What is click fraud and how can you prevent it?

With the rise of programmatic buying in South Africa, the click fraud threat has been raised by publishers and advertisers alike. While click fraud is a reality if left unmonitored, there are ways to maximise brand safety in private exchanges by ensuring that the correct validation processes are in place. Below, we discuss the various types of click fraud, their implications and the ways in which ad exchanges can protect advertisers and publishers.

What is click fraud?

In the simplest terms, click fraud is a method of generating illegitimate clicks on digital advertising. This can either be done manually through click farms or more commonly, through automated software and online bots. Click fraud affects both Pay Per Click (PPC) advertising, which creates an artificial price hike for search terms and on display ad impressions, where fraudulent clicks create a false economy in which advertisers think they’re getting higher click-through rates (CTR) than they’re actually achieving.

While statistics on the amount of click fraud that occurs frequently differ, in 2014 it was reported that 36% of traffic on some of the major programmatic platforms was generated through click fraud. The goal of generating this fraudulent traffic is to create fake clicks so that publishers earn more advertising revenue. It is both an unacceptable practice and a heavy toll on brands-which land up paying for illegitimate traffic.

Types of click fraud

Click Farms

Click farms are the ‘sweat factories’ of the online world. Thousands of low paid workers are hired to sit behind a computer and literally continually click on specific adverts. While this is probably the least prevalent form of click fraud, it occurs more frequently than one would assume. This form of click fraud is more widely used on Social Media to increase ‘likes’ or ‘followers’. Click farms are very difficult to track as fraudulent clicks, as it is almost impossible to differentiate them from legitimate clicks.

Online Bots
Online bots are software applications that are driven by code in order to impersonate human behaviour. Over the years bots have become very advanced at mimicking human behaviour and are therefore quite difficult to detect. In essence, bots are coded to click on ads in order to increase the number of impressions bought without that advert ever reaching any actual eyeballs.

Bots have become sophisticated enough that they can run forever, if not detected, and at a very high rate of clicks. There are different types of bots, but the most intrusive to users are ones which invade PCss and infect them with malware which in turn conducts click fraud based on the user’s behaviour. Ultimately, these bots increase media spend on online ads, as brands think that a certain campaign is doing well based on a high click through rate, meanwhile in reality the campaign has not reached its target audience.

In turn, this enables fraudulent publishers to sell a large amount of inventory on ad exchanges which are never actually seen.

Impression Fraud

This is also known as ad stacking or ad stuffing and is a common form of fraud. This is when a fraudulent publisher stacks ads on top of one another by placing the ads in invisible iframes with zero to zero pixels and zero visibility. This, coupled with an online bot to load the pages so that the stacked ads get ‘viewed’, makes the advertiser think their ad is being seen even though only the top ad is actually shown to users.

Fake sites and domains

This is when websites are built with the sole purpose of advertising and offer no other content; these are also integrated within a larger network so that the site does not seem suspicious. Online bots are also used to load and reload pages, creating false traffic. Advertisers think that it’s a legit website, with tons of traffic and end up spending money on useless impressions.
How does one detect click fraud?

There are many technologies that can be used to detect click fraud, however a tell-tale sign that a campaign has fallen prey to a fraudulent site is where click through rates are close to 100% on a specific site. Often when one looks at a campaign report from a programmatic buy, there will be thousands of different sites on which the adverts have appeared and to save time, the sites where the bulk of the campaign has run are then assessed. But the key is to check those sites which hosted only a few impressions, if their click through rates are extremely high, then the chances are that these are fraudulent sites and should be added to a block list for future campaigns.

How can publishers and advertisers be protected?

Click fraud is widely known and discussed within the online advertising industry, especially within the programmatic buying environment. And whilst it is a real concern, it certainly hasn’t deterred advertisers and publishers from programmatic buying, and neither should it. There are a number of ways to ensure that you minimise the risk of falling victim to click fraud. In fact, ‘premium’ programmatic buying comes to the fore as part of the solution to click fraud as it saves advertisers significant budget building up block lists.

Advertisers need to find an exchange platform that guarantees a high degree of brand safety. In private exchanges, such as SouthernX, there are a number of validation and vetting processes that occur for both publishers and advertisers. Only validated, premium publishers are represented, ensuring that brands are buying legitimate inventory on legitimate websites. Advertisers and publishers alike also have the option to choose who they do business with, ensuring that all parties are satisfied that impressions, ads and traffic are legitimate.

Further safety measures include both automatic validation of creative as well as human validation, making sure that publishers do not get unwanted adverts on their platforms such as pornographic content or spam. In addition, each publisher and website is also verified and audited by both automatic systems and the SouthernX quality control team which ensures a supply of genuine inventory to advertisers. All of this ensures that both publishers and advertisers are given peace of mind whilst trading.

As new technology is introduced, fraudsters and hackers will always try and find a way to create disorder, but this shouldn’t cause us not to adopt the technology. Before entering into an ad exchange, both publishers and advertisers should educate themselves on what validation processes are in place to ensure that they are protected. In essence, publishers and advertisers shouldn’t be deterred from entering into programmatic buying by the fear of click fraud. Premium, private exchanges can provide protection for most of the types of click fraud discussed in this article. In addition to the various prevention methods, well informed programmatic buyers will also be able to identify click fraud by interpreting the campaign data, and stop it before any real money is lost on worthless impressions.