Taking Aim At Cyber Scammers To Ensure Traffic Quality
by News
on 18th Mar 2014 inCameran Harman, OpenX, managing director, EMEA, explains that while the programmatic industry as a whole is generally in rude health, it is one besieged by nefarious parties, and how combatting fraud is everyone in the industry’s job.
An Industry Under Siege
Digital advertising is all about scale and confidence. The greater number consumers advertisers can reach, the more opportunities they have to grow their business. Publishers want scale too, because the more readers they can claim, the higher the CPMs they can charge.
And yet scale has exposed the industry to an alarming amount of fraud. According to Steve Sullivan, IAB vice president of advertising technology, fraud is a $6 billion crime perpetrated against advertisers each year.
Advertisers aren’t the only ones who suffer. When fraudsters place low-quality inventory on ad exchanges, they damage the monetisation prospects of every publisher on that platform. Premium publishers are particularly at risk of having the value of their inventory diluted by such practices.
Who are these scammers, and what can the industry do to prevent fraud?
Types of Fraud
According to internet security authorities, bad actors create botnets to generate massive amounts of traffic to phoney sites. These sites look attractive to unsuspecting advertisers who pay to advertise on them – ads that will never be seen by consumers.
Fraudsters also generate fake page views, which create phantom impressions that are then sold on automated platforms. To conduct their nefarious practices, cyber scammers have a host of tools at their disposal, including botnets, browser hijacks, impression-laundering (i.e. misrepresenting risky inventory as legitimate), and hidden ads.
A large portion of fake traffic also from legitimate publishers who purchase traffic. “To publishers, buying traffic for pennies a click or less from a network sounds like a dream come true,” according to John Murphy, VP of marketplace quality at OpenX.
“What they don’t know is that those networks are often buying traffic further downstream from botnets or other nefarious sources,” he explains.
Combatting fraud is everybody’s job
In December 2013, the IAB released a paper Traffic Fraud: Best Practices for Reducing Risk to Exposure, stating that “both the buy-side and sell-side need to play a role in defending against traffic fraud and improving the digital ecosystem.”
The document provides concrete steps that buyers, publishers and ad networks should do in order to combat fraud. Does that mean that the ad-tech providers are off the hook? Absolutely not. “OpenX decided that we need to take a lead role in combatting bots and fake inventory in the OpenX Ad Exchange,” says Mr. Murphy.
OpenX’s multi-pronged approach traffic quality
When it comes to ensuring traffic quality, there is no magic bullet. That’s why we have a multi-pronged approach that encompasses rigid publisher and buyer approval processes, proprietary technology, third-party partnerships, and a dedicated team of data scientists to monitor traffic quality 24/7.
Publisher approval process
All ad exchanges need strict, multi-tiered policies that govern all supply it sells. Policies should encompass content type, number of ads per page, and portion of ad that must appear above the fold.
OpenX has introduced several policies designed to protect its buyers, including a ban on so-called ‘ghost’ sites (sites with little or no original content) as well as sites that support piracy, host un-moderated user-generated content, or contain excessive ads, with no signs of user engagement. As a result, we reject more than 30% of the sites submitted to OpenX Ad Exchange.
Proprietary Technology
As noted earlier, bot traffic stems from legitimate sources as well as fraudulent ones. For this reason, the ad exchanges must take a leading role in filtering all traffic. To this end, OpenX has built an internal traffic filtration system that scans an impression, IP and domain in real time to determine if it’s bot traffic. If it is, the ad request never enters the exchange. It reaches OpenX, but it’s blocked. On average, OpenX rejects 12 – 15% of all impressions we see.
Third-Party Partnerships
True traffic quality goes beyond scanning for fraud; advertisers need to know their ads will have a fair chance of being seen, as well as leading to a conversion. Our industry is fortunate enough to have plenty of tech companies that can lend their expertise to the ad exchanges’ efforts to ensure quality.
For instance, OpenX works with Integral Ad Science (IAS) to flag pages that advertisers consider inappropriate for their brand. Sites and pages that score poorly are then reviewed by IAS team members.
And we also partner with Adometry, the industry’s leader in click-scoring. Adometry scores pages based on the likelihood that a conversion will occur (and since bot traffic never leads to a conversion, Adometry serves as another checkpoint for OpenX). Thanks to the combined proprietary technology and processes, OpenX ranks 30% higher than Adometry’s industry benchmark.
Dedicated Team
Ensuring traffic quality isn’t something you can add on to an existing group’s plate; it requires dedicated resources who focus on nothing else. At OpenX, we have a team of 10 full-time people who dedicated to quality. Most of them are PhD holders in data science, computer science or statistics. The team is constantly developing new algorithms to detect bot and other fraudulent traffic.
Ad VerificationAdvertiserattributionDataM&AMartechMedia SpendProgrammaticTargetingViewability
Follow ExchangeWire