You can adjust all of your cookie settings by navigating the tabs in this window.
This is the first post in our new “TIL (today I learned)” blog series that puts a spotlight on specific digital advertising technologies and strategies. Today we give you the lowdown on Facebook’s “Minimum ROAS Bidding.”
Minimum ROAS Bidding is the next logical progression for Facebook’s objective-based buying strategies. Facebook’s algorithm technology has evolved over the last decade from handling upper funnel metrics such as Installs and Registrations to down-funnel objectives like “In app purchases” and “add to carts.” Depending on your advertising goals, Facebook can now algorithmically show your ads to people more likely to help you achieve your objectives.
More recently, Facebook has allowed advertisers to bid on users more likely to drive downstream revenue with a feature called “Value Optimization.” Minimum ROAS bidding adds refinements to that feature. It allows users to define the minimum amount of ROAS (return on ad spend) they would require these ads to produce and then bid on users more likely to drive that specific minimum amount of revenue — and ideally more than that.
The biggest problem that Minimum ROAS Bidding solves is that it gives marketers more control over their Value Optimization bids. Value optimization can only be used in combination with Highest Value bidding (formerly known as Autobid) — a method that puts the onus on Facebook’s machine learning and artificial intelligence to find users who will drive the highest value. Users with high value are those who resulted in a purchase or install at the lowest cost possible.
"Let’s say an advertiser wants the minimum amount of return to be $3. That number will tell Facebook to only show ads to users who it believes will drive at least three dollars in return on ad spend."
As a result, marketers using Value Optimization were in effect using Facebook’s algorithms to show ads to users who would drive “value”, but without defining what “value” really meant.
Contrast that with a “Lowest Cost with Bid Cap” bidding strategy, which can be used with goals such as installs, purchases, and registrations. This type of bid effectively has a cost ceiling, which gives Facebook the leeway to locate the ideal installer, purchaser, or registrant, but within defined cost perimeters.
Minimum ROAS Bidding does something similar for Value Optimization as “Lowest Cost with Bid Cap” does for purchases or installs, but in reverse. Think about it as “Highest Value with Bid Cap” where the minimum ROAS is the value floor. Let’s say an advertiser wants the minimum amount of return to be $3. That number will tell Facebook to only show ads to users who it believes will drive at least three dollars in return on ad spend.
While it’s still early days for this feature, Minimum ROAS Bidding could be a boon for any direct response advertiser, but especially those that take a complex but structured approach to their Facebook advertising.
Both Nanigans and Facebook have always encouraged advertisers to vary their campaign objectives based on the types of audiences they target. For example, we recommend Value Optimization for targeting groups that are geographically broad and/or haven’t proven to be successful in terms of conversions or purchases. We also recommend optimizing toward installs or purchases with more focused and historically high-converting audiences, and showing your ad to as many users as possible in this group.
Minimum ROAS Bidding allows highly analytical advertisers to add more tiers to their optimization strategy. They can separate their targeting groups into “Value Optimization only” for audiences that have shown some success in the past, and “Value Optimization with a ROAS floor” — using Minimum ROAS Bidding — for new and unproven audiences that need to be explored more cautiously.
Even advertisers with a less structured approach to their direct response ad spend will find Minimum ROAS Bidding useful. They can now give Facebook’s industry-leading bidding algorithms more data regarding revenue goals. And when there are more signals for the Facebook algorithms to work with, the results tend to be better for the advertiser.