Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Cloud

5/30/2017
05:10 PM
Connect Directly
Google+
Twitter
RSS
E-Mail
100%
0%

Bot-Driven Online Ad Fraud Losses Decline

But counterfeit ad inventory the next big worry for online advertisers.

For once, a bit of relatively good news in cybercrime: online ad fraud losses to corporate advertisers fell 10 percent this year even with an increase in digital ad spending.

That's the finding of a new study conducted by the Association of National Advertisers (ANA) and security firm White Ops, which tracked online ad traffic patterns of 49 ANA members, many of whom represent some of the nation's largest companies, as well as other digital ad activity from October 2016 through January 2017. ANA members who participated in the study include Coca Cola, Unilever, and Walmart, for example.

Online ad fraud losses for 2017 will drop to about $6.5 billion worldwide from $7.2 billion in lost revenue for corporate advertisers in 2016, the study found. Digital ad spending is forecast to increase by 10% this year, so the decline in fraud represents a positive trend in combating fraudulent bot activity, according to ANA and White Ops. ANA members overall will lose even less in revenue - some $3.3 billion worldwide, the study shows.

"This is the first year [of this study] that we have seen genuine decline" in losses to ad fraud, says Michael Tiffany, CEO of White Ops. "The cost of bot traffic that defeats detection has gone up; it used to be a fraction of a penny per visitor. Now the market price for a realistic-looking bot has risen to about a penny per visitor … It costs more to perpetuate [this crime]."

Bill Duggan, group enterprise vice president at ANA, says companies are getting savvier about online ad fraud. "Three years ago when we started this journey" to measure and prevent online ad fraud, he says, the problem was just a blip on ANA member radar screens. "Marketers are taking a bigger interest in fighting fraud" today, he says, and the data bears that out.

Criminals use bots against advertisers by giving the illusion of real ad views, and the botnet operators who run those phony website visitors make money via cash-out points.

Bots continue to look more and more like humans in order to evade detection as phony traffic. Some 75% of the fraud spotted in the study was the handiwork of a computer plus a human on the same endpoint. Meanwhile, mobile ad fraud represents a tiny portion of the fraud landscape, with under 2%, the report found.

Craig Spiezle, founder of the Online Trust Alliance, says programmatic advertising marketing may be efficient, but it comes with security and trust risks. "There's a difference between efficiency and effectiveness: yes, you sell ads quickly, and inventory is great, but effectiveness [is low] because you don't know if your ad is being viewed or seen," he says.

White Ops' Tiffany says some programmatic ad-buying operations are improving now, given the fraud problems. "Several programmatic platforms have gone above and beyond instituting new security controls to tackle the problem," he says.

The next big thing: counterfeit inventory, he says. "We are seeing an increasing amount of counterfeit inventory. That was the kind of attack that Methbot did," Tiffany says. "Not just fake visitors but entirely fake inventory that purports to be real ads running on real websites, but it exists in the ether."

Methbot was a massive online advertising fraud operation run by Russian cybercriminals that was stealing up to $5 million per day from big-name US advertisers by posing as some 6,000 major US media sites, including The Huffington Post, Fortune, ESPN, CBS Sports, and Fox News, and generating fake ad impressions. Discovered in late 2016, Methbot had been operating in plain sight for three years, using its own internal botnet infrastructure rather than than the traditional ad fraud model of infecting unsuspecting consumers to do the dirty work.

Related Content:

Kelly Jackson Higgins is Executive Editor at DarkReading.com. She is an award-winning veteran technology and business journalist with more than two decades of experience in reporting and editing for various publications, including Network Computing, Secure Enterprise ... View Full Bio

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
REISEN1955
50%
50%
REISEN1955,
User Rank: Ninja
9/7/2017 | 1:15:45 PM
IBM and Watson
If I read one more post anywhere about how Watson can save every single problem in the western world through AI knoelege ..... while Lady Ginny at IBM is gutting cloud and Watson staff like crazy???    Can Watcon improve IBM revenue gone bad for 21 quarters???  AI may be great but it is NOT a total solution until itself.  Anybody remember what WOPR was and damn near did?
Eduardo.Master
50%
50%
Eduardo.Master,
User Rank: Apprentice
9/7/2017 | 12:05:29 PM
Re: Ad fraud is a big problem
Can Machine Learning Help Organizations Improve Data Security?
Kelly Jackson Higgins
50%
50%
Kelly Jackson Higgins,
User Rank: Strategist
5/31/2017 | 9:39:26 AM
Re: Ad fraud is a big problem
Given that losses are still in the billions of dollars, indeed it is still a huge problem. Cleaning up the pollution in the ad ecosystem entails the cooperation of all players, and continuing to make it too expensive for the bad guys to target this space.
floydophone
50%
50%
floydophone,
User Rank: Author
5/30/2017 | 11:28:40 PM
Ad fraud is a big problem
Even though it seems to be declining, ad fraud is still a big problem. There's a lot of great coverage on the Software Engineeering Daily podcast about ad fraud and various mitigation techniques. At the end of the day, I think CPA mitigates a lot of the problem but it'll be a long time until we're fully there.
10 Ways to Keep a Rogue RasPi From Wrecking Your Network
Curtis Franklin Jr., Senior Editor at Dark Reading,  7/10/2019
The Security of Cloud Applications
Hillel Solow, CTO and Co-founder, Protego,  7/11/2019
Where Businesses Waste Endpoint Security Budgets
Kelly Sheridan, Staff Editor, Dark Reading,  7/15/2019
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Write a Caption, Win a Starbucks Card! Click Here
Latest Comment: "Jim, stop pretending you're drowning in tickets."
Current Issue
Building and Managing an IT Security Operations Program
As cyber threats grow, many organizations are building security operations centers (SOCs) to improve their defenses. In this Tech Digest you will learn tips on how to get the most out of a SOC in your organization - and what to do if you can't afford to build one.
Flash Poll
The State of IT Operations and Cybersecurity Operations
The State of IT Operations and Cybersecurity Operations
Your enterprise's cyber risk may depend upon the relationship between the IT team and the security team. Heres some insight on what's working and what isn't in the data center.
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2019-13360
PUBLISHED: 2019-07-16
In CentOS-WebPanel.com (aka CWP) CentOS Web Panel 0.9.8.836, remote attackers can bypass authentication in the login process by leveraging knowledge of a valid username.
CVE-2019-13383
PUBLISHED: 2019-07-16
In CentOS-WebPanel.com (aka CWP) CentOS Web Panel 0.9.8.846, the Login process allows attackers to check whether a username is valid by reading the HTTP response.
CVE-2019-13603
PUBLISHED: 2019-07-16
An issue was discovered in the HID Global DigitalPersona (formerly Crossmatch) U.are.U 4500 Fingerprint Reader Windows Biometric Framework driver 5.0.0.5. It has a statically coded initialization vector to encrypt a user's fingerprint image, resulting in weak encryption of that. This, in combination...
CVE-2019-13605
PUBLISHED: 2019-07-16
In CentOS-WebPanel.com (aka CWP) CentOS Web Panel 0.9.8.838 to 0.9.8.846, remote attackers can bypass authentication in the login process by leveraging the knowledge of a valid username. The attacker must defeat an encoding that is not equivalent to base64, and thus this is different from CVE-2019-1...
CVE-2019-13615
PUBLISHED: 2019-07-16
VideoLAN VLC media player 3.0.7.1 has a heap-based buffer over-read in mkv::demux_sys_t::FreeUnused() in modules/demux/mkv/demux.cpp when called from mkv::Open in modules/demux/mkv/mkv.cpp.