"Data-Driven Thinking" is written by members of the media community and contains fresh ideas on the digital revolution in media.
Today’s column is written by Bosko Milekic, co-founder and vice president of technology at AdGear.
After scoring space for your client’s ad on a few premium websites, the reports rolling through your demand-side platform show high volumes, low CPMs and good viewability. But then your client calls because they can't find their ad on any of those premium websites, and when you look, neither can you.
You’ve been gamed. I call it “arbitrash.”
“Arbitrash” is the reselling of impressions bundled in a way that doesn’t add value, often while falsely declaring that the impressions are from high-quality websites. It is yet another kind of ad fraud to combat. Ad technology companies are still working out a foolproof way to counter these attacks, but there are ways that marketers can fight back.
There are no bots involved, the web traffic is from real browsers and the spots being resold may score highly when looking at viewability data. Since falsely declared ad traffic blends in with real ad traffic, advertisers inadvertently buy underlying ads at CPMs exceeding the true value of the ad space. Everyone in the value chain benefits, except the advertiser.
Taking Out The ‘Arbitrash’
Demand-side technology is getting better at detecting and avoiding bot traffic patterns, but it is harder to detect “arbitrash” because the impressions are from real sources. Most ad exchanges audit creative tags to ensure that ads identify a single, specific, legitimate brand, yet auditing is hard to fully automate and, in practice, creative tags foil audits time and time again.
While technology platforms work to stop “arbitrash,” marketers can also take measures against this kind of fraud. They should take advantage of RTB features such as deliberate seller selection, transparent inventory, seller identity and connectivity to each seller and marketplace. Avoid opaque marketplaces, where underlying publishers are not identified. Brand marketers should understand the layers of deception at work in the digital wild and learn how they can be avoided.
In video marketplaces, beware of channel arbitrage – sometimes without domain identity theft – from sellers purchasing display ad inventory and rebranding it as video. While it’s not always necessarily deceptive, it’s good to avoid.
If buying media from an ad network or desk through an IO, ask about the ownership of the bundled impressions being purchased, as well as representation details. Determine how it is being delivered from the web page onward. Understand what value the network or trading desk adds, in terms of know-how, intelligence and execution. Marketers should insist on seller identity or avoid the inventory altogether if the seller doesn’t provide publisher identity or actively curate its publisher and network relationships.
Understand also that media metric or KPI arbitrage makes sense, but only when an ad network offers a media product on a fixed cost-per-click or cost-per-action, and takes the risk of brokering it on the open markets, disclosing the inventory sources and publishers from which the media is purchased. When running branding campaigns, buyers should be selective with sellers and the various domains and channels of inventory where their ads will appear.
What else can you do to avoid the call from an unhappy client who has discovered their ad is nowhere to be found on the premium website you promised? Following precautions against “arbitrash” will help, but there is no guarantee the organizations you work with are telling the truth about ad space identity.
It’s really still a precarious game of trust. Winning the game isn’t out of the question. Do due diligence, promote transparency in the programmatic ecosystem and encourage publishers to declare their identity openly in marketplaces. This will help to gradually change the system.