Understanding Click Fraud

Click fraud is becoming a hot issue in the online marketing industry. Google faces a $90 million settlement involving fraudulent clicks to customers using the company’s AdWords pay per click advertising system.

Click fraud occurs when a search engine visitor clicks on an ad for the intended purpose of charging the payer of the ad for each click. Some website owners click on their competitors’ ads to drive up their costs and lower their return on investment.

Fraud operations have cropped up to take advantage of this technique. These groups employ scripts that simulate humans clicking on ads. This type of fraud is a bit easier to catch. Huge numbers of clicks appearing to come from just one, or a small number, of computers or a single geographic area look highly suspicious to the advertising network and advertisers.

Although paid search providers such as Google and Yahoo! Search Marketing are making efforts to detect and prosecute click fraud, fraud can be extremely hard to detect, whether it’s a script or a human. A wide network of people can be clicking on ads from different regions, without a steady pattern. Advertisers across the globe must ask themselves the same questions: How do I detect click fraud? How can I protect myself? What are the engines doing? Here are some tips that can help you protect yourself and your advertising dollars from click fraud.

Detecting Click Fraud

Detecting click fraud is not impossible when you are armed with the proper knowledge. Most of the burden rests on the engines to detect and deter fraudulent clicks. However, there are plenty of things web owners can do, and should do, to stop it before it starts.

1. Carefully monitor your clicks. Pay attention to your daily, weekly or monthly clicks so that you understand the trends of your different keywords and campaigns. You’ll be able to detect sudden increases in clicks or click through rate and report anything unusual.
2. Log server activity. Keep detailed logs of all the pages requested by browsers. The logs will allow you to pinpoint any suspicious clicks.
3. Remove your ads from certain pages. Temporarily or permanently removing ads from certain pages can protect you from being a victim of click fraud. This works well if you can figure out which ads are being targeted.
4. Always report suspicious activity. If you suspect something isn’t right, report it to the search engine. Be sure to provide them with as much information as possible, including information from your server logs.

When you are monitoring your data, look for the following information:

• Clicks from IP addresses outside of the country you are advertising in
• Keywords that don’t normally spend a lot of funds suddenly spending well
• Sudden increases in spend (aside from seasonal increases in your sector)
• Sudden increases in traffic from any one affiliate in the PPC engines network
• Lots of clicks from one IP address

Reporting click fraud is your best defense, so you must be watching your campaigns and be prepared to detect suspicious activity. Click fraud is like any other business risk; you should take appropriate steps to reduce the risk and gather firm data of any fraud. If you do have evidence of fraud, calmly and clearly present your data to the engines. Serious engines will listen to you and investigate on your behalf.

So what are the search engines doing about it?

Google and Yahoo! are the two largest paid paid search advertising providers, and the security of their clients is very important to them. They consider detecting click fraud to be crucial in serving their customers to the best of their ability.

Google uses technology that analyzes clicks and impressions to find patterns. This system aims to identify invalid clicks and filter them out before a customer is billed for them. Google will also refund any invalid clicks that are later found to be invalid. Some of the items that Google looks at to determine validity are the IP address, the time of the click and duplicate clicks. Google also has a team that uses specialized tools and techniques to examine individual instances of invalid clicks.

Yahoo! Search Marketing has a “Click Protection System” to help protect their users from paying for invalid clicks. Yahoo! evaluates nearly 100 criteria to determine legitimacy of clicks, including IP addresses, session information, cookies, networks, browser information, and time of the click.

The Click Protection Software is monitored for weaknesses. They never want to assume that they know every way an invalid click can be caused.

All search engines have a compelling interest in running a fair and trusted marketplace. Although click fraud can be difficult to completely stop, the engines know that this is something they must control.