Click fraud is a very real concern for anybody who is doing CPC marketing. Click fraud involves any artificially created click on your ad. A human being can create these clicks, or a software program called a clickbot can create them.
These clicks unfortunately cost you money.
The hard part is identifying click fraud. In order to do this you need to constantly monitor your campaigns, keeping a watchful eye out for any abnormalities. You need to pay attention to your campaign’s established trends and watch for any divergence in those trends. For example, suppose you have an Adgroup that averages an 8% CTR. Minor variations to this CTR are normal and expected. What happens, however, if this Adgroup spikes to a 15% CTR without warning? What do you do?
The first thing you need to do in such a case is check to see if there are any logical reasons for this spike. For example, was there some media coverage on your industry that would cause an increase in searches for your keywords? There can be perfectly reasonable explanations for unexpected traffic spikes. Did you experience an increase in sales? This is another good indicator of real traffic. If your sales correspond to your traffic everything should be fine.
If the traffic still seems fraudulent you need to gather as much information as possible on the incident. Record the time period when it happened and what keywords were affected. Check your server logs and look for any suspicious looking IP addresses or user agents. Once you have collected all relevant information you need to contact Google support and allow them to take over the investigation. They will want access to your server logs and any other pertinent information. They are very good at what they do, and you need to be patient as their investigation can take some time.
For example, at one point I had gone away on a camping trip and had no access to a computer, or an Internet connection to monitor my campaigns. When I returned I found the costs and traffic associated with one campaign had spiked dramatically, costing me several thousand dollars. I instantly knew that a click bot hit me. I contacted Google and an investigation was started. A few months later the investigation was concluded and it was verified that I was the victim of click fraud. I was reimbursed for all the money the click fraud had cost me and life went on.
If you spot click fraud Google is very competent with their investigation and they will reimburse you if click fraud is indeed proven. If it is determined that click fraud did not occur, you will be given an explanation describing how this was determined.
There are many different reasons for click fraud. The most common is for financial gain. This usually occurs on the content network. Here people get paid every time someone clicks on a Google Adsense ad on their site. Some people get greedy and try generating artificial clicks in order to make more money. Sometimes they try to click on the ads themselves, but this doesn’t work well because of the repetitive IP address.
Software programs called click bots might be used. These programs use multiple proxy servers to beat the repetitive IP problem, and can also randomize behaviour to make the traffic seem more natural. These programs vary in sophistication and are harder to spot.
In the recent past it was discovered that there are companies in places like India, where labour is very cheap, which use real people to click on ads. These people are spread out in different locations making the IP addresses all different and non-proxy. Moreover, they are real humans so their actions can seem more natural. This scheme can also be difficult to spot.
Other reasons for click fraud include malicious competitors. In this case a competitor wants you to either stop advertising on a keyword, or lower your bid. This often happens as a result of bidding wars. It also often takes place on the actual search pages. Any of the various methods of generating artificial clicks can be used, and the result is still the same. Increased costs.