Google on Friday was sued for fraud, business code violations, and unjust enrichment, claims arising from the company's alleged sale of low-quality ads.

In the parlance of online marketing, "low-quality ads" refers not to shrill infomercials but to ads that generate a poor response or show a poor conversion rate due to problems with placement, audience targeting, or related factors.

The class-action lawsuit was filed in U.S. District Court in San Jose, Calif., by lawyers from San Francisco-based Schubert Jonckheer Kolbe & Kralowec. The plaintiff is attorney Hal K. Levitte, who advertised his legal services though a Google AdWords pay-per-click campaign last year.

According to the complaint, the "Levitte International" online ad campaign ran from June 1, 2007, through August 18, 2007, and received 202,528 impressions from parked domain pages -- placeholder Web pages with auto-generated links related to a pre-determined search keyword or the hosting domain name.

Google (NSDQ: GOOG) runs an advertising program called AdSense for Domains that seeks to make money off of parked pages. "Parked domain pages generally have no content; however, by adding targeted ads, we hope to help users find what they are looking for," Google's online documentation explains. "Using Google's semantic technology to analyze and understand the meaning of the domain names, AdSense for Domains delivers targeted, conceptually related advertisements to parked domain pages to improve the user experience on these pages."

Despite Google's semantic technology, the complaint states that the ads Levitte placed just didn't work. Out of the 202,528 impressions on parked domain pages, Levitte got 668 clicks and zero conversions.

Levitte's ads also appeared on error pages, through Google's AdSense for Errors program. With 1,009 impressions, 25 clicks and zero conversions, he had little to show for it.

"Domain and error page ads accounted for approximately 16.2% of all clicks on plaintiff's ads during his campaign, yet did not result in a single person completing the online form on the site, or contacting the plaintiff by phone or e-mail," the complaint states.

Levitte spent $136.11 for ads on parked domains and error pages, which works out to 15.3% of his $887.67 ad campaign.

In seeking class certification for the lawsuit, Levitte's attorneys hope to represent other aggrieved Google advertisers. "We believe it's a problem that affects all [Google's] advertisers equally," said Kimberly Kralowec, partner at the law firm representing Levitte.

If the lawsuit moves forward as a national class action and it turns out that all of Google's AdWords advertisers have spent a similar percentage of their ad budgets on low-quality ads, Google could be liable for a significant sum.

"Google includes millions of parked domains and error pages that have little or no content, and that result in practically zero conversions, in both its Content Network and its Search Network," the complaint alleges. "Given the low quality of these parked domain and error pages, advertisers would not want to spend their advertising budgets on these distribution networks. However, Google designed its network in such a way that it was virtually impossible to opt out of the AdSense for Domains and/or AdSense for Errors programs."

The complaint states that while Google allows advertisers to decide whether to place ads on Google Search, Search Network, or Content Network, there's no setting screen that allows advertisers to opt out of the domains or errors networks.

The situation changed somewhat in March, when Google altered its Site Exclusion Tool to allow advertisers to identify entire Web site categories where their ads would not appear (previously sites had to be specified individually). But the complaint claims that the Site Exclusion Tool only allowed opting out of the entire Content Network, not out of the parked domain or error pages.

Google has made further changes since then and advertisers can now find out more information about how ads on parked domains perform. Kralowec, however, said that Google hasn't fully addressed the issue though adequate disclosures and visibility into its ad system.

Google did not respond to a request for comment.

Levitte spent $136.11 for ads on parked domains and error pages, which works out to 15.3% of his $887.67 ad campaign.

Great way to advertise a business without paying Google but using Google

Lol, we should of done this Nick. Our sites would be HUGE

Lets wait and see how much damages google will claim after winning the case

He's suing over $136.11?

Lets wait and see how much damages google will claim after winning the case

I wonder how much he's paying for legal fees. I'm sure Google can provide stats from other advertisers whereby there are conversions from parked pages. I wonder why he didn't include the conversion info from his other $751.56 spend on Adwords. Probably low to zero as well which means that the problem lies not with parked pages but with his own website and marketing.

I wonder if the lawyer read his agreement with Google.

http://levitteinternational.com/

Here's the really funny bit... He IS the lawyer!

...with a crappy single page website that couldn't convert in a women's prison with a fist full of pardons!
http://levitteinternational.com/

Yesterday, we learned of a lawyer who felt that domain parking can’t convert for advertisers because he didn’t have success spending $136.11 on parked domain advertising. I met an advertising network CEO a few weeks ago who said his blood boils when someone doesn’t get conversions and claims he received “bad” or “fraudulent” traffic. After looking into his advertisers’ claims, that CEO said he often finds that other advertisers are converting rather well with the same traffic.

We have anecdotal evidence that parked domain names convert for advertisers. Efficient Frontier released a case study, and owners of large portfolios have run their own campaigns with success. I’ve even had great conversions on Dark Blue Sea’s ROAR network, which I understand is delivered via pop-ups on parked domains.

So I decided to run the numbers on a campaign I’m running on Google. The campaign promotes a free report download and requires the visitor to provide their email address. In other words, it’s a lead generation campaign similar to the lawyer’s campaign.

Not to my surprise, parked domains covert quite well.

Here are my stats for the month to date:

Search + Search Network 19.8% conversion rate, $.77 per lead
Content Network (not including parked domains) 23.9% conversion rate, $.74 per lead
Parked Domains 23.3% conversion rate, $.75 per lead

As you can see, both the content network and parked domains are outperforming search ads! Get this — some of the domains that are converting are being promoted through arbitrage, too. And for those people that complain it’s not fair that parked domain names are sometimes served as part of the search network, here’s another surprise: almost all of the conversions I have on parked domains are from the search network feed, not the content feed. In other words, without include parked domains in the Search Network numbers google provides, the search network would be performing relatively worse.

Proof that Google’s smart pricing works? Perhaps, but I’m paying the same amount per click on the search network as the content network.

If I want to further refine the campaign and get better conversions, all I have to do is block the domains that aren’t converting well. But one thing is for sure, I’d be stupid to block the domain channel for my ads. And it’s clear that parked domains names can provide value to advertisers.