Click Fraud: A Guide To Protecting Your Pay-Per-Click Campaign
By John Hill
Click fraud is the latest 'hot topic' circulating the online
marketing arena, but what is it? And how does it affect you as a
merchant running a pay-per-click campaign?
Spending on Internet advertising is growing faster than any
other sector of the advertising industry and is expected to
surge from $12.5 billion last year to $29 billion by 2010 in the
U.S. alone, according to the researcher eMarketer Inc. With
around 50% of this spending being spent on pay-per-click (PPC)
advertising.
Here we offer you a complete guide as to what this phenomenon
is, who is likely to commit such an act, how to identify &
prevent click fraud and how to best report instances of
suspected click fraud on your PPC campaign.
What is Click Fraud?
According to Wikipedia "Click fraud is a type of internet crime
that occurs in pay per click online advertising when a person,
automated script, or computer program imitates a legitimate user
of a web browser clicking on an ad, for the purpose of
generating a charge per click without having actual interest in
the target of the ad's link"
Click Fraud is estimated to range from 5% - 15% of
pay-per-click traffic (some estimates are as high as 20% - 35%)
although Google estimates click fraud at only 2% due to the
rigorous detection methods they claim are in place.
In a recent study by Click Forensics, click fraud reached a new
high of 14.2% in the last quarter of 2006 with the average rate
of click fraud on 'content networks' as high as 19.2% for the
same quarter.
So who is likely to commit Click Fraud?
The click fraud villain is most likely to fall into one of
three categories:
- Online vandals with nothing better to do than cause a
nuisance
- A competitor clicking on your search network PPC ads, with
the sole intention of increasing your cost-per-acquisition
(CPA). This could be interpreted as click fraud, although
currently the search engines do not consider this kind of
activity as click fraud
- Search Engine advertising affiliates who generate self-income
from fraudulent clicks on 'content network' adverts displayed on
their own websites. This practice, at it's extreme, involves the
use of unscrupulous 'paid to read' or PTR sites, which are
basically click-fraud rings, some with hundreds or thousands of
participants, paid to click on your ads with no regard for your
return on investment (ROI) as the advertiser
What are the Search Engines doing about it?
Both Google and Yahoo claim that they filter out most
fraudulent clicks. The costs involved for these clicks are
either not charged or are reimbursed to advertisers who have
been wrongly billed.
To combat click fraud Google applies four layers of fraud
detection:
1. Automatic detection - this filters clicks from both the
search and content networks in real-time with the goal of
removing them before their existence is ever shown to the
advertiser
2. The "Flagging system" - an automatic process to remove
invalid AdSense clicks
3. The "Manual review" - this process has more than two-dozen
Google employees tasked with manually reviewing and removing any
suspicious AdSense clicks
4. If the first three layers of protection fail then the fourth
and final layer of click fraud detection falls to the advertiser
and 3rd party click fraud detection companies. Google refers to
this layer as "requested investigations"
Googles main aim is that the first three layers of filtering
will identify all invalid and fraudulent clicks. These layers
currently filter more than 98% of invalid clicks.
And should you be in any doubt, both Google and Yahoo have, in
the past, released the following statements:
"We think click fraud is a serious but manageable issue" says
John Slade, Yahoo's senior director for global product
management.
"Google strives to detect every invalid click that passes
through its system" says Shuman Ghosemajumder, the Google
manager for trust and safety. "It's absolutely in our best
interest for advertisers to have confidence in this industry."
As a positive for the future, Google is currently testing a
cost-per-action (CPA) platform, which should effectively deal
with click fraud. With CPA ads you don't pay by the click but
instead pay when the customer reaches a certain goal: buys a
product, fills an enquiry, etc.
How to identify click fraud on your pay-per-click campaign
Before you can even contemplate identifying click fraud you
must have effective tracking tools implemented on your website
and, if possible, access to your server logs. With tracking
tools in place, the most obvious way of spotting click fraud is
to simply observe any spikes in traffic where there is no
particular shift in your conversions.
Once identified, these spikes can then be analysed by looking
for repeated clicks from sources that look similar. This
similarity could be an IP address or an IP range; it could be a
combination of IP range; browser version; operating system.
Basically look for data in groups that looks fraudulent.
If all this is just 'a bit too heavy' for you then there are a
number of companies out there that can help.
- AdWatcher: claims to able to spot click fraud so that you can
report it. Covers other aspects of PPC marketing, by helping you
track your ROI, email success, etc.
- Click Auditor: offers the ability to check whether your
competitors IP is the one performing any abusive clicking, and
says it will stealthily gather your competitors IP addresses for
this purpose
- ClickSentinel: focuses on helping you get refunds on
fraudulent clicks, as requesting a refund from your PPC provider
can often be very difficult for the un-initiated
- Click Tracks: reportedly has automatic click fraud reporting
along with other click tracking (analytics) tools
Reporting suspected Click Fraud
When reporting suspected click fraud, you must include as much
captured data as possible to increase your likelihood of
obtaining a refund or credit.
The following guidelines are recommended:
- Clearly state, at the very beginning of your claim, that you
are reporting suspected click fraud
- Provide a full explanation to support your claim
- Include your account details (do not include your password or
payment information)
- State the exact keyword, ad and campaign where you suspect
click fraud has occurred
- State the exact time, date and IP address of each instance of
suspected click fraud. This data can be gleaned from your server
logs or 3rd party tracking tool
- Finally, state whether you are requesting a refund, credit or
investigation
If you were using any software tools, such as those highlighted
earlier, to help you track and report click fraud then include
any reports generated by these in your claim.
Lowering the risk of Click Fraud happening to you!
Always bear in mind that your PPC objective is to get
conversions and not just clicks.
The more you have researched the demographics of your intended
client base the better your chance of avoiding click fraud. Are
your clients from a specific country or location? When are they
likely to search for your product or service? What are the key
search-terms they are using?
With demographic data in-hand you can target your ad campaigns
more effectively and lower your risk of click fraud.
About the Author: E-Gain New Media (http://www.e-gain.co.uk)
are a specialist UK SEO company specialising in UK Search Engine
Optimisation, Online PR and Pay-per-click Management
Source: http://www.isnare.com
Permanent Link: http://www.isnare.com/?aid=126757&ca=Internet