Using Ads for Impression Fraud
March 25th, 2009
Watch this video:
How does it work? Pretty easy –> just traffic an ad with some JS that keeps writing out IFRAMES with ad tags. User never sees an ad and as long as the page is opening you are generating what appear to be valid ad requests from valid users!
Hopefully Right Media’s fraud filters are catching this!
Related Posts:
- About
- Business or Tech
- Redirects and Integration, Part II: Hacking Around the Browser
- The Coming of the Real-Time Exchanges
- Premium vs. Remnant — (Part I — Supply)
March 25th, 2009 at 12:22 pm
If the adserver serving up the creative is worth a damn this traffic will probably get flagged as either spider or bot activity and the metrics adjusted accordingly. X number of sequential requests from the same IP within a predetermined time span will generally trigger this, at least that was how we adjusted for this kind of stuff when I worked at my previous gig AdShuffle.
March 25th, 2009 at 1:06 pm
Sure, if it loads the samet ag over and over again. But it’s not hard to get a little smarter here — but what if my javascript randomly loaded from one of 200 ad tags? Now you generate 1-3 views per tag on a single 300×250. Would your adserver be able to identify these 3 requests as fraud?
I have to say — this almost makes me miss the days of ‘inqwire.com’ when at least the fraud was blatant and obviously visible in a pop-under.
March 25th, 2009 at 1:36 pm
I would think the variance in ad tag would be irrelevant. The telling factor isn’t that the ad request is always the same (although that could be one additional factor to take into consideration), it’s the high number of sequential requests from the same IP address in a very short amount of time. 60, 100, or how many ever ad calls in less than 2 minutes is pretty suspect.
March 25th, 2009 at 3:26 pm
Hmmm. There’s a whole lot of ad networks out there, over a couple of minutes I bet someone could load about three tags from *each*. Hell, one tag from a dozen is relatively easy, and that’s still a ton of extra money.
Awesome! I love the creativity of scumbags. I suppose this could be caught by measuring clickthrough rates and detecting abnormalities, but those clicks could be faked too, I bet. (It’d be ironic if every company had to eventually measure CPA just to make sure their CPM ads weren’t fraudulent…)
By the way, why is your test ad spot running a ton of dating ads?
March 26th, 2009 at 3:48 am
Awesome, the ads just don’t stop loading
March 26th, 2009 at 4:50 am
Mike,
I came accross this tow days ago on a pretty big filehosting website with two ad networks showing ads
while my charlesproxy showed 200+ rows of urls in less than 2 minutes. Nice scam.
Michael
March 28th, 2009 at 10:41 am
[...] posted the video (which is below) and commenters, as well as Mike(OnAds) Nolet, noted that this could potentially prove difficult to catch with a bit of misdirection: “… it’s [...]
March 30th, 2009 at 1:11 am
Hi Mike,
I’ve certainly seen this carry-on before, but never such an extreme example!
Sandi &c.
July 7th, 2009 at 1:26 pm
Holy crap. This make me look at bad traffic from media buys in a whole new light.