Price floors let buyers in a programmatic auction know they won’t be able to buy an impression cheaper than at a certain price.
So, if you set up a price floor of $5 and receive the following bids: $4, $5, $6 and $7, then only the last two bids will be considered. This is because you have eliminated everything else with your price floors.
Publishers set price floors to ensure that their inventory is not undersold. But how do you go about setting these floor prices? There are a number of parameters basis which the floor prices can be set:
Country
This is one of the most common parameters that publishers use to set floor prices. If you are a publisher with both Indian and US traffic, you would ideally want to start with running higher floors on the US inventory (where advertisers are willing to pay more) as compared to the Indian inventory (where CPM’s can be as low as $0.01 in some cases).
Device
Let’s take the example of a pizza brand. If you are in the vicinity of a local pizza outlet (and they can track that via your mobile device location turned on), there is a high chance that they will be willing to pay more for you on mobile devices. Alternatively, if you are a B2B company, and there is little chance someone will end up directly buying your product via mobile, floor prices for desktop can be higher.
Day parting
Most retailers who run stores prefer spending only during the hours their store is actually open. If you have the kind of inventory where these stores can target, it makes sense for you to actually raise the floor during the store opening time and reduce these once the store closes.
Seasonality
It is known that seasonality affects CPMs and publisher revenues. Take advantage of seasonal trends. If you are a financial site in India that helps viewers fill up taxes, you know that the period around the end of Q1 is something you can target where most advertisers will be willing to pay more. Even if you do not have a niche website, you can still try out things like increasing floors around Christmas (when advertisers are willing to pay more and exhaust their budgets).
Audience
If you are a news website that has different sections for Sports and Politics, for instance, you can leverage these audience sets to be targeted by different kinds of advertisers. During the Super Bowl, it is best to raise floor prices on the sports section, but during the US presidential elections, raising prices on the Politics section will drive results.
Ad Size
This is the lowest hanging fruit of them all. There are certain sizes of ad units that tend to work better for some kind of publishers in terms of buyer (DSP) demand. If for example, your 300×600 ad unit is outperforming the square 300×250 ad unit on average, it is advisable to set higher floors on the vertical unit as compared to the later.
Also read: Best Price Floor Optimization Partners to Increase your ADX Revenue
Manual or Automated?
All of this seems good on the surface but requires a lot of commitment in terms of man-hours. Someone has to continually monitor the bid ranges through a period of time to move the needle on price floors and all of this has to be incremental in nature (it makes no sense for you to set the floor price as $10 if on average, if your bids end up in the range of $2-$5).
How do you go about doing this if not manually? You can do it with Dynamic Price Floor Optimization (DFO) provided by some vendors. DFO allows you to determine the market demand for a particular audience in a particular time frame and price it optimally. DFO also automatically sets price floors, for instance, in your AdX based on calculations of factors mentioned above. These calculations and the speed of reaction (ability to quickly adjust prices) are the core of a Dynamic Price Floor Optimization.
Whether you try to do it manually or end up taking the help of a vendor, it is best if you fetch the bid ranges reports from time to time to get a better idea of where exactly is that most of the bids are being won/lost. Another important thing is to always set up some passback tags (the tag the impression passes to in case no DSP is able to match the floor price mentioned) to make sure none of the impressions go unsold.
Although a process that will not produce results overnight, floor prices have always been a go-to strategy for most big publishers, and most often than not, they end up reaping rewards in the long run if done right.
AskAdOps is a weekly column in which Shubham Grover, Product Specialist at AdPushup, answers commonly asked questions about ad operations. If you have a question, send in an email at askadops@adpushup.com