Secrets of Pay Per Click Advertising

In part one, Secrets of Running a Website, I introduced my background in running various websites. Here in part two, I’ll look at how advertising works, starting with pay per click programs like Google AdSense, AdBrite, and Yahoo! Ads.

You might be wondering what kind of massive ad revenues my site generates. It’s no MySpace, but I do drive some significant traffic. This last weekend, a popular story on market share drove 20,461 unique visitors and 35,269 page views. That means the majority of readers who came to the site were interested enough to continue looking around. What kind of bonanza of advertising revenue did I earn from all this traffic?

Apart from direct contributions from readers, Yahoo! Ads amounted to $2.69, affiliate links from the iTunes Music Store were $1.14, and Amazon links paid me commissions of $0.70.

To those worried that my expressed opinions cater to whims of these advertisers, let me assure you that the subjects I write about are not determined by less than $5 of potential revenue (I have to accumulate $100 of earnings before I can actually receive payment from an advertiser).

Money and the Web

If you are considering adding sponsors to your own website, those numbers might sound depressing. They also help to explain why intelligent content can be hard to find on the Internet: there isn't much of a market for providing smart, free information!

Apart from those writing for fun, that leaves a lot of space to fill by companies trying to dish out ads. Without a real revenue stream to pay for smart content, that space ends up getting inflated by low interest, high volume windbags, who write sensationalist tripe to generate teapot tempest controversies that drive waves of traffic.

When the commercial Internet started to emerge in the early 90's, newspapers, TV channels, and other traditional publishers all thought the web would provide them profits from a new huge paid audience. It turned out people didn't want to pay to access every site they visited, partly because there was so much free content available. The throngs of visitors browsing the web didn't even want to bother with having to log in to receive free content.

That was a great disappointment to the get rich quick schemers who thought they could just throw some junk on the web and then bill billions of web visitors to see it. As it turns out, it’s rather difficult to sell information; I've noted before that people don't want to pay for ideas, they want to spend their money on tangible things. That's why it’s easier to sell hardware than software.

Problems with Web Advertising

Of course, if you multiply my small ad revenues a thousand times, and then repeat the performance every day, that trickle of ad revenue starts to add up to real money. Web properties started figuring this out in the late 90's, when the same players who had struck out on a pay per view content model decided to try pitching the advertising supported model common to print publishing.

The returns for advertising on the web, and the returns in producing ad supported content on the web were both negligible. The real winners were the middlemen handling advertising sales. While the majority of dot coms failed miserably, there were significant success stories in Overture, the inventor of pay per click advertising, and Google, who copied Overture’s lead.

The market had a hard time trying to determine how much web advertising was really worth. In some ways, ads on the web were like ads in newspapers; readers might ignore them, but they still created valuable awareness. Most advertising does not aspire to generate an immediate sale. Trying to determine the immediate results of an ad campaign typically isn't very easy, so establishing a value for ad placement is tricky.

In print marketing, an advertiser might pay a lot of money to place an ad that a lot of readers would see, but those ad impressions are tied to something real: a paper copy. On the web, an advertiser paying per ad impression might end up with a bill for millions of virtual pages hit by machines impervious to their advertising message.

It is also harder to define a regional audience for a web site, because viewers can be anywhere. A merchant wouldn't benefit from thousands of people in another country knowing about their product if it didn't result in additional sales.

Additionally, advertising on the Internet is too easy. Nobody can print their own newspaper or run their own TV broadcast and hope to reach viewers without spending millions on distribution, but anyone can send out a glut of commercial messages by email or post them on the web at little cost.

All those marketing messages tend to water down the overall effectiveness of Internet advertising, because the sheer volume of ads overwhelms the audience.

The result is similar to the problems behind attaching a price to media downloads: should advertisers pay next to nothing for billions of ads because the individual effect of each ad is tiny, or is the very limited ad space adjacent to popular content worth a lot?

Advertisers want their ad budget to return results, and ad sponsored content generators want to get paid for hosting those ads. The result is a dramatic conflict between quantity and quality that both sides are losing as they blanket web pages with ads that flash, pop into the viewers' faces, and take over the screen to display interstitials. As ads get more annoying, viewers simply avoid sites that have become too obnoxious to visit, or find ways to block their ads.

Advertising Options

I originally added advertising to my site to fill in the spartan white space on the right hand side. I can't really expand my web page content into that area, because it wouldn't be accessible to readers who use a limited resolution display.

Having my ads on the right also makes them optional, as readers can simply resize their browser window to avoid seeing them. I also decided it would be clever to place them within a div tag with an alpha transparency setting, to tone down their impact and subtly send them into the background

Recently, I stopped using pay per click advertising, and have moved entirely to affiliate ads. That means I'm only paid by viewers who find my links useful enough to use. That means I can't rely on pure ad annoyance or hyper-ad placement to induce accidental clickage, but instead have to find affiliates that are a benefit to my readers.

For example, I found a magazine site that pays me a commission for every reader who signs up to receive a free subscription to one of several business magazines.

Why have I given up on pay per click? Part of this was a growing cancer gnawing on my conscience: the death of the web at the hands of what I call...

The DvorMac Effect

Personally, I decided I would rather read the blog of some nobody with an interesting take on things, than continue to swarm with the idiot traffic that flops upon every morsel of insanity gushing from the keyboard of professional rabble rousers who have made a career out of generating noise about subjects that are not even interesting, let alone sane or practical.

Playing the role of a reactionary Defender of Common Sense, and refuting the nonstop flow of inane ideas from pay per click sensationalists, I was inadvertently generating more pay per click revenue to support them. I've since decided to mimic the suave style of the Daily Show's John Stewart and Steven Colbert. They dismantle bad ideas for comic effect, and highlight smart stuff to consider on the side. That's where I'm aiming anyway.

Pay per click ads are sponsoring the death of the web by encouraging mindless, reactionary traffic buzz around tired and stupid ideas. In order to really embrace pay per click traffic, I would need to start writing fight articles about advocacy topics that simply aren't interesting to me. I'd also need to align myself with populist viewpoints that generate the right kind of dumb traffic.

All of that would make me feel dirtier than shopping at Wal-Mart or buying an off road SUV to drive around on paved city streets. Fortunately, I wasn't making any real money on pay per click, so I can feel righteously smug about all this without actually suffering any personal sacrifice: Del.ici.ous!

The Fickle Finger of Google AdSense

Over the last several years, I have learned a bit about dealing with various web advertising and affiliate programs, and I'm happy to share what I've learned. Prior to Google AdSense, ad serving was handled by companies like DoubleClick, which served loud, garish ads that tried to jump off the page to frantically grab web surfer's attention.

The public responded by actively diverting their attention away from annoying ads, or installing tools to block ads entirely. I was always too disgusted by the lameness of Internet advertising to use anything prior to AdSense.

Google AdSense introduced subtle, text based advertising that employed the company's search technology to present ads the were at least somewhat related to the surrounding content. The program also paid fairly significant fees to website owners, particularly for web site clicks. Depending on the value assigned to an ad category, a single click on a Google ad can pay the website owner nearly a dollar.

With that in mind, you should regularly click the ads in websites you like; it costs you nothing! Web site owners can't tell you that, because ad contracts forbid owners from pleading with users to click their ads. Of course, if you click too many all at once, pay per click companies like Google are likely to terminate the website's AdSense account.

Until this spring, Google was paying me around $100 a year on my site. It didn't even pay my basic web hosting expenses, but it was a nice perk to get a mystery deposit from them every few months.

The problem with doing online business sight unseen with a billion dollar corporation, is that you have no leverage if said company decides to forget you exist. When I returned from vacation this spring, I asked Google why I'd received an email saying they'd terminated my AdSense account.

They invited me to write up a detailed appeal, then eventually got around to telling me they didn't want to consider my appeal, and that the $100 balance they were due to pay me for the last year of displaying their ads would be "returned to their advertisers" because they suspected click fraud on my account.

So as much as I like playing with Google's free apps, like SketchUp, Google Earth, and Maps, dealing with Google itself is like doing business with the mob. I sure wouldn't want to keep my important data online in Ajax enabled Google SpreadSheets if the company reserves the right to unceremoniously dump their peon users at the drop of a hat.

That's a major flaw in Google's future business strategy of shifting everyone to breathing their brand of oxygen. What happens when they decide to cut the air supply?

If Google really investigated some fraud connected with my site, why haven't they got around to figuring out that the billions of ads they serve to obvious fraudsters posting garbage content on fake search result pages, which are dutifully included in Google's own web searches on any and every subject, are also a complete waste of advertisers' ad budgets?

It's because Google makes crap loads of money watering down the actual content on the Internet and delivering increasingly worthless search results for their users! At the moment, nobody has really taken notice, but how long can Google continue to crap where they eat before their impressive sounding market valuation dumps into the toilet?

Google, welcome to my Suspiciously Shady List.

AdBrite's Dim Prospects for Revenue

Next up, I tried AdBrite, another text based ad server that happens to be located right here in San Francisco. Their improvement over AdSense is to include a link encouraging advertisers to directly sponsor the site; following the link, a potential advertiser can see what kind of traffic the site generates, and get an immediate quote for placing their own ads on the site. I thought that sounded like a clever idea.

Without a direct advertiser jumping in my lap, and without spending any efforts to court one, I found that AdBrite failed to pay anything at all for ad placements. After leaving them on my site for several weeks, they recorded 53,859 hits and 62 clicks, and assigned me all of $2.70 in earnings. For that kind of money, I can afford to advertise the beauty of stark expanses of white on my website.

Yahoo! Ads! Suck!

Everybody knows Yahoo! has no taste; their search site always put me in mind of a Google re-mastered by Microsoft's marketing department. Lately however, Yahoo! has attempted to be more like Google, and is even trying to beat them at their own game.

Part of this effort involved paying $1.7 billion in 2003 to buy Overture, the inventor of pay per click advertising. Overture was formerly known as GoTo.com, one of the many dot coms to spring from Idealab.

Shortly before the purchase, Overture sued Google for copying their patented ideas about paying for ad clicks and charging advertisers to appear prominently in search results.

After buying Overture, Yahoo! ended the dispute with a truce that involved Google giving 2.7 million shares of its common stock to Yahoo! in exchange for a perpetual license for the pay per click business model.

Yahoo! has since copied Google's text ad program down to the last dot, so I figured Yahoo! ads would pay something similar to AdSense. While I didn't do a scientific comparison, from my numbers it looks like that isn't the case at all.

Apparently a lot of things about Yahoo! don't ad up. I'd heard from industry insiders about the corporate drama between the remains of Overture and their new Yahoo! overlords, and that Yahoo's recently rebuilt ad relevance matching software doesn't actually work and is actually secretly cheating their advertisers.

Still, I was somewhat surprised to see that the number of ad impressions Yahoo! was trying to give me credit for had no relation to the number of actual pages I was serving up. I have web statistics that are calculated using both my own server's access logs, as well as an outside, javascript based analytics system.

Yahoo! was posting three or four ads on each page I served, but they reported less than half as many ad impressions as I had pageviews! I find it hard to believe that the majority of users are blocking ads; if they were, they’d also block my stat counter. In any event, Yahoo! is either grossly incompetent, willfully defrauding, or a little of both.

Doing business with Yahoo! feels like drinking cloudy water in Mexico. They sell the ads behind the infamous Gator (aka Claria and Gain Ad Server), and work to actively distribute scads of adware and other garbage to unsuspecting PC users trying to install their Yahoo! IM or web search toolbar.

Along with Google, Microsoft, AOL, Cisco and Skype, Yahoo! has contributed to efforts by China to censor the Internet. Yahoo! has even turned over documents sent via Yahoo! mail to the communist government to rat out Chinese journalists and send them to prison. And of course, they were caught sending pornographic images to kids in school who were using ‘safe search results’ in Yahoo! image search for such racy topics as "www."

I have therefore decided to wash my hands of Yahoo! ads as well. All in all, if you're trying to earn revenue for your site, placing pay per click ads isn't a useful way to do it, unless you are a trash magnet posting garbage on billions of pages and interlinking them to gain a fake page rank.

If you are tired of seeing AdSense and Yahoo! Ads, click them into oblivion, so their web hosts gain a few bucks of revenue, and advertisers see that their advertising budget is getting spent up with no results.

As for you Yahoo!, you are now on my Ripe Stank of Evil List, along with Flash, Edgar Bronfman Jr., Verizon and Microsoft.

All About Affiliates

Earlier, I alluded to affiliate advertising as the alternative to pay per click. The later pays for indiscriminate swarms of traffic that might accidently click on those flying monkeys and ads earnestly asking viewers:

Install this SPYWAYR Remover NOW!!! CLICK!!!

Affiliate ads, however, only pay for actual business. So if I link to the iTunes Music Store, Amazon, the Apple Store, Netflix or other network affiliates, I only get paid after a viewer makes a purchase, or otherwise makes the advertiser happy, as is the case with the free business magazine offer.

In my next article, I'll cover how affiliates work and how valuable they are. I'll also look at advertising alternatives, the details of web traffic, and take apart what generates good traffic and bad traffic. You might be surprised by some of the details.

I really like to hear from readers. What do you think? Leave a comment or email me with your ideas.