Commentary
Agencies And Networks Should Lead The Click Fraud Fight -- Stop Saying It's Up To All Of Us
- by Sean Hargrave , Staff Writer, March 24, 2015
That was the agreement at Ad Week Europe yesterday, and it's a conclusion that has been voiced time and time again. Now, I'm not going to say that nobody has any responsibility in combating any type of crime when we will clearly should all do the equivalent of bolting doors and putting lawn mowers and bicycles in locked sheds. If that's the spirit of Ad Week Europe's conclusion, then it would be hard to argue against it.
It's just that I have a lingering doubt that more hasn't been done sooner against click fraud because it wasn't in the interests of the huge trading desks to start lifting stones for fear of what might crawl out. I was chatting to a pal the other day who has left a major agency group that has a very well-known trading desk, and he confirmed that when ISBA claims as much as half of what a brand pays goes in bills and click fraud is just about on the money. So just to get it straight, it is not unusual for clients to have a bill that isn't broken down transparently, but which sees half of the total amount spend on fees and robots rather than media seen by humans.
Pretty scary stuff -- and that's why I just can't believe that brands haven't been more vocal on this. Or at least, I'm surprised they are not more publicly vocal. From what I hear, the issues over click fraud and fees can be explosive, but they are nearly always private.
It's probably a good time to remind ourselves that although we have an IAB definition of viewability, which can now be measured so vendors can be certified, the fraud equivalent is unlikely to surface for another year. This hardly sounds like an industry rushing to deal with a problem, does it? This is not me taking issue with the IAB because they're an industry body that can only go at the pace the industry wants to move at. It's more a case of pointing out the obvious -- why would agencies and their trading desks break their necks to combat a problem they could put down to fraudsters, rather than anything they have done wrong themselves?
The trouble is that although the cat is out of the bag and a lot of brands are building their data management platforms -- which, among many uses, will almost certainly be utilised for buying display programmatically without being completely reliant on a major agency's trading desk.
This is why I find it all the more surprising that fraud is apparently everyone's responsibility. Okay -- so we all have a part to play, but we all know what happens if we don't put up a rota for cleaning the office kitchen but instead leave it as everyone's job.
Agencies and their trading desks will have to step up to the plate and stop saying that everyone has a role, unless they are prepared to say, in the same breath, that they are taking the lead. Ultimately it's the ad exchanges they link into through their trading desks that have to be clean. If the agencies rounded on them, it is difficult to see how they could resist the demands to install the best anti fraud measures possible. You can throw technology around throughout the buying process but surely it makes sense to insist that any add exchanges has the best measures possible. Checking media where it is bought and sold has to be the best way of ensuring that it comes from a quality site. The quickest way to make this happen is for the agencies to only work with exchanges that have taken all the required anti-fraud measures required for advertisers to be as assured as possible they are paying for good quality media.
There may well be an argument that it's everyone's role to combat fraud -- but until someone has the courage to say "we're in it together, but we're taking the lead on this" then brands will surely continue to pay double the price of their media to run a campaign.
If that doesn't constitute a broken model, then it's hard to imagine what would.
Sean Hargrave is MediaPost's London Editor. You can reach him at seanhargrave@btinternet.com.
Good article Sean and I certainly agree that simply saying 'more work needs to be done' (i.e. but not by not the person saying it) gets us nowhere and is an annoying habit.
Click fraud, viewability and brand safety are certainly hot topics this year. Putting my official hat on: BuzzCity, as a global blind ad network with 13k directly integrated publishers- i.e. not exchange traffic- takes click fraud extremely seriously, and has done for several years- as audience quality is fundamental to the growth we've managed.
In short, we have various manual audit processes in place to keep a close eye on our publisher base as well as automated sweeps by the system, and if any 'suspicious' traffic is found the credits are refunded to the advertiser by default, and the publisher is put through a process that will generally end up seeing them blacklisted. Plus we have extra steps and scrutiny for our white listed, 'brand-safe' traffic.
So the question is, taking my official hat off again: what are "all the required anti-fraud measures" and how do you define "good quality media"?
I think as this topic gains traction in client/agency discussions, the fact that the above questions do not have a simple, single answer will become relevant- misconceptions and a lack of client education can come in.
Yes context and individual needs are important, but I sense a possibility for reactions to the problem to be confused or mistargeted- e.g. does 'transparency' automatically mean 'safe', 'quality' traffic? Is 'premium' spot-buy traffic inherently better than exchange or ad network traffic?
Of course not. Naturally, environment and context is a factor but when you're looking at such a critical industry-wide problem, it's fundamentally about reaching >humans< and not robots, in a cost-effective manner, and let the back-end results determine everything- assuming you're properly measuring what you're trying to achieve.
Robots won't pay for a service, or engage on a landing page in a way that is indistinguishable from a human, so all advertisers need to do is follow what works.
What they should NOT do is prejudge what will work without >testing<, unnecessarily limit their options or introduce half-baked rules on the whim of a client- i.e. do you really want to trawl through a huge site list to check its 'quality' (and how do you decide the criteria?), or do you go to who is doing the most to get you a quality audience and pre-vetting their traffic? Do you pay over the odds for a 'premium' site because it feels safe or do you let the raw data and results decide for you?
The media is only as 'quality' as the organic content of its audience.
If the demand side holds the supply side to account then click fraud will be reduced or suppliers will go out of business and good riddance. The demand side also needs to hold itself to account however and make sure it is making educated, evidence-based decisions.