The DSP's optimization technology gives you different approaches to maximizing tactic performance:
Group budget optimization (GBO) automatically shifts the group's budget between tactics based on performance, with the goal of allocating more of the budget to the best performing tactics whenever possible. Enable GBO at the group level.
Algorithmic optimization (AO) adjusts the CPM bid price and status based on tactic goals that you set.
Machine learning optimization (MLO) leverages machine learning and a unique algorithm to calculate the best price per impression based on the overall probability of achieving the tactic goals that you set.
You cannot enable algorithmic optimization and machine learning optimization for a single tactic at the same time.
Disclaimer: Even after enabling optimization, you should still check tactics regularly to make sure that they're achieving the results you're looking for. For help troubleshooting tactic performance, contact support.
Algorithmic optimization automatically adjusts the bid price for your tactic's placements to an optimal value to achieve a KPI goal. This lets you recognize which inventory is performing better for your tactic objectives and concentrate spend on it, while investing less in inventory that isn't performing as well.
For information how to set up algorithmic optimization for your tactic, see Enabling Algorithmic Optimization.
When algorithmic optimization is enabled, the tactic bids on impressions, just as it would without optimization, and evaluates which domains and placements are helping the tactic meet the goal. After a learning period, the tactic creates rules that adjust bid prices depending on how a particular domain and placement is performing. Optimization rules can stop bidding altogether on under-performing domains and placements.
If your goal type is eCPC, eCPA, or eCPVC, you control the learning period by setting a learn budget, a dollar amount set aside for open bidding before any rules are created.
If your goal type is CTR or VCR, you control the learning period by setting an impression threshold, the number of impressions that will be won and analyzed before creating rules.
Machine learning optimization leverages machine learning and a unique algorithm to maximize a tactic's performance. To learn how to turn MLO on for tactic, see Enabling Machine Learning Optimization.
Machine learning operates in two distinct modes:
Learning mode: At this stage, the tactic uses the Default Bid (along with bid multipliers) to bid on impressions, and the optimizer captures information to create a model. This can take some time, as it must capture a certain number of actions (such as clicks, conversions, or viewable impressions) to create a useful model.
Optimized mode: When the learning phase is over and the tactic is optimized, the model that was created is used to decide if the DSP should bid on an impressions, as well as how much to bid, up to the maximum bid amount that you define. This overrides the tactic's default bid and ignores bid multipliers.
You can see a tactic's optimization status in Analytics.
Note: It might take some time for MLO to create a model and take over bidding. If the tactic has been running for some time and still isn't optimized, contact support for assistance.
In optimized mode, the campaign uses the KPI Goal value and the Maximum Bid to calculate the price to bid for the impression based on the overall historical probability of achieving the event that affects the KPI. Bid prices are unique for every impression and depend on many factors, including the inventory available, the timing of the impression, and the device that will display the impression.
To make sure that the algorithm builds an adequate model, we recommend a minimum spend of $5,000 per brand over the course of 2 weeks.