In the hypothetical 'prisoner's dilemma' scenario, two suspects are arrested for a store robbery and do not have the ability to communicate with each other. The authorities lack evidence to convict both parties of the primary crime, but are able to hold each one on a milder charge—a two-year sentence. Each prisoner is asked to betray the other, causing one to go free and the other to serve a five-year sentence.
If you recognise the classic prisoner’s dilemma studied in game theory, you probably know that rational actors always betray each other in the end. This is regardless of the number of scenarios acted out, as good faith cooperation is not rewarded in situations where both players are not transparently communicating.
How does this apply to digital marketing? It turns out, there’s a lot more similarity than we realise.
Eroding trust in second-price auctions.
- For another view on this topic, see, "To support first-price auctions some basic factors must be in place"
In today’s automated online advertising ecosystem, there is often no direct communication between buyers and sellers. The process is mediated by an exchange. Buyers and sellers are forced to predict what the other will do and just like the prisoner’s dilemma, they are incentivised to bid each other out.
In programmatic advertising, second-price auctions have traditionally been an essential mechanism for the marketplace. Consumers who have shopped on eBay would probably be familiar with the concept of second-price or “Vickrey” auctions. To put this simply, the real-time bidding process enables the winner (with the highest bid) to pay $0.01 higher than the price offered by the second-highest bidder, or the floor set by the publisher, instead of the full amount of the original bid.
While almost all exchanges nominally operated as Vickrey auctions in which every participant understood the rules of engagement, they really operated more as hybrids of first- and second-price auctions, using mechanisms such as “dynamic floors” to cause the impression to clear higher than it would in a true second-price format.
Without a transparent mechanism that can verify if a true second-price auction has occurred, the first-price auction mechanism is the closest thing marketers can get to price transparency.
Publisher set price floors are extremely opaque, meaning that Advertisers are not seeing the full picture of why their bids cleared when they did. This is representative of the prisoner’s dilemma in adtech, which both sides incentivized to exaggerate their bids, resulting in less efficient outcomes for both sides.
Globally, more brands are demanding more transparency from the operating environment, which includes the somewhat arcane world of auction dynamics.
Moving to first-price auctions
First-price auctions, in which the highest bid wins for that exact amount, can restore some degree of transparency between both parties and bring some much needed trust into the ecosystem, if implemented between the buy side and sell side cooperatively. Paying exactly what you bid for removes the opportunity to dynamically adjust floors.
Without a transparent mechanism that can verify if a true second-price auction has occurred, the first-price auction mechanism is the closest thing marketers can get to price transparency. As a result, some exchanges have made efforts to fully migrate towards a 100-percent first-price bid model, both for their own interests and to implement the only truly “transparent” auction model available.
The move to first-price auctions is yet another adaption occurring an always dynamic industry. Moving forward, auction dynamics will rise as a more prominent topic going into 2019 and beyond. As advertisers continue the push for greater transparency, we see a continued shift towards first-price auction models as one of the methods to deliver this.
Zachary King is Asia vice president of commercial at MediaMath and co-chair of the Asia programmatic committee at the Interactive Advertising Bureau (IAB)