Sunday, April 20, 2008

Click Fraud: Defining the Context of a Problem

Click Fraud: Defining the Context of a Problem
Copyright © 2006-2008 Jim Hedger, All Rights Reserved

Everyone involved in search marketing considers Click Fraud a problem issue though to what degree problems exist depends on who is looking at the issues. Pay-Per-Click (PPC) search advertising is the golden goose that makes the major engines financially viable but the success of the PPC model, mixed with the nature of the 'net can make PPC advertising a double-edged sword for advertisers.

This is the first piece in an ongoing series on Click Fraud we will be running in SiteProNews and SEO-News. Many in the search marketing and search advertising industries would prefer the issue be avoided altogether. Paid search advertising has grown to become a rapidly expanding multi-billion dollar industry, much of it based on the success of PPC.

According to a recent study, "US Online Advertising Forecast, 2005 to 2011" (http://www.marketingvox.com/archives/2006/07/24/ online_advertising_to_reach_nine_percent_of_total_ad_spend_by_2011/), by JupiterResearch, $6.5-billion will be spent on search advertising in 2006, a figure that is projected to jump to $11-billion over the next five years. Click Fraud, which is estimated to range from 5% - 15% of PPC traffic (some estimates are as high as 20% - 35%) could therefore be a factor in $350-million to $975-million worth of click-transactions. There is a lot of money involved in PPC and the tremendous weight of financial interest lurks somewhere behind virtually any discussion of the issue, especially a discussion about Click Fraud.

The share values of giants such as Google, and to a lesser degree, Yahoo!, Time Warner, AOL, IAC and MSN rely on advertiser confidence in paid search advertising. A massive amount of search marketing spam is created to generate high PPC payouts, which are shared between the search engines and web publishers. There are a number of legitimate SEM agencies turning a tidy profit by charging a percentage of the monthly or annual ad-spend of their clients. With so much money at stake, it's little wonder the biggest players in the search engine and search marketing industries would want to keep the issue under wraps.

This series is going to explore issues associated with Click Fraud from several angles, starting with an attempt to define the issue from a few different points of view. In subsequent articles we will focus on motives and opportunities, technologies and techniques, detection and deception, and alternative search based advertising methods outside the realm of PPC. We expect to touch on a number of other points, many of which will overlap from article to article, during the development of the series.

A good place to start would be an attempt at defining what, exactly, constitutes Click Fraud. In this case, the breadth of the beast is measured in the eye of the beholder.

The Wikipedia (http://en.wikipedia.org/wiki/Invalid_click) offers the best definition saying, "Click fraud occurs in pay per click online advertising when a person, automated script, or computer program imitates a legitimate user of a web browser clicking on an ad, for the purpose of generating an improper charge per click. Click fraud is the subject of some controversy and increasing litigation due to the advertising networks being a key beneficiary of the fraud whether they like it or not."

Of the search engines themselves, Google provides the simplest, most direct definition. Yahoo! offers a 13-point FAQ that fails to actually provide an exact definition. MSN and Ask also provide fairly good descriptions of the issue and remedies for advertisers. None of the major PPC providers offer advertisers an indication of how complicated click fraud analysis can be.

Google's Definition

Google doesn't like to use the term Click Fraud, preferring a phrase that conveys far fewer implications, "invalid clicks".

According to the Google AdWords Learning Center (http://www.google.com/adwords/learningcenter/text/19457.html), "Invalid clicks are clicks generated by prohibited methods. Examples of invalid clicks may include repeated manual clicking or the use of robots, automated clicking tools, or other deceptive software. Invalid clicks are sometimes intended to artificially and/or maliciously drive up an advertiser's clicks and or a publisher's earnings."

Google provides instruction on how to deal with and report suspicions of click fraud and outlines how to initiate an investigation.

Yahoo!'s Definition

Yahoo! doesn't mind using the phrase Click Fraud. A prominent text-link on the front page of the Yahoo! Search Marketing site leads to the Y!SM Click Fraud FAQ (http://searchmarketing.yahoo.com/legal/clickfraud.php) page.

On it, Yahoo! says, "Click fraud is generally considered to be clicks made with bad faith with the sole purpose of generating a charge to the advertiser with zero possibility of a legitimate site visit or transaction occurring. We agree." No clear definition of Click Fraud is provided, aside from Yahoo!'s agreement with a general statement describing it. Apparently, defining click fraud is complicated.

The FAQ suggests it is difficult to judge the purpose of clicks through its network because it is technically impossible to read people's minds. (Ed. note: We agree.) Instead, it is forced to fall back on analysing click behaviour, spotting identifiable patterns in click traffic, trying to define the faith of each click. Yahoo!'s FAQ continues, saying, "In terms of identifiable behavior, we define click fraud as detected illegitimate bots and certain repetitive clicks."

The next paragraph on the FAQ page should be of concern to advertisers. It introduces the phrase, "Traffic Quality Clicks", a phrase describing all clicks that are not generated by (identifiable) click fraud. This includes, "... clicks from domestic and international users and clicks from our various distribution channels and products." A small but scary number of Yahoo!'s distribution channels and products have been linked to malware, spyware and click-bot networks, as a coming article in this series will examine. Come to think of it, someone at Yahoo! needs to take a look at that FAQ page immediately, we're going to and we thing Y!SM marketers should as well.

Microsoft's Definition

MSN, which broke into the PPC market later than its rivals did, shares Google's aversion to the phrase "Click Fraud", agreeing to refer to the problem of "Invalid Clicks" (http://advertising.msn.com/microsoft-adcenter/faqs/invalid-clicks). An FAQ entry offers a short and direct definition of what they consider an Invalid Click saying, "An invalid click is a type of non-billable click resulting from user error, malicious activity or other factors potentially indicative of fraudulent activity."

In a subsequent statement, MSN acknowledges invalid click activity will occur saying, "Any unexplained changes in click activity should be reviewed for validity." The FAQ section goes on to explain how to monitor and appeal click charges.

Ask's Definition

Ask's new Sponsored Listing PPC considers Click Fraud to be an industry problem. Ask's Traffic Quality page (http://sponsoredlistings.ask.com/trafficquality.php) opens with the headline, "We protect you against click fraud."

Ask suggests, "Valid clicks are clicks on a listing that are generated by humans, whose intent we judge to be to engage the advertiser's site (such as to make a purchase, register for services, or navigate content.). Invalid clicks are clicks generated by robots, systems or software whose intent we judge not to be to engage the advertiser's site."

The rest of the page goes into great detail describing how Ask identifies invalid click behavior, mentioning several factors it uses to assess the validity of each click.

Industry Analysts

In a March 26, 2006 iMediaConnection (http://www.imediaconnection.com/content/8830.asp) article analyzing Google's settlement of a major click fraud case called Lane's Gifts v. Google, Clickfacts (http://www.clickfacts.com) cofounder, Michael Caruso provides a well rounded description,

"With pay-per-click ads that show up next to search results, the more times an ad is clicked, the more the advertiser pays. With contextual ads, where ads are placed on third-party websites, pay-per-click ad revenue is split between the publisher and the search company, such as Google.

Sometimes publishers click their own sites to get more revenue from the search engines. Webmasters sometimes form partnerships to click on each other's advertisements. Sometimes people click on their competition's ads, just to drive up their costs. CNET reported, "...the chief executive of an internet marketing company, enjoys clicking on his rivals' text ads ... his competitor must pay as much as $15 each time." There is also bot-like software all over the internet through anonymous "proxy" servers scattered in far-flung locales, creating the illusion that visitors are logging on from all over the place, masking the traffic's true origin.

Then there are even software companies with names like Fakezilla that sell traffic simulators online intended to make sites appear popular and thus boost ranking on search engine pages. They are also used for click fraud."

Six definitive acknowledgements of invalid click activity later, we can all agree that click fraud is a very complicated and very real problem. Stay tuned as we explore it.

About The Author:

Search marketing expert Jim Hedger is one of the most prolific writers in the search sector with articles appearing in numerous search related websites and newsletters, including SiteProNews, Search Engine Journal, ISEDB.com, and Search Engine Guide.

He is currently Senior Editor for the Jayde Online news sources SEO-News (http://www.seo-news.com) and SiteProNews (http://www.sitepronews.com). You can also find additional tips and news on webmaster and SEO topics by Jim at the SiteProNews blog (http://blog.sitepronews.com/).

No comments: