No two campaigns are the same. Monitoring real-time performance and continually refining your strategy are crucial to ensure every dollar is spent efficiently—but this presents a challenge when you’re managing several campaigns simultaneously.

A Bid Model is a high-performance machine that you can steer to optimize your campaigns. You create a set of rules that the Bid Model deploys as the campaign runs. With intuitive controls, you can adjust bids with Bid Modifiers, allocate budgets with Spend Ratios, and define priorities with Priorities to optimize performance in real time.

All advertisers can manage Bid Model components for their campaigns; however, a campaign must meet certain conditions before it becomes eligible. These conditions differ based on the individual Bid Model component that you’d like to apply.

Refer to these resources for more information on component-level eligibility requirements:

Campaigns support Bid Modeling on up to 100 dimension values.

When you apply a Bid Model setting, you define rules across one or more Bid Model components for a particular dimension. These components include Spend Ratios, Priorities, and Bid Modifiers. The Bid Model also includes the option to add individual dimension values to a campaign’s Allowlist or Blocklist when those dimensions appear in the campaign’s ad-serving data.

While these granular adjustments are optional, they allow you to take greater control of your campaign’s optimizations and, ultimately, its performance.

Refer to Spend Ratio overview, Priority overview, and Bid Modifier overview for more information about these individual Bid Model components.

Dimensions are the different targeting parameters that you can adjust by applying one or more Bid Model components to them. Bid Modeling is supported across the Creative, Location, Device, and Inventory dimension categories.

* Note that only Bid Modifiers aren’t supported for the Deal ID dimension.

** Note that Spend Ratios aren’t supported for the Publisher Category dimension. Refer to Spend Ratio overview for more information.

Bid Model components can be applied individually or together. When you apply multiple settings for the same dimension, ad opportunities pass through a decision-making process.

The Bid Model engine assesses whether serving is enabled for a dimension value at each stage. This assessment is based on the Bid Model settings you defined and the campaign’s current performance. Once the campaign confirms that the ad opportunity matches every campaign and Bid Model setting for that particular dimension value, it places a bid in the programmatic auction.

* This graphic is a simplified version of the logic that a Bid Model applies to determine whether or not it should bid on an ad opportunity when multiple Bid Model components are defined for the same dimensions.

Refer to the examples included Spend Ratio overview, Priority overview, and Bid Modifier overview for more details and examples of how these Bid Model components work individually and in tandem.

You have the option to manage Bid Model settings based on a given dimension value’s performance. You’ll add up to two performance filters by choosing a metric, selecting a condition (e.g., greater than, less than), and entering your desired value. For example, you may want to only review and model the ZIP Codes that earned more than 50K impressions.

When you choose two metrics for the same dimension, you’ll choose either an “OR” or an “AND” relationship between those filters. With an “AND” relationship, the Bid Model table will display a dimension if it meets both of the metric conditions you’ve set. With an “OR” relationship, it’ll display a dimension if it meets either of the conditions you’ve set.

Refer to these resources for step-by-step instructions on setting up performance filters:

You have the option to group together a set of dimension values within the same dimension. These groups are called Bundles, and should be reserved for instances when you’d like to apply the same Bid Model settings to multiple items.

A Bundle is a container or shell that you’ll use to manage the Bid Model settings of two or more dimension values from the same dimension. Instead of configuring their settings individually for each dimension value, you’ll apply them once at the Bundle level, and those settings automatically apply to everything inside.

While Bundles are optional, we recommend using them to improve the efficiency and organization of your optimizations. You can apply Bid Model settings once instead of many times, and ensure the settings of a set of subdimensions all stay in sync.

Refer to Spend Ratio overview: Spend Ratios applied to Bundles for more information on how your Spend Ratio settings are applied for the dimensions you’ve added to a Bundle.

Refer to the following guidelines while creating, managing, or applying Bid Model settings to a Bundle.

* Type of bundled dimension values: A single Bundle can include only items belonging to the same dimensions. For example, while both Cities and ZIP Codes belong to the Location dimension category, a single Bundle can include only Cities or ZIP Codes–not both.

** Child (dimension-value-level) advanced modeling: When you add dimension values to a Bundle, you’ll add Bid Model settings at the Bundle level. This means that modeling settings for each of the bundled items are disabled. You can remove a dimension value from a Bundle at any time to apply different Bid Model settings to it, if desired.

Refer to these resources for step-by-step instructions on saving supported dimensions to a bundle in the Bid Model app:

In certain cases, the granular campaign rules that you apply via Bid Model settings can begin to limit the number of bids that your campaign places in programmatic auctions. If a campaign featuring Spend Ratios, Priorities, and/or Bid Modifiers is not performing as expected, we recommend generating a Campaign Bidding Insights report for it. The report highlights which targeting parameters are limiting the campaign’s number of bid requests, which you can use to apply strategic optimizations and increase scale.

Refer to Campaign Bidding Insights overview for more information, and review “Reporting and analytics” collection overview for a complete list of reporting and insights resources.