In addition to a major global economic downturn, click fraud is considered one of the few things that could impede the growth of the click advertising business. No company knows this better than Mountain View, California-based Google Inc, which relies on click ads for the vast bulk of its fortunes.

Google was sued early last year by advertisers who claimed the search engine collected revenues on fraudulent clicks. The advertisers won a $90m settlement, which was recently approved by a US court.

Google has been working to improve transparency about invalid clicks for our advertisers and we’re pleased to join an industry-wide effort with the same goal, said Google’s business product manager or trust and safety, Shuman Ghosemajumder, in an email.

And while Google said it manages invalid clicks well to ensure small losses to its advertisers, the company has joined the anti-fraud group to also ensure advertisers have trust in the industry overall, Ghosemajumder said.

The new group is called the Click Measurement Working Group and is run by the Interactive Advertising Bureau and the Media Rating Council. The goal is to devise a set of Click Measurement Guidelines to determine the difference between legitimate and fraudulent or invalid clicks on the links sponsored by the search vendors’ advertisers.

The group, whose members also include Ask.com and LookSmart, plans to foster an auditing and certification recommendation for companies that rely on clicks for revenue.

The guidelines would provide marketers with a standard for the consistent and reliable measurement of their performance-based marketing, said an MSN spokesperson, in an emailed statement. Microsoft also pointed out its adCenter has various proprietary technologies to protect advertisers from click fraud.

Google said the guidelines would raise standards and promote transparency throughout the industry, in addition to its internal methods of keeping the problem in check.

We believe that the percentage of invalid clicks that may escape detection – and for which advertisers may have been charged – is very small, said Google’s Ghosemajumder. It is unclear how exaggerated estimates of click fraud are put together, but it’s very likely that they include attempted fraud already caught and discarded by our automatic filters.

Not so, say some of the companies that track click fraud. Click Forensics LLC, which sells anti-click fraud software and runs a click-fraud detection service, said the research it publishes is the average click fraud rate of the more than 1500 advertisers that use its anti-click fraud tools.

Click Forensics chief executive Tom Cuthbert also noted that independent analyst group Outsell Inc came up with similar click-fraud figures through research that had nothing to do with ours.

Its also misleading for anyone to point the blame at independent, third-party monitoring services, who experience no financial gain when click fraud occurs, unlike search providers who get paid no matter what, Cuthbert said. We do know that the number of clicks that our clients are billed for is consistently much higher than the number of valid clicks our system identifies.

San Antonio, Texas-based Click Forensics recently released a new survey that said 14% of clicks on Internet ads were bogus. Tier one search engines, such as Yahoo and Google, has the lowest average click fraud rate, 12.8%, compared to smaller search providers, who had an average rate of as much as 27%.

Outsell recently released its own study too, of 407 advertisers that account for about $1bn in ad spending. The report estimates click fraud is a $1.3 billion problem that is significantly reducing advertisers’ pay-per-click spending and threatening the core business model of search engines like Google.

The single most telling finding in Outsell’s research was that 27% of advertisers have reduced their spending on pay-per-click advertising, by an average of 33%, and an additional 10% of advertisers said they intend to reduce their PPC ad spending. This result does not depend on definitions of what is a fraudulent click, said Outsell VP and lead analyst Chuck Richard.

A big step Google could take to support its assertion that its estimates are accurate while others are exaggerated is making available its click fraud data and methods, Richard said. He noted Google recently said it would take steps to begin to do this, and added that Yahoo and Microsoft should also be more transparent.

To determine which clicks are fraudulent requires marrying data from search engines and advertisers, Richard noted.

It is essential to know what happened on the advertisers’ sites after the search engines send the traffic to them – was there a purchase? did the user visit multiple pages? how much time did they spend on the advertisers’ pages? did they register? request additional information? – in order to get a full picture of which clicks are fraudulent, he said.

Burlingame, California-based Outsell recommends a third-party audit bureau serve that purpose of marrying data from search engines and advertisers. Richard said the newly formed industry group might lead to that happening.

He also noted that the advertisers that participated in Outsell’s research were explicitly requested to provide the net fraudulent click percent they were billed for after removing any click fraud refunds they may have received.