One of the big concerns when advertising on Pay Per Click services, is the potential for click fraud. The smaller networks aside from Google and Bing are particularly prone to this issue. Despite the measures in place that all these engines have to prevent it, it will still sometimes slip through the cracks.
However, another important thing to remember is that poorly performing campaigns aren't always indicative of click fraud. In fact what may look like click fraud may actually just be due to poor keyword management and bad ad copy. For example, if you are not using Match Types properly, you could inadvertently be causing your ads to be showing up for far too many loosely related search queries. If you are not effectively targeting users with a specific intent to convert on your site, it will take a toll on your conversion rates and eat up your budgets.
For example, if your site only sells car insurance, and you are bidding on the term "insurance" set to broad match, you are attracting all sorts of other user search queries, such as people searching for pet insurance, renters insurance, people just doing some research on insurance companies, people looking up job information about a given insurance carrier, and more. A majority of these broad matched search queries are likely not from people interested in purchasing car insurance from your site. And considering how expensive the bidding is on the keyword "insurance" it will likely eat up your budget very quickly. A better strategy would be to not bid on the term "insurance" at all, or if you do, set it to Exact Match and keep the bid within reason so as not to eat up your total budget. You should also have it isolated into it's own ad group, with a comprehensive list of negative keywords applied to it (such as "pet", "life", "home" and others). It would make more sense to go after more specific keyword phrases, such as "car insurance quotes" or "compare car insurance" and set them to either phrase or exact match. That will force the engines to only send you traffic from queries that closely match those which you are bidding on.
I won't go into too much on how to spot click fraud, since it's rather self explanatory - if you suddenly see shifts in traffic and conversion rates, that would be your first red flag. Then you need to identify the sources of that traffic - are there any referring sites or IP addresses in particular that the traffic is coming from? Another way you can take the IP address analysis a step further, is to consolidate your data down by an IP range. Sometimes the click fraud sources will use a range of related IP addresses to send their traffic to your site in an effort to make it harder to detect. So while looking at an IP address report, if no one IP address in particular stands out, try chopping off the last three digits after the final period, and doing a consolidation in Excel to see if any ranges now stand out, if nothing remarkable is there either, do it again with the next three digits.
However, another important thing to remember is that poorly performing campaigns aren't always indicative of click fraud. In fact what may look like click fraud may actually just be due to poor keyword management and bad ad copy. For example, if you are not using Match Types properly, you could inadvertently be causing your ads to be showing up for far too many loosely related search queries. If you are not effectively targeting users with a specific intent to convert on your site, it will take a toll on your conversion rates and eat up your budgets.
For example, if your site only sells car insurance, and you are bidding on the term "insurance" set to broad match, you are attracting all sorts of other user search queries, such as people searching for pet insurance, renters insurance, people just doing some research on insurance companies, people looking up job information about a given insurance carrier, and more. A majority of these broad matched search queries are likely not from people interested in purchasing car insurance from your site. And considering how expensive the bidding is on the keyword "insurance" it will likely eat up your budget very quickly. A better strategy would be to not bid on the term "insurance" at all, or if you do, set it to Exact Match and keep the bid within reason so as not to eat up your total budget. You should also have it isolated into it's own ad group, with a comprehensive list of negative keywords applied to it (such as "pet", "life", "home" and others). It would make more sense to go after more specific keyword phrases, such as "car insurance quotes" or "compare car insurance" and set them to either phrase or exact match. That will force the engines to only send you traffic from queries that closely match those which you are bidding on.
I won't go into too much on how to spot click fraud, since it's rather self explanatory - if you suddenly see shifts in traffic and conversion rates, that would be your first red flag. Then you need to identify the sources of that traffic - are there any referring sites or IP addresses in particular that the traffic is coming from? Another way you can take the IP address analysis a step further, is to consolidate your data down by an IP range. Sometimes the click fraud sources will use a range of related IP addresses to send their traffic to your site in an effort to make it harder to detect. So while looking at an IP address report, if no one IP address in particular stands out, try chopping off the last three digits after the final period, and doing a consolidation in Excel to see if any ranges now stand out, if nothing remarkable is there either, do it again with the next three digits.