This morning, many Yahoo advertisers received an email from Rust Consulting letting them know, they are included in a legal settlement over Yahoo search ads. The settlement web site, which is at inreyahoosettlement.com explains that the “lawsuit alleges that customers contracted for targeted ad placements through two products, “Sponsored Search” and “Content Match” (and predecessor products provided by Overture Services, Inc. and GoTo.com, Inc.) and that Yahoo! breached its contract with its customers by allowing Yahoo! ads to be displayed in spyware, domain name parking sites (bulk registration sites), pop-ups, pop-unders and typosquatting sites.”

Those eligible in the settlement get a refund of $20, whereas the lawyers get $4,170,000.00, plus reimbursement of expenses of approximately $100,000. You can see all the court documents over here and the email sent to eligible advertisers over here.

Finally, if you want to get your $20, you need to submit a claim online by March 22, 2010. If you are not happy with the $20, you can always sue Yahoo directly, but make sure to exclude yourself from this settlement by December 14, 2009.

Today, I received two click fraud reports from two different companies. The first was from Anchor Intelligence and the second was from Click Forensics. Anchor Intelligence reported a click fraud rate of 22.9% for the second quarter of 2009, up from 21.7% in the first quarter. While Click Forensics reports a click fraud rate of 12.7% in the second quarter of 2009, down from 13.8% in the first quarter. That is a 10 plus point difference in the click fraud rate, reported by these two agencies. Who should you believe?

Click Fraud can be measured on many different levels. Anchor broke out their click fraud rates by two types of “invalid” clicks, “attempted click fraud” which was the 22.9% rate and “innocuous invalid” which was 4.2% rate, a total of 27.1% in invalid clicks for last quarter.

You can obtain the full report from Anchor Intelligence over here. Click Forensics report can be seen over here

Each company has their own data sources, measurements and definitions for reporting on click fraud. Like always, when reporting on click fraud, you need to keep these things in mind.

The New York Times reports that Microsoft is taking the click fraud battle to court.

“Microsoft filed the civil complaint on Monday in United States District Court in Seattle against Eric Lam, Gordon Lam and Melanie Suen, of Vancouver, British Columbia, along with several corporation names they were believed to have used, and several unnamed parties.”

The Times quotes Tim Cranton, a Microsoft attorney, who says they’ve decided to get more active in fighting fraudulent clicks on web ads. “The theory is you can change the economics around crime or fraud by making it more expensive.”

The lawsuit stems from auto insurance advertisers, who complained as far back as March 2008 about unusual spikes in paid traffic. The Times says Microsoft investigated those complaints and found similar oddities in other paid ads, including those related to the World of Warcraft game. Their investigation ultimately led to the defendants, who refused to comment for the Times’ article.

The Times does explain what Microsoft thinks the defendants were doing:

“Microsoft’s theory is that Mr. Lam was running or working for low-ranking sites that took potential client information for auto insurers. The complaint said that he directed traffic to competitors’ Web sites so they would pay for those clicks and exhaust their advertising budgets quickly, which let the lower-ranking sites that he sponsored move up in the paid-search results.”

Microsoft seeks at least $750,000 in damages.

Click fraud has been estimated to affect as many as 17% of paid clicks, but search engines have disagreed with those estimates.

As businesses become more virtual in organization and structure, and more workers become digital nomads, the question of  whether having a physical, real-world address on your website will have any effect on your rankings becomes more important. In this article I’ll take a look how a real world address can have some impact on your sites organic listings.

Let’s look at the auto repair sector. We’ll be looking at national chain/franchises, whether they have a corporate address in the footer or on the site, public or private whois data, does the whois data match and do they have a crawlable/indexable location directory. Below is a matrix of all the data:

* – on subdomains

With all of the possible cities in place we’ll look at searches for [ciytname auto repair] we’ll be looking at two separate issues: the local/map onebox listing and the standard listing in the normal SERP’s.

Data for Google local onebox results comes from a variety of sources including telephone listings, so having an address on the site provides very little benefit to getting your website to appear there.

At the surface it would seem that having an address on the site has very little or no influence on organic rankings, as Meineke and Midas are the only websites that are showing up in the organic listings. Looking back at the first chart we can see that while everyone had public whois data available, the only two who had matching on site addresses and whois data where Meineke and Midas, are the only two who showed up in the organic listings.

But look at the SERP’s for Chicago Auto Repair and Charlotte Auto Repair. Midas and Meineke are the only websites that rank—but they are also the only ones that don’t have the city name in the title—nor do they have the address or city name anywhere on the page.

The key point is not to think of having an address or having a matching address as an on/off switch for ranking. Instead think of it as part of a websites overall trust score. Having a private registration, no address, or non matching addresses may not act as a negative, but having matching on-site addresses seem to act as positive. Again it’s not required for a website to rank, but it does seem to help.

It’s also important to note that Google was able to attribute the city data to the home page, when in this case that information was only present on about us and contact pages of the site. Trust has been a component in Google’s algorithm for some time even before Eric Schmidt’s comment about the internet being a cesspool and brands were the key to sorting it out. If you go back to the Google Librarian newsletter of 2006 , you’ll see Google gives some advice on things to look for when determining if a site is trustworthy. It doesn’t take a huge leap of faith to assume some of these factors are probably already built into the algorithm.

How can webmasters and site owners take advantage of this to help their websites seem more trustworthy to Google? Using fake or false whois data is not only risky but put you at risk of losing your domain as it’s against ICANN regulations. Additionally if you’ve spent any time testing Google maps you’ll also know that Google recognizes real address, or least those it has never seen before, identifying them with “did you mean” response. Using a post office box isn’t a viable solution either as Google isn’t able to locate them on a map.

Another option is using a Mailboxes, Etc. or UPS mailbox location. In a post 9/11 world the postal regulations have required these companies to stop using the “suite” designation for mailboxes. They are now required to use PMB # (Personal Mail Box) for all addresses. In practice if you put suite or just # you would probably get your mail. How closely ICANN regulations are in line with 9/11 postal homeland security restrictions is uncharted territory.

If you’ve come to the realization that Google is all but forcing you to create a profile , you might have learned something in the account verification process. One of the two methods Google uses to verify profiles is via telephone. I’ve had several phone numbers, many of which have been around for almost 5 years, however the only one Google was willing to verify was the listed one in the phone directory. Is Google using the same or similar technology to verify address data that’s on your site or listed in your whois file, only the folks working at the Googleplex know for sure. However I’ll firmly grab the ear-flaps on my conspiracy theorist tin foil hat and say at the very least it’s plausible, and it wouldn’t surprise me if address information worked it’s way into the algorithm in the next 3 to 5 years.

In conclusion, putting a real world verifiable address on your website appears to increase its chances of looking more algorithmically trustworthy to Google. The effects may not be immediate, but in my opinion, are forward-looking steps to give your website a bonus your competition may not be getting.

Click fraud is worse today than ever according to an announcement from tracking company Click Forensics.

Click Forensics estimates the overall average click fraud rate for Q4 2008 at 17.1%, the highest level since they began tracking it in 2006. That rate is up from 16.0% in Q3 2008 and from the 16.6% rate reported in Q4 2007. Click fraud estimates on the search engine content networks are said to be at 28.2%, up from 27.1% in Q3.

They also estimate that botnets were responsible for 31.4% of click fraud in the last quarter of 2008, up from 27.6% in Q3, and also the highest level yet.

“We’ve started to see old schemes like click farms reemerge,” says Tom Cuthbert, Click Forensics’ president.

As always, keep in mind Chris Sherman’s advice when looking at these figures:

At first glance, these numbers may seem alarming, but they may not take into account the discounting of questionable clicks done by most search engines. In fact, Google has stated that click fraud amounts to just .02% of all clicks after it allows for other non-converting clicks. See Danny’s Google: Click Fraud Is 0.02% Of Clicks for a detailed look at how Google analyzes clicks and why its number of fraudulent clicks is so much lower than that reported by Click Forensics.

Click Forensics has released their estimated 3rd quarter click fraud numbers that they track. The numbers remain pretty much the same, at least as it pertains to the overall estimated industry average click fraud rate. The rate is now 16.0%, down from last quarter’s 16.2%.

The main change was that botnets were said to be responsible for 27.6 percent of the click fraud, up 10% from last quarter.

The average estimated click fraud rate of on the content networks, including Google AdSense and the Yahoo Publisher Network, was 27.1 percent, which is down from the 27.6 percent rate of the previous quarter. Click fraud was said to originate mostly from Russia (4.9 percent), France (4.8 percent) and the U.K. (3.5 percent) according to Click Forensics.

Like always, keep in mind Chris Sherman’s advice when looking at these figures:

At first glance, these numbers may seem alarming, but they may not take into account the discounting of questionable clicks done by most search engines. In fact, Google has stated that click fraud amounts to just .02% of all clicks after it allows for other non-converting clicks. See Danny’s Google: Click Fraud Is 0.02% Of Clicks for a detailed look at how Google analyzes clicks and why its number of fraudulent clicks is so much lower than that reported by Click Forensics.

Google allies with click-fraud-detection firm Click Forensics from ComputerWorld reports Google has now agreed to work with Click Forensics to aid in the detection and reporting of search ad click fraud.

Specifically, Google said they would now accept click fraud submissions through the product, FACTr. FACTr is a product Click Forensics created with the help of Yahoo to gather and submit click-quality reports. Google will now be accepting these reports electronically, hopefully streamlining the process for advertisers to get refunds for approved click fraud. Looksmart and Miva are also now accepting them, along with Google, in news that Click Forensics announced last week.

As you may remember, Google and Click Forensics have not always seen eye-to-eye. But this is a nice step to see from both parties involved. As a matter of history, Yahoo partnered with Click Forensics back in March of this year.

Click Forensics has released the second quarter results for 2008, showing a slight drop in the estimated click fraud. Last quarter, Click Forensics reported a 16.3 percent rate, and they are now reporting a 16.2 percent rate for Q2 of 2008. However, the rate is up from a year ago, which was a 15.8 percent rate in Q2 of 2007.

The main number that stood out was that “traffic from botnets was responsible for more than 25 percent of all click fraud traffic in Q2 2008,” said Click Forensics. Click Forensics believes the botnets increase was the reason why the click fraud rate did not drop more than it should have.

Here are the latest charts:

As always, keep in mind Chris Sherman’s advice when looking at these figures:

At first glance, these numbers may seem alarming, but they may not take into account the discounting of questionable clicks done by most search engines. In fact, Google has stated that click fraud amounts to just .02% of all clicks after it allows for other non-converting clicks. See Danny’s Google: Click Fraud Is 0.02% Of Clicks for a detailed look at how Google analyzes clicks and why its number of fraudulent clicks is so much lower than that reported by Click Forensics.

The only difference now is that Click Forensics and Yahoo have teamed up to build a tool to track click fraud.

Google Sued For Selling Ads On Parked Domains from InformationWeek reports Google was slapped with a class action lawsuit over the AdSense for Domains product. In short, advertisers are upset with the quality of traffic driven through that program and feel that their money was taken from them without any return.

The suit was filed on Friday in a U.S. District Court in San Jose, California. The law firm, Schubert Jonckheer Kolbe & Kralowec, is based in San Francisco. The one suing is actually a lawyer, suing under Levitte International for his advertising campaign he ran from June 1, 2007 through August 18, 2007. The campaign received 202,528 impressions via the AdSense for Domains program. The campaign also was shown on the AdSense for Errors program, which received 1,009 impressions, 25 clicks, and zero conversions. The AdSense of Domains and AdSense for Errors traffic and impressions cost him $136.11.

Levitte hopes that other advertisers join his lawsuit. “We believe it’s a problem that affects all [Google's] advertisers equally,” said Kimberly Kralowec, partner at the law firm representing Levitte.

It is important to note that in August 2007, when this lawyer’s campaign was still active, Google announced that they will be enabling a feature to allow advertisers to opt out of AdSense for Domains traffic. Google began beta testing this feature in November 2007.

In addition, the domain park (AdSense for Domains) is not new to getting Google in trouble. We reported it landed Google in a trademark case back in January 2007.

Click Forensics, Yahoo take on click-fraud cases from News.com reports that in part of the Click Forensics & Yahoo partnership, they have teamed up to build a new software tool to detect click fraud, named Factr.

Factr, Fully Automated Click Tracking Reconciliation, automatically queries Yahoo’s search ad platform when Click Forensics detects click fraud. The tool provides automated detection of click fraud and streamlines the refund process with Yahoo.

Steve O’Brien, vice president of marketing for Click Forensics, said:

We’re trying to automate what has been a tedious manual process for online advertisers. With a click of a mouse, advertisers can say, ‘I want this report submitted to Yahoo, and I want verification from Yahoo that I’m getting what I paid for.’ We handle the rest in (the) background.

O’Brien added that they are “in conversations with” Google about using a similar tool in the AdWords platform.