If ad tech were a medical patient, a doctor would diagnose it with, among its other ailments, a bloat problem – specifically, bidstream bloat.
According to eMarketer, the volume of bid requests transmitted through the programmatic bidstream increased roughly 2.3 times between 2020 and 2023.
During that same period, spending on open web programmatic (outside of the walled gardens) grew only 18%, according to recent data from Jounce. This suggests the volume of traffic available to buy programmatically is increasing at a far faster pace than the budgets to purchase it.
Though some portion of this supply explosion is attributable to changing media consumption habits that arose during COVID-19 and the growth of new formats like CTV, the bulk of this bloat is due to more systemic causes.
With header bidding, for example, publishers commonly make their web inventory available to multiple SSPs simultaneously, creating many repetitive pathways for buyers to access the same supply. Mediation platforms play a similar role in creating duplicative bloat in the in-app ecosystem, while bid duplication and the rise of made-for-advertising sites also contribute to the problem.
More supply equals more costs, but not always more spend
Regardless of the causes, DSPs have finite capacity to process bid requests. Without a commensurate increase in ad budgets or QPS on the buy side, this surge in traffic volumes only generates more processing cost and carbon waste without creating any new value for advertisers, media owners or platforms.
SSPs, too, face rapidly increasing distribution costs for inventory that DSPs aren’t likely to bid on.
With efficiency, cost-savings, SPO and environmental sustainability top of mind for almost every programmatic stakeholder, it’s high-time to dust off a few of the existing curatives in the ad tech medicine cabinet.
Can traffic shaping be a cure?
Traffic shaping, which has been around in various forms since the early days of RTB, is the intentional optimization (or shaping) of the programmatic bidstream to send buyers more of the traffic they want to buy and less of what they don’t.
If done well, traffic shaping enables DSPs to reduce their listening costs without sacrificing campaign delivery or performance. SSPs, similarly, can enjoy lower distribution costs and higher win rates with DSP partners.
The ecosystem overall benefits from less waste and a considerably lower carbon footprint.
Unfortunately, the traffic-shaping tools of yesteryear aren’t always up to the challenge of today’s incredibly complex programmatic supply landscape, and it’s unclear who in the supply chain should own this process.
Deploying traffic shaping more broadly to increase programmatic efficiency and reduce waste will require investment and ecosystem-wide collaboration. But how do we get there?
Here are a few suggestions for how each party can get started.
1. DSPs: Reward top shapers
Start by analyzing the efficiency of your current SSP partners, looking specifically at the inbound traffic-to-bid response ratio. If possible, look at your cloud costs for listening and processing inbound traffic, and compare it to the fees you collect on media spend by partner. Work with lower-margin partners to optimize their connection or migrate spend to more efficient pipes.
Provide explicit guidance to SSPs around your traffic-shaping expectations. Ask what tools they have in place and how effective they are at optimizing the bidstream to your buying patterns.
Most importantly, always reward suppliers that fill up their QPS allocation with highly relevant supply and more spend.
2. Publishers and SSPs: Embrace your role as a shaper
DSPs increasingly prioritize supply partners based on their ability to send highly relevant, easily monetizable inventory. This represents both a challenge and an opportunity for SSPs.
For those with deeper pockets and strong data science experience, investing in your own traffic-shaping solution might make the most sense. This allows for maximum flexibility, customization and control. Others may choose a partnership approach to ensure prediction effectiveness and achieve faster time-to-market.
In either case, make sure you have a testing framework in place to understand the impact of bidstream compression on win rates and media spend, and the flexibility to adjust optimization levers as needed to balance buyer demands with publisher fill expectations.
Constantly evaluate the data available to you to improve your prediction models and work closely with your demand partners to ensure the service is adding the value they expect without degrading trading activities with added latency or reduced spend capacity.
3. Advertisers, agencies and media owners: Use your sway
As the parties that control spend and supply, you have a key role to play. Include traffic shaping in your vendor selection criteria and partner ESG questionnaires.
After all, working with partners that have effective bidstream optimization tools in place is good for your sustainability initiatives and may even drive up your programmatic performance.
Luckily, bidstream bloat is a lot like how the human body experiences bloat: It can be incredibly uncomfortable, but it’s rarely permanent.
What’s most important is putting the right tools, processes and incentives in place to effectively remove the junk and make room for more of the good stuff.
“The Sell Sider” is a column written by the sell side of the digital media community.
Follow Criteo and AdExchanger on LinkedIn.
For more articles featuring Barry Adams, click here.
