Understanding Datasets (Formula Sources)

When writing a Formula in the Composer, the first crucial step is selecting the Dataset. The Dataset acts as the Source for your formula, determining the level of your advertising structure that your formula will analyze and filter.

What is a Dataset?

Each Dataset represents a specific layer or type of entity within your Amazon Ads account structure. Think of it as choosing the table of data you want to work with. The Dataset you select dictates:

1. The Level of Granularity: Are you looking at overall Campaigns, specific Ad Groups, individual Keywords, or something else?
2. Available Properties: Different properties are available depending on the dataset chosen. For example, the budget property is available for the Campaigns dataset, while the bid property is available for Keywords and Targets. Time-based metrics like clicks and sales are aggregated differently depending on the dataset level.
3. Available Workflow Properties: Different properties are available to be modified through Workflow actions depending on the selected dataset. For example, to change the bid of a keyword you'll select the Targeting or Keywords dataset.

Available Datasets

Here are the primary datasets you can select as your Formula source and their available Workflow properties:

Campaigns

Represents your top-level advertising campaigns. Formulas using this source will filter and analyze data aggregated at the campaign level. Properties like campaign name, budget, and state are available.

Workflow Properties: state, budget

Ad Groups

‍Represents the ad groups nested within your campaigns. Formulas analyze data aggregated at the ad group level. Properties like ad group name, default bid, and state are available.

Workflow Properties: state, default bid

Keywords

‍Represents individual keywords within your manual targeting ad groups. Formulas analyze performance specific to each keyword. Properties like bid, match type, and state are available.

Workflow Properties: state, bid

Targets

‍Represents product targets (ASINs, Categories) in manual targeting ad groups, and automatic targeting groups (Close Match, Loose Match, Complements, Substitutes) in automatic campaigns. Formulas analyze performance specific to each target or auto-group. Properties like bid, match type (for auto-groups), and state are available.

Workflow Properties: state, bid

Targeting

‍A combined view representing both the Keywords and Targets datasets. Useful for analyzing overall keyword and target performance together within ad groups.

Workflow Properties: state, bid

Choosing the Right Dataset

Select the Dataset that corresponds to the level you want to filter or analyze. If you want to find specific keywords to adjust bids for, choose Keywords. If you want to find underperforming campaigns to pause, choose Campaigns.

For a complete list of properties available for each dataset, see the Syntax Reference page and use the "Datasets" filter.

‍