Monday, July 31, 2017

What is Click Fraud ?

If not redirected, please click here https://www.thesecuritybuddy.com/anti-spam/what-is-click-fraud-and-how-to-prevent-it/




What is Click Fraud ?


In online advertising, advertisers often use ad-networks to place a clickable advertisement on a publisher’s website. A user interested in the advertisement clicks on it to get more information. The advertiser then pays a predetermined amount of money to the publisher through the ad-network. This type of advertising is also called Pay Per Click advertising or PPC advertising, as the money as advertiser pays to the publisher is directly related to the number of clicks generated from the website of the publisher. Click fraud is a type of fraud where a fraudulent person himself or by using some automated scripts imitates a legitimate user and generates lots of clicks on an advertisement without having any actual interest in the advertisement. It is often done with the purpose of making the advertiser lose lots of money.


What is the purpose of Click Fraud ?


Perpetrators may have several purposes behind click fraud.

  • To cause the advertiser financial loss – Perpetrators often use click fraud to cause financial harm to the advertiser. It may be done by a competitor who advertises in the same market. The perpetrators may not directly profit from the click fraud, but they can force the advertiser to pay large amount of money for irrelevant clicks and thereby weaken or eliminate the source of competition.
  • To frame a publisher – The perpetrators may do click fraud to look as if the publisher is clicking on the advertisements of his own website for his own financial gains. The advertising network may then terminate the relation with the publisher. If the publisher relies heavily on advertising for revenues, the click fraud may cost him heavily.
  • To cause the publisher financial gains – Sometimes supporters of a publisher like his friends, family, fans or political party supporters perpetrate click fraud as they think that would cause financial gains to the publisher. But, it gets backfired instead, once it gets detected and causes harm to the publisher instead.
  • Other malicious intentions – Sometimes the perpetrators may have other motives other than causing harm to the publisher or the advertiser. The perpetrators may have political or personal vendettas. These causes are very difficult to track down and it becomes quite difficult to take proper legal actions against the perpetrators.


How to detect Click Fraud ?


Many ad-networks use several techniques to detect click fraud.

  • Automatic Filter – Sometimes ad-networks use automatic filters to filter out invalid clicks in real time.
  • Manual Analysis – Sometimes ad-networks use manual analysis to investigate anomalies and fluctuations on clicks and remove invalid clicks once detected.

Anytime an invalid click is detected by an ad-network, credits are issued to the corresponding account.

But, techniques used by the ad-networks may not be enough always. So, it is always good for the advertisers to monitor the clicks on the advertisements and do their own analysis to detect click fraud.

There can be several signs that may indicate a click fraud :

  • Number of times the advertisement shows on a search page increases unusually.
  • There are unusual number of clicks on the advertisements.
  • There is no significant increase in the number of conversions even when there is an unusual increase in the number of clicks on the advertisements.
  • The average number of pages visited per visitor drops.
  • The bounce rate becomes high.

An advertiser should always look for these warning signs and report it to the ad-network after doing own analysis.

Click fraud can be analyzed by the advertiser using various techniques. Whenever there is a click on an advertisement and the user arrives at the advertiser’s website, it is always good for the advertiser to collect a few data and do his own analysis. For example, information on the following can always help :

  • IP address of the user who clicks on the advertisement
  • Click timestamp and action timestamp
  • User agent

Information on IP address – If there is a huge number of clicks from a particular IP, then there is a possibility that it may be a click fraud. But, it is always good to collect some more information on the suspicious IP address before deciding anything. It may not be a click fraud if the IP address is of a proxy server serving many Internet users.

Click timestamp and Action timestamp – Click timestamp is the time when a user clicks on an advertisement and visits the advertiser’s website. Action timestamp is the time when the user performs the intended action on the website. If there is a huge number of clicks with click timestamp and no action timestamp, then it is likely to be a click fraud.

Information on User Agent – Information on User Agent is helpful in detecting whether clicks from a particular IP is from a single person. Information on the device being used, browser, software etc also help as they can identify whether two people using the same IP are the same person or not.

So, to summarize, advertisers should detect click fraud using the following steps :

  • Identify metrics to audit the clicks on the advertisement
  • Use a specialized software to track these metrics
  • Audit metric values to create a benchmark
  • Review current metric values against benchmark values
  • Take proper actions when benchmark thresholds are exceeded.


How to prevent Click Fraud ?


Advertisers should take the following actions to prevent click fraud.

  • Instead of relying on the ad-network alone, the advertisers should do their own analysis to detect click fraud as explained above.
  • If a IP address is associated with fraudulent clicks, the IP address can be blocked from displaying the advertisement.
  • Advertisers can use remarketing to display advertisements to those who have visited and shown genuine interest in the advertiser’s website.
  • If a competitor is suspected to commit click fraud, one can always block his IP, zip code or city from displaying the advertisement.
  • If most of the clicks from a particular city or region are click fraud, one can always exclude the region from showing the advertisements. But, one should also make sure not too many good traffic are eliminated.

Click fraud is a serious problem. But, with proper approach this problem can be dealt with. So, keep updated about click fraud and stay safe, stay secured.