launch of Bitpass and have used it to view McCloud's latest work and buy a pile of MP3s from various bands. I've used all these systems but I realize their limitations: uptake is minimal when compared to most web surfers, and few people probably go through all the trouble to pay anyone for content.

Clay Shirky recently tore McCloud and Bitpass a new one over this (Scott wrote a response), and after using micropayment systems like Bitpass, and compared to something like Google's Adense, I've come to the conclusion that micropayments indeed suck. There are a whole host of reasons for their suckitude, but the biggest obstacle has to do with user experience. People are lazy. You are lazy. I am very lazy. I don't want to lift a finger to do anything generally, unless it is really worth it. When you're reading stuff online and you're hit with signup forms, registration forms, or worse of all, payment forms, most people close their browser or go somewhere else. It's just how we work. I haven't read the LA Times website since last year when they instituted a draconian registration process that requires your email for their spam cannon. When users are faced with what looks like work when they were expecting to enjoy reading something, you can bet they will leave, be pissed, or both. Now don't get me wrong, I love Scott McCloud's work and I'd gladly pay him a buck a day to produce the high-quality comics he's capable of, but if I was randomly hitting Bitpass "pay me to see the next page" screens in my web wanderings, I wouldn't be wandering for very long. I'd be leaving. And no matter how far micropayment companies take the technology, there is always going to be a cost (no pun intended) involved with paying for content. You'll have to click buttons, log into extra screens, authorize your accounts and the like. Not all of those steps can be automated and the process will continue to be slightly painful for anyone going through with it (the vast majority will not go through with it). With unobstrusive textads like Google's, the experience for users of your site is only slightly changed. They now have to see an ad (depending on how well you integrated it into the site), and a few of them may very well click on the ads. As the site owner, you get money trickling in, everyone and anyone can view your site, and some people will even find better information or bargains on stuff they wanted. Your readership can grow as much as you want, and usually this carries with it higher click-thrus and more money. Compared to micropayments, unobtrusive textads look a million times better for both the author and the reader.

Google's innovations

Google doesn't get much credit for what it has done to revitalize online advertising. Google's textads aren't just a great technology because they are less of a hassle for readers (compared to micropayments or banner/popup/popunder ads), they are also better than any other textads I've ever seen. I built my own system in fall of 2001 for MetaFilter, and while it did mimic the format, it didn't have any of the content-sensitive filtering, nor did it have any performance limits built-in. When I launched MetaFilter's textads, the first ads were terrific. They were often members' blogs or new services other members had launched. People clicked on them like crazy because they were loaded with fun new sites. It was ads by the members, for the members of MetaFilter. This is typical of other services I've launched online — the early adopters are knowledgeable and enthusiastic, so their first contributions are terrific. A few articles were written about MetaFilter and the new ads and soon after several copycat systems sprang up. People talked about a new revolution in unobtrusive advertising on corporate and hobbyist sites. I enabled anyone to advertise on MetaFilter, and although I kept some standards to keep erotica sites and spammer sites out of the system, eventually most people advertising weren't members and didn't know MetaFilter too well. Many placed ads based on the promises the highly positive textad articles talked about. Their ads were often for things that no one was interested in, even though I warned many advertisers before taking their ads and their money. As people realized the majority of ads weren't that great, they started to ignore the textad box on the front page. The click-thru rates fell like a brick, from 5% in the first month down to the typical advertising industry average of 0.5%, and like that, the revolution of textads was over (there are still ads at MetaFilter and a few dozen people use them for some fairly high-quality ads these days). Google did two amazing things to prevent the same activity in their own system. First off, most everyone knows they do content-sensitive filtering on ads. This is vitally important for sites like PVRblog, but at MetaFilter, any ad was randomly placed next to any content. Google goes to great lengths to make sure that any ads you see have something to do with your searches or the content that accompanies them. The second innovation is their performance metrics and lower limits imposed on advertisers. Simply put, if your ad sucks, you will get booted, so the quality system-wide of the ads never goes down. It's brilliantly simple, but if your ad was written poorly or you picked the wrong keywords, the click-thru rates will reflect that, and Google often ends campaigns that don't meet a minimum of criteria. This Darwinian system ensures the strong survive, that ads are highly targeted, well-written, and used by a good percentage of searchers and readers, making the system as a whole work. Simply put, Google not only invented the textad, they perfected it, and thanks to those innovations, Adsense carries those benefits to site owners and readers.

Adsense advantages

I've met with some great success from Adsense, beyond my wildest dreams, but one of my first thoughts was "what does this mean for blogs." When I asked a few friends during my initial freakout phase, even those that had low opinions of blogs in the past saw this as a ray of hope. People creating websites for the fun of it have never really had a way to be compensated directly for maintaining high quality sites. "This is a wonderful thing" they said. "For once, writers can get a break, by finding a way to pay for hosting and maybe even encourage them to write more" was something else I heard.

I think the success of Adsense means a lot for not just blogs, but for any website produced by writers, historians, artists, and fans. One of the first great things about the web was the plethora of amazingly informative sites maintained by passionate individuals in their free time. There's no reason any topic-focused site can't benefit from Adsense. It's my hope in writing this piece to show how this is a potential boon to other website authors. Make no mistake, last month PVRblog made a lot of money from Adsense. When was the last time you heard someone say they received a check for advertising on their hobby site that could be used to purchase a fully loaded Aeron chair? Sounds like something you might have heard in 1998, no? Well it's true today, and I hope a lot more people meet with the same success.

Tips for a successful Adsense site

I know that a lot of the PVRblog's luck with adsense was just that: accidental luck. I had no idea so many people would like the idea of a blog about DVR technology. I didn't know TiVo was such a sought after keyword at Google. I had no idea my previous projects' Google ranking would help my site out when I first linked to the site from my blog. The whole thing was one big happy accident, but I've noticed some trends between what I did and what others have tried. What follows are things I've noticed worked to my advantage. 1. Pick a topicBlogs are about anything and everything and it isn't every day that you find a good blog focused on a topic. In order to have any remote chance of success gaining an interested audience and getting good on-topic ads showing up, pick a narrow topic you are passionate about and run with it. I would guess that I do just as good or better than Gizmodo on textads (Gizmodo certainly covers the same area of PVRblog, just not as in-depth) even though I probably have 1/10th the traffic because my site is more tightly focused. If there's anything in this list that requires a drastic change on the part of website authors, this is it. Focused blogging isn't that popular but I'm convinced it's the only way to have a chance to carve out a niche on the web. If you want to proclaim yourself as an expert on a topic to both an audience and search engines so that people will know you're the one site to go to for information, you'll have to focus. Focus and be as specific as you can. 2. Consider your topic as it relates to the webIf what you are aiming for is ad revenue, it helps if your topic is something you can buy products related to it. It also helps if those products can be bought online and people are comfortable with it. One of my favorite topic sites (arguably slightly blog-like) is Kicksology. Professor K knows everything and anything about basketball shoes and about once a week I drop into the site to see what's new in cutting edge shoe design. Often when I see a rave review on a cool looking shoe, I want to know how much it costs and if I can buy them. It's an impulse buying thing, but if you notice Kicksology recently added Google's ads to the site, but they're not super-focused. Ideally, if I was reading about a new shoe, I'd want ads offering the same shoe for purchase right now. I've checked out a few dozen of the reviews, but the new air jordan review is the only one that carries with it targetted ads. Generally speaking, Kicksology is about something not normally ordered or sold online and the ads are often a poor fit for the content (no one's fault really, people just don't buy that many shoes online). TiVos are very close to the web. People buy them online, they look up tips and hacks for them, and resellers have tons of TiVos to move. I didn't really think about it when I started the site, but thanks to the mass availability and customers looking for deals on them, the web's a natural place to shop for a new TiVo. If you're really interested in knowing how well a topic might work out, try going through the process of placing a Google Adwords ad. During the process they'll tell you how much a keyword will cost you, and you can use that to determine if writing a blog about goldfish is going to be more lucrative than the one you could be writing about golden retrievers. 3. Be passionate and write your ass offDon't start a blog just to turn a buck, because it's going to be clear to your audience that you don't really care about the topic if you don't offer much beyond press releases from companies. If you want to have a site that ranks highly at Google, write how-to article after how-to article and offer content no one could find anywhere else. I love this guide to ranking higher in google because it doesn't focus exclusively on HTML tricks or stoop to tips on gaming the system, it simply says: write the most useful website on earth and everyone will link to you, which will make you #1. I started PVRblog because I've been following the space for the past three years and I have dozens of in-depth tutorials I've written and want to write about the subject. I'm enthusiastic about the topic and I look forward to spending a few downtime hours writing articles, conducting interviews, or reviewing books and hardware for the site. 4. Designing for Google and your audienceDon't underestimate the power of Google and google-ability of your site. About half of all the traffic to PVRblog is from a Google search. If people are looking for information on how to upgrade a tivo, they might find my articles about it, and alongside every article are four links to upgrade kits at various prices. I wouldn't be surprised if the click-thrus are crazy high on those links, for those users. I do the same thing myself, often looking up reviews on cellphones and following ad links to help find the best prices I can. On the technical side of things, having an accessible, valid XHTML site, with good semantics, good page titles, and good filenames helps Google index your site. Typepad does all these things extremely well right out of the box. After I launched PVRblog, Google indexed the entire site within hours and reindexes it often. The site shows up in the top ten for many common TiVo hacking or TiVo feature searches. Searchers are often looking for info to help a purchase, and are likely to click an ad, so it's worth thinking about them. Nick Denton recently wrote about the design of a weblog may change based on Adsense, and I'd say he's got a lot of good points, but be careful that you don't go too far, forcing people to make extra clicks just so you can stream more ads at them. Your audience will pick up on this eventually and bail. What not to doOf course now that I've given you a few tips, it's important to reiterate what you shouldn't do. Don't just slap ads on your blog and expect to get rich the next day. If every blog about anything on earth is going to carry adsense boxes, their utility is going to go down and people won't be likely to click on them. Don't be disappointed if you're not pulling down big bucks on your topic-focused, well-googled site. It takes time to build an audience and gain links from people that find your content useful. If you follow these guidelines, it's quite possible you'll be able to pay for your own hosting. Eventually, you might make more.

The downsides

Like anything, it's not all roses and Adsense is far from absolutely perfect. It's got two big drawbacks: the approval system and the terms of service. The approval system is evidently run by humans, and they state upfront they won't accept personal sites. I would guess that's likely because most personal sites aren't focused on much of anything besides a specific person, and hence would be hard to advertise to. For even focused sites, some people have had real trouble convincing the Adsense folks to approve them, even when their sites could potentially produce great on-target ads. The decisions are sometimes arbitrary and will likely work against Google if it continues to mistakenly deny legitimate sites. The other big problem is the terms of service for Adsense which have received a lot of scorn recently after a few people were booted from the program without any recourse. Google took it a step farther and beefed up the legalese to even prohibit the discussion of the TOS, which is kinda nuts. Of course, I've benefited greatly from Adsense, and it's probably no surprise that the TOS issues weren't a dealbreaker for me personally. Google's between a rock and a hard place on this issue. If they were more transparent, like they have been with their search results, people would no doubt spend their lives gaming the system. What Google hasn't stated upfront about how pagerank works, people have reverse-engineered to great effort. There are whole sites and thousands of people that dedicate their lives to getting their sites as high as possible in Google, whether or not their site has good information or helps out web surfers. These folks see no problem in making Google less useful for searchers if it means their clients are happy with their rank. Google's spent every day since their launch in a cat-and-mouse race to beat those that seek to game the system. If Adsense were transparent, say if they told you how many clicks were allowed per hour from a page or an IP, you can bet that within a couple hours people would produce adsense link clicking bots and bot farms that carried hundreds of IPs solely to produce fake click-thrus for cash. So it looks like they are trying to keep a lid on this, keep their methods as secret as possible and went as far as to put a gag order on the entire subject. I don't know what's going to work out for them in the end, but I think they could potentially lose a lot of money whether or not they tell everyone how Adsense works exactly. The future will tell us if they made the right move here. Areas for improvementThe tools offered for authors using Adsense are pretty paltry. It's a new service and I expect it to change, but it'd be nice if the results were more granular so you could compare which pages produced the highest click-thrus, and tailor your content to that. It'd be nice if you could see a breakdown of each domain/site you were using the ads on. The new custom colors and layouts are great, but it'd be nice if you could track how different sizes/colors were performing to enhance your design. Currently you just see a number for views, a click-thru total, and a dollar value assigned to that.

Conclusions

A long-ago promise of the web was that people could share their expert opinions and thoughts on anything, and others in the world would find them. Search engines have helped the web live up to that promise and it works well today. Google's Adsense goes one further, in that someone writing passionately and expertly on a topic can now also make some money doing it. Nick Denton has been saying niche blogs could turn a profit for a while now, and after my experiences I'd have to say he was right. The opportunity is now there — if you've ever wanted to write a topic-focused blog and wondered if you could get paid for it, Adsense could make it all work for you.