Ad fraud has been a problem for some time, and a well-known problem at that. Back in 2016, a Mashable headline read, “Ad fraud could become the second biggest organized crime enterprise behind the drug trade.” Even Virginia Senator Mark Warner relayed this stat in a tweet to emphasize the problem.
For most advertisers, ad fraud remains a problem to this day. It is estimated that 25 percent of all clicks on ads are fraudulent. The reason why it’s still a problem is twofold. For one, ad fraud makes the bad guys money. Secondly, most people don’t know how to fight back against ad fraud.
Fighting back against ad fraud is tough, though. You have to know what you are fighting against. To do this, you need to know the different types of ad fraud, which are broken down into bots controlled by malware and human fraud.
Ad fraud bots, or zombies, are computers that are infected with malware and report back to a command-and-control (C&C) server. The C&C server instructs the bot on what to do. If the instructions are to go out and click on ads, that is what the bot does.
The Chameleon botnet—a network of 120,000 host machines infected with malware that was discovered in 2013—for example, was specifically instructed to commit ad fraud, to the tune of $6 million worth of fraud per month.
Mobile apps are getting in on the game as well. Apps such as Clone Camera, which was downloaded close to a million times, contributed to the more than $6.5 billion that was expected to be lost to bot-generated ad fraud in 2017.
Humans also do their part to defraud advertisers. Large groups of people make up human fraud farms, and much like the malicious bots, their job is to go into websites to click on ads, fill out lead gen forms, or even complete e-commerce transactions using stolen credit card numbers. Because this type of fraud mimics real traffic more effectively than botnets do, it is harder to stop if you don’t have the right ad fraud detection solution in place.
This type of fraud, termed sophisticated invalid traffic (SIVT), is specifically meant to avoid patterns and behaviors that may raise a red flag for ad fraud detection solutions. Humans, and even many bots, generate this type of traffic when committing fraud.
General invalid traffic (GIVT), on the other hand, is generally seen in legitimate bots such as search engine crawlers. They are programmed to move from page to page after a set amount of time, follow links, and so on. This type of low-level invalid traffic is much easier to detect.
The Cost of Ad Fraud
Regardless of whether it’s perpetrated by malicious botnets or humans behind keyboards, fraud costs advertisers money. Botnets are cheap to rent, with some costing less than $9 an hour. For just more than $2,000, a bad actor can run an ad fraud campaign that lasts a month.
Mobile apps are cheap to create as well. With enough downloads, an army of mobile devices are out there working for illicit individuals to cash in. In fact, ad fraud is so profitable for criminals that it was expected to cost advertisers $42 billion in 2019.
Ad Fraud Detection Software Is the Answer
No company wants to contribute their advertising budget to a scammer. Those ad dollars should be going toward bringing in new customers, alerting customers to new products or services, and growing your business.
Ad fraud solutions are able to detect fraudulent traffic by gathering data points from your existing traffic and analyzing behaviors on your ads. These solutions are able to determine whether traffic and clicks are generated by botnets to eliminate the chance that you are lining criminals’ pockets rather than building your business.
The best ad fraud solutions, however, are able to go even further than spotting botnet traffic. With sophisticated algorithms, they can determine whether traffic is originating from a human fraud farm and even stop fraudulent transactions from happening to reduce charge-backs generated by fraud.
Stopping financial losses is not the only way an ad fraud solution helps your company grow. By eliminating fraudulent traffic, you also have the ability to look at a much cleaner data set for your advertising metrics.
Each time a bot or someone from a human fraud farm sees your ad, you have a fraudulent impression. If they engage with that ad, if they convert, or even if they ignore it, your data is less accurate. That traffic isn’t real, but it is showing up in your numbers.
By eliminating the pollution that illegitimate traffic causes, you can tell which campaigns are the most effective because you are looking at reliable data. Your time, energy, and money are better spent on these high-performing campaigns. When you are able to focus on the ones that are working for your business, you have a better opportunity to improve those campaigns’ performance even more.
Think of all the opportunities that are missed because of ad fraud as well. Illegitimate impressions run up numbers. When you set limits on pay-per-click campaigns, fraud can cause you to reach those limits without getting your ads in front of the real people you created them for. The fraudulent activity not only keeps potential customers from seeing your ads, but it also puts you at a competitive disadvantage when your competitors’ ads start showing up in the place of yours after your daily limit is reached.
Finding the right solution to fight back against ad fraud is not always easy. With any problem come solutions that aren’t always the right ones. If you are counting on an ad fraud solution to help grow your business, you need to be confident in its ability to address fraudulent activity while avoiding false positives that turn legitimate business away.
To see how a best-in-class ad fraud solution works to help companies grow, download this case study, which shows how Anura helped Digital Media Solutions address this very problem. By applying similar methods, you, too, can take the money, time, and resources that ad fraud was costing your business and leverage them for growth instead of waste.
Richard Kahn
