The main advantage of advertising online vs. traditional channel is that it is more measurable, thus can be made more effective. However, what undermines this efficiency is the proliferation of fraud in the form of “non-human” traffic.
Being a worldwide industry problem, China’s digital advertising market seems to be one of the worst affected ones. Chinese internet is the second most popular advertising media after television and was estimated at $20 billion by the end of 2013. Unfortunately, advertisers’ losses could account for up to third of that amount, mostly lost to non-human traffic and fraudulent CTR (click through rate).
The report by Miaozhen Systems, released about a year ago, claims that auto and telecommunications were hit the worst with up to a third of the traffic being non-human, i.e. generated by bots and crawlers, while food & drink and the fast moving consumer products industry saw the lowest rate at 8.52% and 11.88% respectively.
…many local computers still run on Windows XP that is no longer supported by Microsoft. Bots can infect such machines much easier and, in fact, a proportionally larger share of fraudulent traffic seems to be coming from lower tier cities where more people still use older Windows OS.
How do the losses occur exactly? A small piece of malicious software, a “bot”, is placed on a host computer from which it visits various sites in the background without user noticing anything abnormal with their machine. It’s even harder to spot for advertisers because it shares the real user’s unique “cookie” identifier.
Other pieces of software can stack hundreds of ads on top of each other on a website, or place the entire website into a small pixel on a page. This way, it would seem that a particular ad was displayed while, in fact, it was never seen by a visitor.
CPC (cost per click) ads are also affected as such bots can be smart enough to mix real clicks with the “fake” ones, making it harder to identify as a non-human activity.
There are number of reasons why China is one of the worst affected markets. First, many local computers still run on Windows XP that is no longer supported by Microsoft. Bots can infect such machines much easier and, in fact, a proportionally larger share of fraudulent traffic seems to be coming from lower tier cities where more people still use older Windows OS.
Second, not all the fake traffic is caused by bots, some is accounted for search engine crawlers. Since search engine market in China is much more fragmented than in the West, there are simply more crawlers visiting each site.
A third reason that is also specific to China is the fact that many people would willingly install special traffic software on their systems in order to attract more followers to their social media profiles, making their social activity appear more popular. However, such programs would do many other things without user’s knowledge or consent, effectively plugging such computer into a larger bot network.
Lastly, there is a lack of habit and willingness to go after the offenders and many in the industry still consider this problem to be the necessary evil that nothing much can be done about. Although, there is a technology in place to combat the problem, its adoption in China has been slow.
A new alarming trend is the increasing proliferation of online fraud in the growing mobile space. Android is by far the most popular mobile OS in China but the official Google Play store is not the usual place to go for apps. Most people would go to many smaller apps market which exercise very little control over quality.
While, at the moment, most non-human traffic in China is still PC based, it is reasonable to assume that with more people using Android devices to access internet, the worst of mobile advertising fraud is yet to come.