Are you done shivering for fear of click fraud yet? I didn't think so... Now, keep alert for here comes to be added another one, the “trendiest”, impression fraud.
Although not much discussed in the past, for it was far overpassed in the caused prejudices (especially cost-related ones) by the very spread phenomenon of click fraud, impression fraud has appeared a long time ago. Since the recent revival of impression-based advertising in Google's advertising system, the problem of impression fraud is brought again into the spotlight. Naturally, if until not long ago impressions did not cost anything, since they've begun being paid for, fraud threatens. However, not much emphasis is put on its importance.
Impression fraud means creating false page impressions with the purpose of wasting competition's advertising money, as a competitor, or making money fraudulently, as a publisher. The ads are caused to be artificially served many more times than in normal circumstances. Thus, prejudice is done to both CPC and CPM advertisers.
You probably are aware that a form of impression fraud, causing CPC ads to drop in position, has been unscrupulously used by some ever since the beginnings of the CPC system. The advantage for the offender is that it has been hardly detectable, everybody watching for fraudulent clicks. What was the mechanism? The more impressions for the same number of clicks, the lower the CTR, the lower the ad positioning, until it drops out of the search rankings.
The phenomenon affecting CPM advertisers has not acquired great proportions yet and it does not exert much pressure. Some have begun to wonder on this, on whether it has started to occur, on its effects and on the measures that can be taken to prevent it.
Effects
Advertisers are the first category to be affected. Invalid impressions make an advertising campaign useless, burning out an advertising budget and eliminating a campaign from competition.
Publishers have begun to wonder what effects will this have on them, considering the definite position on click fraud in AdSense Program Policies -- click fraud is absolutely not tolerated, publishers' accounts are simply disabled in such cases without much research on whether the website owner did it or not. That's why this new possibility of fraud (which is even easier to occur) frightens many, making them wonder what are the limits that can be considered safe and what is Google's position on this.
No reports of extraordinary negative effects of impression fraud have been made so far by AdSense publishers. Google's rules are not very strict in this regard, the primary concern is still for click fraud.
Cases were when publishers, noticing unusual performance parameters in their AdSense stats, or fearing their own repeated reloading of AdSense pages, have e-mailed Google. Responses were reportedly very sketchy on the impressions fraud problem, emphasising more on click fraud and on the interdiction of clicking one's own ads, as you can see in the fourth post on this page or in the sixth post on this page on WebMasterWorld.
Many publishers have created an issue around their fear of generating impressions that could be considered fraudulent while working on their own sites. An official statement from Google says that AdSense publishers are allowed to view the CPM ads on their own websites without the risk of creating invalid ad impressions, still excessive reloading of one own's pages is to be avoided. Great, isn't it? The only problem with this would be that they did not explained what "excessive" stands for. So, it's up to you, I guess...
Prevention
For advertisers:
- Choose only high-profile sites for impressions-based advertising;
- Watch out for sudden decreases in the performance of keywords.
For publishers:
- Keep a close watch at the performance of your pages. A sudden drop in CTR might indicate fraudulent impressions. Reporting to Google anything that looks suspicious would ensure safety.
- Avoid excessive reloading of your own pages. Yet, if you have to do this, reporting it cannot hurt.
Conclusion
Though impression fraud was not considered so much of a threat in the past, it might as well become. The phenomenon might just take a huge leap ahead in development, being more easily achievable than click fraud and harder to detect, because of the great amount of browsing taking place on the web. Once this will become being considered a real threat, real means of efficient detection will be developped and serious measures against it will be thought of. Until then, we're on our own!




