I was in London a few weeks ago talking over a few Guiness Extra Colds with my main man Jake King and he dropped some gems on me. I admittedly know much less about the guts of RTB than I ought to. Now I know a little more. Thought I'd share (and see whether any RTB-ers can refute this).
Disclaimer: I am pro RTB and believe it's the future of all ad buying, but a little dose of cynicism never stopped a bandwagon did it?
When you buy via RTB, you’re buying based on an ID value in the system instead of actual referrer data. What this means is when you win the bid for that impression, you are winning it and paying for it based on what the exchange technology “thinks” it is, not where it is actually showing up.
Why does this matter? Because the exchange technology is no smarter than typical ad serving technology. If a publisher or network downstream of the exchange decides to place the tags anywhere, has dynamic page generation, adds sites to network, etc., the platform cannot tell this has occurred and therefore cannot uphold whitelists and blacklists.
This is due to the fact that the platform aggregates each inventory source and assigns an ID to it, but then downstream each traffic sourcing entity uses its own ID’s, creating the daisy chain that plagued ad nets and gave rise to verification.
So, the whitelist you are buying on, cannot be guaranteed unless the platform has a direct relationship with the publisher. In reality, <15% of the comscore top 100 pubs are accessible via exchanges... The right question to ask when buying RTB – "can you pick up the phone and call the publisher?”