What is Ad Fraud?
It’s common to think of ad fraud as consisting solely of bots repeatedly visiting websites and clicking ads. These bots artificially deflate CPM (Cost per 1000 impressions) and CPC (Cost per Click) for the site, motivating advertisers to continue serving ads on the site. But unfortunately, this is just one kind of ad fraud. The complicated nature of the RTB process, combined the number of parties involved in the Programmatic buying process, has caused a wide range of fraudulent activity to develop.
For example, a click farm employs people to click and engage with ads on certain sites to improve figures. These employees aren’t bots – they’re real people. But this is still an example of fraudulent activity where the goal is to manipulate the metrics which often guide RTB strategy (clicks and impressions).
Domain spoofing has a different goal. A fraudster can masquerade their own site as another more reputable site to win advertiser spend. Advertisers, naturally enough, want their ads showing on sites with premium content. A major example of domain spoofing happened just last year. The Financial Times was having their site mimicked and up to £1,000,000 a month was being spent by advertisers on serving on a site that was not the real FT.com. One more example is ad stacking. Bad actors place multiple ad spaces in the same spot on their website so that ads are ‘stacked’ on top of each other. Obviously, only the top ad is visible to the user but many impressions are counted because the display ad is serving multiple times.
What’s the big deal?
The problem is that for many digital marketing campaigns there is a rush towards high-volume low-cost inventory. You might think: CPM fees ought to be kept low and CTRs high. Bad actors are all too happy to play to this simple bidding strategy.
Alternatively, fraud occurs because Programmatic advertising can be complicated because of the number of parties involved. If a brand wants to serve a display ad programmatically they have to use a Demand Side Platform (DSP) which communicates with an Ad Network which communicates with a Supply Side Platform (SSP) (and they might use an agency to do this all for them!). Unfortunately, opportunities for fraud occur at each step in the supply chain.
How do Merkle | Periscopix reduce the risk of ad fraud?
We’re platform super-users - Whilst the DoubleClick stack might be our forte (we’ve been DoubleClick partners since 2015) we also have experience using Amazon Advertising Platform, BrightRoll and AppNexus. We bid on inventory using these platforms so it’s important to know them back-to-front.
Transparency is at the core of our business - When reporting on the performance of your campaign your account manager builds up the clearest possible picture of your account and its performance. Regular checks, calls and reports make it hard for fraud to go unnoticed.
We can read between the lines of performance data – to analyse performance well, we don’t just take metrics on face value, instead they need to be tied back to business outcomes and the wider industry landscape. We also like to use a wide range of performance metrics which makes spotting fraudulent activity easier.
Tried and tested ad verification measures - Over the years we’ve been building up a best practice that aims to make the most of your advertising budget. All major third-party vendors of ad verification have been tested and audited by our team of specialists – meaning that, if you choose to use 3rd party vendors, we can recommend the best for your needs.
We also demand transparency throughout the supply chain - From publishers to DSPs, we demand transparency on our partner’s own ad fraud measures and level of sophistication. This is a part of our onboarding process, but also stitched into the relationships we maintain with them.
We communicate regularly with publishers on what they’re doing about ad fraud - The earlier example of the FT experiencing domain spoofing illustrates this point nicely. They chose to tell all their agency contacts about the fraud they discovered. By doing their part to clean up the programmatic buying ecosystem they re-directed the advertising spend back into the legitimate programmatic supply chain and, in this case, back to the Financial Times!
We’re continuously improving - Any sites which are clearly fraudulent end up on a blocklist which is added to all your campaigns. Whether we find them from our own analysis or through our partnership with the City of London police who regularly publish an infringing website list. Read about operation creative here.
It’s a group effort.
To resolve the problem of ad fraud all players within the programmatic ecosystem should step up their game and do their part to call out bad actors and make fraudulent practice as difficult as possible. We at Merkle | Periscopix are committed to playing our part, and we’re eager to hear from anyone interested in bringing about a fraud-free future. Get in touch!