During the period where Google was on a second-price auction model and before it moved to the first-price model, there was some kind of a cat-versus-mouse game, where publishers endeavoured to position their soft floor on a sweet spot that sits exactly between the highest bid and the second highest bid.
Let’s explore some very simple examples from the the old world:
Scenario 1 – No soft floor :
Scenario 2 – A better targeted soft floor :
Scenario 3 – A SUPER optimised soft floor:
At the time, for the same auction, a well optimised soft floor used to be a huge game-changer in terms of revenue. Many publishers had created their “secret sauce” based, for example, on the average of bids or on many other strategies.
The second-pricing auction model had even created a lucrative market where many companies specialised in the subject, by creating predictive algorithms that aimed to continuously position your soft floor as close as possible from the highest bid, in order to maximise revenues and at the end mimic a first price auction.
Moreover, the uplift was easily measurable which was one of publishers’ favorite optimisations.
Now things have changed!
Unless you’ve been living under a rock during the last six months, chances are you’ve heard Google AdX has moved to first-price auction this last September. Google was the last to do so, long after Appnexus and Rubicon, etc.
Meaning second-price auction belongs to the past now.
So, what would we get if we replayed the same aforementioned scenarios in a second-price auction world?
Scenario 1 – No soft floor :
Scenario 3 – A SUPER optimised soft floor:
You got it, no more magic formula or secret sauce that will allow you to go get an easy uplift! Which by the way was one of the reasons that pushed Google to switch to first-price and thus limit all the auction manipulation that used to annoy advertisers.
This, as you may have understood, made things much more complex and by the same way much less lucrative for companies which specialized in the topic. Those same companies must have had a very hard time after Google’s announcement.
No, floor pricing is not dead. You still need to protect your inventory, as advertisers keep trying to lower their bids, even in a first-price model, in order to optimize the efficiency of their budgets.
Theoretically speaking, there is one mitigating factor that consists in a self-regulating market (poke Adam Smith). Which means, that in a very competitive market, if an advertiser lowers his bid, he could lose it to the benefit of another advertiser, without floor prices being involved.
That being said, the problem is mathematically very complex: there are as many bidding strategies as there are campaigns out there. The question is, how to prove that a certain floor price properly protects your inventory and how to push algorithms to bid higher ? This is where all the problem stands!
Simply said, floor price optimisation in a first-price auction world is built around the idea that an advertiser goal is to keep lowering his bids in order to win the auction while paying as little as possible. If you want to prevent them from doing so, you need to protect your inventory by increasing your floors.
However, we highly recommend not to take this path without having very clear ideas and by being prepared to dedicate a lot of time to properly deal with the subject.
Rules have changed!
If you’re not yet entirely familiar with how to set up your floor pricing, we’ve got you covered ! Here’s our step-by-step guide to retain maximum value through your floor setup : 4 steps to setting up floor prices that drive maximal ad revenues.
However if you’ve already started delving into your floor pricing, we’ve also got something to help you out ! Here are 2 floor-pricing mistakes you should absolutely avoid, we hope you’ll find them useful !
.png)
%20(2).png)
