What is Supply Path Optimization in Programmatic advertising?
Exploring the evolution from traditional Waterfall and Header Bidding to today's Supply Path Optimization in digital advertising
Have you ever wondered how a Ralph Lauren ad appears on a news media website? News media plan advertising blocks on their sites, then find ways to sell these spots to advertisers. These ad spaces are their inventory, usually priced per thousand impressions. Media websites use tools like Ad Servers to enable advertisers like NIKE to Ralph Lauren to purchase these spaces, displaying them to the general consumer. Google Ad Manager is a common Ad Server tool that allows media to manage their inventory.
However, media can also have excess inventory. If no ads are shown when consumers visit the site, this leads to wasted inventory. This is where SSPs (Supply-Side Platforms) and Ad Exchanges come in. Their purpose is to help media sell remaining inventory to advertisers. Additionally, some SSPs may exclusively sell ads for specific media.
Traditional ad systems operate through a 'Waterfall' setup, prioritizing traffic to the SSP willing to pay the most. For example, if SSP1 is the first in line for an ad request, it is asked to bid first. If SSP1 decides to bid, its ad content is displayed to the consumer. If SSP1 doesn't bid, it moves to the next in line, SSP2, and so on. That’s why it’s called a Waterfall model.
Although this setup is straightforward, the decision-making process is a complex and opaque black box. For instance, Google Ad Manager has introduced a feature that allows bypassing the news media's set order, giving Google priority in buying traffic.
Around 2015, 'Header Bidding' emerged. Instead of relying on Ad Servers, it allows initial ad bidding to occur on the media's webpage, enabling media to control the actual ad bidding process. Under Header Bidding, it’s possible to simultaneously inquire with 5, or even 50 buyers in the first position. This parallel ad request method means each impression and audience can be bid on by multiple companies, maximizing every sellable spot and exposure for media like news outlets.
This is also beneficial for advertisers. They have more choices for each impression opportunity. We can think of SSPs as pipelines to inventory, allowing advertisers, even for the same media and exposure, to choose the most efficient, cost-effective, and best-performing pipeline to achieve the highest ROI.
However, Header Bidding has a side effect: for the same potential exposure, media may connect with 5 SSPs through Header Bidding, all of which might inquire the same DSP if they're willing to bid. This can result in DSPs receiving several times more ad requests, representing the same exposure opportunity. Even if DSPs bid on each request, they can only buy one exposure, creating significant system costs.
Each pipeline from advertiser to media contains valuable data, such as signals from SSPs, transaction fees, and performance data. Extracting and optimizing this data to find the most efficient path is known as Supply Path Optimization (SPO). Currently, The Trade Desk's OpenPath is an example of buyer-side SPO, while SSP Magnite's ad-buying platform ClearLine is an example of seller-side SPO, allowing direct business with ad agencies, bypassing DSPs.
Technologies and business negotiations like SPO that decentralize the ad platform bring more transparency to ad transactions, offering better prices for advertisers. We see both buyers and sellers gradually joining to share in the pie of decentralized ad platforms. The developments in 2024 are also highly anticipated.