Combat Click Fraud By Blocking Low Quality Traffic
Click fraud is a major concern to most advertisers. While advertisers should monitor their PPC activity and be aware of any possibilities of click fraud, there are several tools within Google AdWords and Yahoo Search Marketing accounts which can help to protect advertisers from potential click fraud.
Many industry observers suspect that the majority of click fraud occurs on the content network, rather than in search results. This does not mean that the content network contains inherently poor traffic; many advertisers report having lower cost per conversions on the content network than search. However, the network is often misunderstood. Therefore, the first item to consider is techniques for optimizing the content network.
Second, it is worth noting on the content network that not all clicks are priced equally. Google uses smart pricing, and Yahoo uses’ quality based pricing to lower the price of a click based upon several quality factors. Therefore, when examining metrics on the content network, items like cost per conversion are much more telling than conversion rates.
However, even beyond the content network, there are several options with PPC accounts that will help combat potential click fraud no matter if it is search or content based.
How to block your ads from showing in certain countries
Yahoo has a wonderful feature that’s hidden in the administration settings section called ‘Blocked Continents‘. This allows advertisers to keep their ads from showing on different continents. If you do not ship internationally, or have global customers, then it is useful to keep your ad from showing to searchers in countries that you can’t service.
Yahoo Search Marketing is also different from Microsoft adCenter and Google AdWords in that country targeting occurs at the account level. Therefore, even if you do target internationally, you may wish to opt for one account to not show ads in Europe, and yet another account to not show ads in the United States.
How to block competitors from clicking on your ads
The IP exclusion tool on Google AdWords allows you to block specific or range of IP addresses from seeing you ads. If someone cannot see your ad, then they can’t click on it. While competitor clicking is easy for engines to catch as it’s pattern based, it can be comforting to know your competitors will no longer see your ads.
Before you go tracking all of your competitors and start blocking their IPs, please make sure you are blocking the correct IPs, or that the IP addresses are not shared. Some host providers may have thousands of users on the same IP address.
See which content network sites are sending you traffic
Google AdWords has recently launched a placement performance report that allows you to see all the websites in the content network which are sending you traffic. This gives you the ability to see where your brand is being placed across the web. The placement performance report also works with conversion tracking. So, why you might optimize your campaign for the content network, you have the option to block sites which are not converting.
How to keep yourself from showing for keyword searches
All three of the major pay per click engines, Google AdWords, Yahoo Search Marketing, and Microsoft adCenter have the ability to block your ad from showing on certain search results. The feature is called “negative keywords.” Essentially, you can tell the search engine not to trigger your ad when a certain keyword is searched.
For instance: If you sell all types of plasma television brands, except for Sony, then you might not want to show for the search ‘Sony plasma TV’. In this example, you could use the negative keyword ‘Sony’ so that your ad does not show whenever someone is looking for a Sony television.
Negative keywords are an essential tool for filtering out search queries when you do not want your ad to be triggered or shown on irrelevant searches.
How to block content network sites
With Google AdWords, there is a tool called Site Exclusion, which allows you to block any website on the content network from sending you traffic. When the tool first launched, there was a limited amount to the number of sites one could block. That has since changed, and now you can block unlimited websites. Feel free to add all of those non-converting sites to the site exclusion tool so they will no longer send you any traffic.
It is important to note that the site exclusion blocking occurs at the campaign level. You may need to enter these sites into each campaign. In addition, if you are running a site targeted campaign, it will not allow you to block websites for that campaign. Since site targeting allows you to target specific websites, hence you have control of where your ad is showing, you do not need to block websites for site targeting campaigns – just remove them from the campaign.
Determine how many clicks you did not pay for
One of the difficulties in tracking down click fraud is that your website’s analytics program rarely matches what a search engine reports for how much traffic you have received. In order to track down any invalid clicks, one must look for the difference in how many you paid for versus how many your analytics program said you received.
AdWords has an invalid clicks report that allows you to determine how many clicks were sent to your website for which you were not charged. This is often the best place to start investigation when your analytics system and your pay per click reports are not close in terms of how much traffic you received from a pay per click vendor.
When combating click fraud, follow this four step process for the most effective results.
Step one is to identify and set baselines. One baseline measure is traffic. How many clicks do various keywords receive each day or month? How many clicks do you receive from each engine? Determine what you traffic levels look like, and then set metrics in place to measure all of those baselines. The second baseline is traffic quality. This can be average page views, time spent on site, conversion rates, etc. Establish baselines for quality and total traffic that you can measure on a regular basis.
Once you have baselines in place, you will want to look for deviances from those baselines. This will give you a starting place for any investigations as well as alert you to changes in your normal traffic patterns.
The second step is to investigate changes from your baselines. When one of your baseline metrics suddenly spikes, investigation should occur. This analysis is comparing your metrics to search engine metrics. Examining referrers, running invalid clicks report, noting changes in search behavior (did a news story just break which is triggering your keyword?), etc, should help you narrow down why you had a change in your baseline.
Third, block non-performing traffic. Not all traffic converts equally, and not all non-converting traffic is fraud. Poor traffic could be because of the keywords you are bidding on, the websites sending you traffic, or changes in matching algorithms. It is important that you examine your traffic sources and block traffic that is not helping you to reach your website’s goals.
Last, if you cannot determine why a spike in one of your baselines occurs, contact the search engines. Every pay per click engine has a fraud investigation team which can help you determine if fraud occurred, and what are the next steps.
The best way to combat click fraud
is to protect yourself by knowing which settings you can utilize in blocking poor quality traffic in the first place. Each industry is very different in terms of the buying cycle, user engagement, and traffic levels.
It is important that you understand how your industry performs online, and then take steps to receive the highest quality traffic for your industry to your website. Being proactive in blocking traffic can save you many headaches, and time, later on from pouring over log files trying to determine if an event was fraud or legitimate. Take the time now to block traffic that has little chance of helping your business meet its goals.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.