Stopping The Vicious Cycle Of Ad Fraud
Michel de RijkMarch 15, 2016
USD$18.5bn – that is how much advertisers are estimated to have lost to ad fraud globally in 2015. This was revealed in a recent study conducted by Ernst & Young that was commissioned by the Interactive Advertising Bureau.
It’s a big number and clients should be worried because it is highly possible that they have contributed part of their marketing spend to it without them knowing. And truth be told, it does affect the quality of campaign results because the numbers you see may not be a true reflection of how people are perceiving or welcoming your ad.
At Xaxis we see prevalent ad fraud happening in the industry, thus we make sure that we protect our clients’ campaigns by working with the right media partners and taking the necessary precautions as I would explain at a later part of this article.
But as an advertiser, there are steps you can and have to take to protect your campaigns and budgets. But most importantly, what can you also do to stop the fraudulent cycle?
Identifying the different types of ad fraud?
There are many types of ad fraud, and the first is a human problem – dishonest media vendors who pay huge numbers of people to click or sign up for an offer thus creating a click farm. These are persistent fraudsters who despite 3rd party ad-serving or click trackers, are still stuffing ads with multiple click trackers to generate many clicks off just one real one.
There are also many media vendors who may not be running the fraud themselves but by outsourcing the delivery to other networks, they are opening themselves and their clients up to the risk of fraud. These are the kind of vendors that you need to avoid.
Secondly, it is the use of poor quality media which are often ghost sites. These sites are thin on content and are replicated many times over under different domain names and are almost never updated, and yet contain multiple ad slots that are often stuffed with many layers of ads. These are uploaded into exchanges and ads are placed by unwitting buyers who are making ‘run of exchange’ buys. Unfortunately, new ones are created as quickly as they are blacklisted.
Thirdly (often in conjunction with ghost sites) we see evidence of bots, and whole networks of these (also known as botnets) are distributed through malware or ghost sites. These are indiscriminate and can fake millions of ad impressions and clicks on all types of websites (mostly on the ghost sites).
But in my opinion, what’s worst than fraud is turning a blind eye to it.
Ignorance is not bliss.
I don’t think a lot of marketers have a clear idea on how bad ad fraud is AND how it affects them, and that is the problem. The first thing the industry needs to do is to acknowledge that ad fraud is a problem and understand its scale. Unfortunately, this stage has not been reached yet because most advertisers are not aware and most media vendors do not want to deal with the issue. However, these issues affect the whole online ad ecosystem.
Advertisers are also not asking tough enough questions. Many are ignoring or downplaying the problem because they want to be able to deliver good perceived click results to their management. The desire for cheap prices and high CTR’s (the dated measurement metric of choice) compounds the severity of the problem. Change needs to come from the top, management teams need to do move on from the archaic way of evaluating the success of a digital campaign.
However I have seen cases where even when clients are advised otherwise, they still insist on pricing, buying metrics and KPI’s that actively encourage media vendors towards fraudulent activities. It’s simple supply and demand. If clients want the cheapest CPC or reward simply high CTR then media vendors will search for placements that offer these metrics. Unfortunately, these placements are often fraudulent.
Why you should be worried
It is only really in 2014 that the above fraud issues have become identifiable using 3rd party technologies. However, these technologies are not widely used in APAC and still have big gaps in what they can track e.g. VAST video formats. This including large media vendors building walled gardens to actively disallow 3rd parties to independently track their platforms to establish the extent of the fraud and work out ways to avoid it. Obviously opening up to this tracking could impact their revenues.
The alternative is that media vendors allow for check and balance but incur higher costs and lower CTR. However they risk not getting the re-book and are therefore turning a blind eye simply to please advertisers – even though it is wrong.
I advise clients to demand (and pay for) 3rd party ad-serving, and heed the advice to set KPIs that measure cost per in target audience reach for brand campaigns and deep beyond the click metrics for response campaigns. Until this happens at scale, fraud is just too easy and clients will unfortunately continue to pay for it.
My concern is that simple economics of the ecosystem is proliferating the problem. For example if you are given the choice of buying $0.25 CPC media and risking 50% of clicks being fake or buying $1 media and knowing that all clicks are genuine, then what is the better buy from a eCPC point of view? On face value, it might seem more cost effective to take the hit of the fake clicks and buy the cheaper option. However, this is supporting crime and making online advertising untrustworthy. Advertisers have to make a stand and pay for quality, and not simply chase the lowest price. But this has to be a collective effort from everyone.
Precautions: How to fraud-proof your ad campaigns
The markets where fraud seems to be most extensive in are the markets where CTR, clicks and cheap pricing are promoted above everything else. Examples of such markets are Vietnam, Taiwan, China and India.
It is a fact that only a handful of clients pay for independent fraud and viewability tracking software. This illustrates the fact that although it is becoming more of an issue, very few clients are still unwilling to pay for technology to identify the media vendors who are delivering fraudulent views or clients.
So how can you make sure that your campaigns are immune from ad fraud? What can you request from your agency to make sure that they carry out their due diligence?
Monitor all campaigns using 3rd party technology so as to identify, block and avoid fraudulent placements. Make sure your agency has a blacklist of fraudulent sites and that they only work with ad exchanges which they can have direct human contact with a trusted employee of that exchange.
Identify suspicious numbers. We have seen campaigns where over one third of impressions are deemed suspicious by our own standards. For example impressions delivered on domains that result in over 1% CTR. Where this level of performance is occurring (on desktop campaigns) it is likely that some kind of click fraud is occurring. Always actively optimise away from domains that deliver over 1% CTR.
Xaxis’ own internal monitoring process using independent 3rd party technology indicates that if unoptimised, a programmatic campaign using RTB media alone can see up to 36% of impressions deemed as suspicious. However, if you use programmatic direct media, blacklist and optimise away from problem domains this can be reduced to almost nothing.
No doubt fraudsters will believe that the growth in purchase of aggregated media through open exchanges in APAC will offer them opportunity to defraud buyers. However, what we see outstripping RTB growth is growth in programmatic direct. Work with agencies that deal with ‘programmatic direct deals’ where they have integrated directly with the publisher and do not rely on buying via exchanges (which are susceptible to fraud). This means there is much more control of the quality of the inventory that is bought and hence can avoid fraudulent instances.
Ensure that you (or your agency) work with publishers who are represented physically and locally with a strict contractual agreements that takes into consideration 3rd party verification, and in cases where you need to use RTB.
For example when it comes to Retargeting, your agency should verify each seller and domain before they add that to their marketplace. This in conjunction with a deeper understanding of the user e.g: Have you seen this user before or is this the first time we are seeing this user? Which IP is this user coming from? Are there are spikes in traffic from this IP etc helps us determine if the impression is fraudulent or not before we buy the impression?
Protecting your campaigns and budgets.
The main issue across APAC and indeed globally is the ongoing obsession with chasing cheap pricing and click based KPI’s. If this is demanded above everything else then ad and click fraud will continue to grow. In my view, it is not about finding the culprits or developing ever more advanced technology to outwit fraudsters but really by cutting off the lifeline to them.
By the above I mean changing what we buy and how we measure online media so as to make it difficult for online criminals to earn money from fake online ads and clicks. This is not a game of hide and seek or a clean up job. It is just a shift in strategy as to what advertisers buy, deem important for their brands and measure.
Reassurance should come from the use of 3rd party ad serving and tracking technologies. This is my primary recommendation. While some companies may have their own stringent internal quality controls and technology, we advised clients to have their own method of assessing delivery, performance, fraud, viewability and also target audience achievement. Any media vendors who do not enable 3d party technology and wish to mark their own homework are potentially hiding something, and that’s when we advise you to bring your business elsewhere.
Michel De Rijk is the Asia Pacific Chief Executive Officer of Xaxis.