Since the advent of paid search engine marketing, click fraud has been a growing area of concern for search engine marketers. Estimates of the scope of click fraud vary from 10% to 35% of all click activity. If not detected, click fraud can rob your company of both your marketing dollars and your sales. Are your campaigns safe? This article provides some tips on how to detect click fraud in your campaigns and some guidance on what to do if you find it.
So what is click fraud? Click fraud occurs when a person or a program, often referred to as a “bot†or a script, clicks on a paid search ad with no intention of legitimately visiting the advertiser’s site.
What is the motive for click fraud? In general, perpetrators of click fraud are motivated by one of two objectives: to gain a competitive advantage; or to make a commission on the click.
Who are the perpetrators? (Does this sound like a CSI episode?). In some cases they are your competitors. Competitors who are trying to get you to burn through your advertising budget quickly so that your ads stop appearing on search result pages. A second and larger category of perpetrators are website operators who are participating in a paid search engine affiliate program such as Google’s Adsense, or Yahoo’s Publisher Network, who increase their revenue share illegitimately by implementing bots or paying individuals to click on your ads. (Note: Google AdSense, and Yahoo’s Publisher Network are affiliate programs in which website owners place paid search ads on their websites and receive a commission on the clicks that occur on those ads. The restrictions for participating in affiliate programs are minimal so they are appealing to fraudsters.)
What campaigns are susceptible to click fraud? All campaigns are at risk, but campaigns utilizing content match, or the expanded search networks, have the highest level of risk since those offerings include affiliate sites.
Can click fraud be detected? The answer is, sometimes. All the major search networks utilize sophisticated click fraud programs to automatically detect and eliminate a significant number of fraudulent clicks, but these tools don’t catch them all. There are still cases where large numbers of fraudulent clicks are billed to advertisers. Search engine marketers need to closely monitor their campaigns to catch the fraudulent activity the search engines miss.
Tips for Detecting Fraudulent Clicks:
The following tips can be used a general guide for detecting click fraud in your paid search campaigns. (Note: some of these suggestions require the use of an analytics tool such as Google Analytics, or Hitbox, etc.)
1. Watch for Spikes in Activity
If you see a sudden and dramatic increase in visitors, be suspicious. Some fraudsters aren’t very subtle in their approach and sometimes click on the same ad repeatedly to drain your budget or to increase their revenue. Use your analytics tool to determine if there was a large increase in visitors from a particular keyphrase or set of keyphrases, and then determine if they all came from the same city, or IP address. If they did, the odds are its fraudulent activity.
2. Lower than Normal Conversion Rates
Fraudsters have no incentive to perform any activities on your site once they click on your ad so a common indication of fraudulent activity is that your visitor traffic increases but your visitor to conversion rate decreases. Of course there are numerous reasons that could cause this to happen, so you should use your analytics tool to identify where the increase in traffic is coming from and whether or not there are any conversions associated with that source, if there are not, the likelihood that its fraudulent activity increases.
3. High Bounce Rates from the Landing Page
Visits associated with click fraud commonly average one pageview per visit since the fraudster leaves the advertiser’s site immediately upon landing there. If you suddenly see an increase in visitors who are only visiting one page before leaving then investigate it. It could be that you have an ad that is attracting the wrong audience and they discover your site doesn’t offer what they want, or it could be another sign of fraudulent clicks. Again try to determine if these visits are being generated from a single location or a location out of your normal marketing area.4. Low Average Time Spent on Site
If you have an analytics tool capable of tracking the average time a visitor spends on your site it can use to investigate click fraud activity in a similar way as looking at pageviews per visitor. If you see visitors staying for a matter of seconds it’s another sign that you are either attracting the wrong audience (fix this by changing your keywords and/or ad copy) or you have a fraud problem.
5. A High Number of Clicks from a Single IP Address
This is one of the best ways to detect click fraud. If you see an inordinate number of visits from a single IP address in most cases its one of two things. Its either multiple visitors coming through a common proxy server such as those used by AOL, or its fraud!
6. Clicks From Foreign Countries
Click fraud can originate from any country but some countries tend to be responsible for a disproportionate amount. If you see a high level of visitor activity from a country in which you don’t normally see much, if any, visitor traffic, investigate it to see if its coming from a single city or IP address. If it is, it doesn’t automatically mean its fraud but the odds are high. Occasionally an article or a link can be posted in foreign city and it can generate an unusual amount of traffic, but its wise to use the tips noted above to evaluate the quality of this traffic before asking for a credit from the search engines.Be aware that poorly performing or poorly managed campaigns can easily be confused with click fraud. For example, if you are seeing high bounce rates on your landing pages it may be the result of a poorly written ad, or a poor landing page selection. Your visitors may simply be demonstrating destination disappointment.
How to File a Click Fraud Complaint:
If you believe you’ve been billed for fraudulent click fraud activity, you can file a claim with the search engine that generated the clicks.
Google:
File a request through Google’s Click Quality Team by filling out the form through the link below:
https://adwords.google.com/support/bin/request.py?clickquality=1&ctx=clickqual
Yahoo:
File a support request form through your Yahoo Marketing Solutions account center, or contact your customer service representative.
MSN:
File a request through Microsoft’s adCenter through the link below:
http://support.adcenter.msn.com/contactus.aspx?productkey=Adcenterss&locale=en-us
Each of the search engines will request similar information, including your account information, the time and dates when the activity occurred, the keyphrases used, and detailed information to support your claim. It typically takes 3-5 days to process requests. If you’ve properly supported your claim, the search engines will issue you a credit.
The Future of Click Fraud:
One possible solution for click fraud that is now being offered on a trial basis on Google is a cost-per-action model in which advertisers only pay if a visitor completes an action on their website such as makes a purchase, or submits a registration form, or signs up for a newsletter. This may help, but it also could raise the stakes for fraud. With a cost-per-action model, advertisers will likely pay 10-20 times more per action than they do for a click to their site. That means fraudsters will have the potential to make 10-20 times more per fraudulent action. Granted it will be harder to complete a fraudulent registration than a fraudulent click, but if the commission on that fraudulent registration is 20 times higher, fraudsters will rise to the challenge. Search engine marketers will need to continue to monitor their campaigns very closely.
