Skip to content

Latest commit

 

History

History
214 lines (169 loc) · 18.6 KB

README.md

File metadata and controls

214 lines (169 loc) · 18.6 KB
Raw

Facebook Ads Transformation dbt Package (Docs)

What does this dbt package do?

  • Produces modeled tables that leverage Facebook Ads data from Fivetran's connector in the format described by this ERD and builds off the output of our Facebook Ads source package.
  • Enables you to better understand the performance of your ads across varying grains:
    • Providing an account, campaign, ad group, keyword, ad, and utm level reports.
  • Materializes output models designed to work simultaneously with our multi-platform Ad Reporting package.
  • Generates a comprehensive data dictionary of your source and modeled Facebook Ads data through the dbt docs site.

The following table provides a detailed list of all tables materialized within this package by default.

TIP: See more details about these tables in the package's dbt docs site.

Table Description
facebook_ads__account_report Each record in this table represents the daily performance at the account level.
facebook_ads__campaign_report Each record in this table represents the daily performance of a campaign at the campaign/advertising_channel/advertising_channel_subtype level.
facebook_ads__ad_set_report Each record in this table represents the daily performance at the ad set level.
facebook_ads__ad_report Each record in this table represents the daily performance at the ad level.
facebook_ads__url_report Each record in this table represents the daily performance of URLs at the ad level. By default, this will exclude ads with NULL url values.
facebook_ads__url_tags Each record in this table is a unique combination of creative_id and the corresponding key, value, and type values contained in the url_tags field. This excludes creatives without url tags.

How do I use the dbt package?

Step 1: Prerequisites

To use this dbt package, you must have the following:

  • At least one Fivetran Facebook Ads connector syncing data into your destination.
  • A BigQuery, Snowflake, Redshift, PostgreSQL, or Databricks destination.
  • You will need to configure your Facebook Ads connector to pull the basic_ad pre-built report and its child basic_ad_actions and basic_ad_action_values pre-built reports. These pre-built reports should be enabled in your connector by default. However, to confirm these reports are actively syncing you may perform the following steps:
    1. Navigate to the connector schema tab in Fivetran.
    2. Search for basic_ad, basic_ad_actions, and basic_ad_action_values and confirm they are all selected/enabled.
    3. If not selected, do so and sync. If already selected you are ready to run the models!

Note: If you do not have your Facebook Ads connector schema change settings set to Allow all, it is very possible that you are missing basic_ad_actions or basic_ad_action_values. If you would like to surface conversion metrics in your Facebook and/or Ad Reporting models, please ensure these reports are syncing. Otherwise, the conversions and conversions_value fields will be null.

Databricks Dispatch Configuration

If you are using a Databricks destination with this package you will need to add the below (or a variation of the below) dispatch configuration within your dbt_project.yml. This is required in order for the package to accurately search for macros within the dbt-labs/spark_utils then the dbt-labs/dbt_utils packages respectively.

dispatch:
  - macro_namespace: dbt_utils
    search_order: ['spark_utils', 'dbt_utils']

Step 2: Install the package (skip if also using the ad_reporting combo package)

Include the following facebook_ads package version in your packages.yml file:

TIP: Check dbt Hub for the latest installation instructions or read the dbt docs for more information on installing packages.

packages:
  - package: fivetran/facebook_ads
    version: [">=0.8.0", "<0.9.0"] # we recommend using ranges to capture non-breaking changes automatically

Do NOT include the facebook_ads_source package in this file. The transformation package itself has a dependency on it and will install the source package as well.

Step 3: Define database and schema variables

By default, this package runs using your destination and the facebook_ads schema. If this is not where your Facebook Ads data is (for example, if your Facebook Ads schema is named facebook_ads_fivetran), add the following configuration to your root dbt_project.yml file:

vars:
    facebook_ads_database: your_destination_name
    facebook_ads_schema: your_schema_name

(Optional) Step 4: Additional configurations

Expand/Collapse details

Union multiple connectors

If you have multiple facebook_ads connectors in Fivetran and would like to use this package on all of them simultaneously, we have provided functionality to do so. The package will union all of the data together and pass the unioned table into the transformations. You will be able to see which source it came from in the source_relation column of each model. To use this functionality, you will need to set either the facebook_ads_union_schemas OR facebook_ads_union_databases variables (cannot do both) in your root dbt_project.yml file:

vars:
    facebook_ads_union_schemas: ['facebook_ads_usa','facebook_ads_canada'] # use this if the data is in different schemas/datasets of the same database/project
    facebook_ads_union_databases: ['facebook_ads_usa','facebook_ads_canada'] # use this if the data is in different databases/projects but uses the same schema name

NOTE: The native source.yml connection set up in the package will not function when the union schema/database feature is utilized. Although the data will be correctly combined, you will not observe the sources linked to the package models in the Directed Acyclic Graph (DAG). This happens because the package includes only one defined source.yml.

To connect your multiple schema/database sources to the package models, follow the steps outlined in the Union Data Defined Sources Configuration section of the Fivetran Utils documentation for the union_data macro. This will ensure a proper configuration and correct visualization of connections in the DAG.

Passing Through Additional Metrics

By default, this package will select clicks, impressions, cost, conversion, and conversion value (using the default attribution window) from the source reporting tables (BASIC_AD, BASIC_AD_ACTIONS, and BASIC_AD_ACTION_VALUES) to store into the output models. If you would like to pass through additional metrics to the output models, add the below configurations to your dbt_project.yml file. These variables allow for the pass-through fields to be aliased (alias) and transformed (transform_sql) if desired, but not required. Only the name of each metric field is required. Use the below format for declaring the respective pass-through variables:

Note Please ensure you exercised due diligence when adding metrics to these models. The metrics added by default (taps, impressions, spend, and default-attribution window conversion values) have been vetted by the Fivetran team maintaining this package for accuracy. There are metrics included within the source reports, for example metric averages, which may be inaccurately represented at the grain for reports created in this package. You will want to ensure whichever metrics you pass through are indeed appropriate to aggregate at the respective reporting levels provided in this package.

vars:
    facebook_ads__basic_ad_passthrough_metrics: # add metrics found in BASIC_AD
      - name: "new_custom_field"
        alias: "custom_field_alias"
        transform_sql: "coalesce(custom_field_alias, 0)" # reference the `alias` here if you are using one (otherwise the `name`)
      - name: "another_one"
        transform_sql: "coalesce(another_one, 0)"
      - name: "cpc"
    facebook_ads__basic_ad_actions_passthrough_metrics: # add conversion metrics found in BASIC_AD_ACTIONS
      - name: "_7_d_click"
        alias: "conversion_value_7d_click"
      - name: "_1_d_view"
    facebook_ads__basic_ad_action_values_passthrough_metrics: # add conversion metrics found in BASIC_AD_ACTION_VALUES
      - name: "_7_d_click"
        alias: "conversion_value_7d_click"
      - name: "_1_d_view"

Configuring Conversion Action Types

By default, this package considers the following kinds of custom, purchase, and lead action_types to be conversions and pivots their metrics (raw event frequencies and monetary values) as columns in each *_report end model:

Action Type Action Description (Meta docs)
offsite_conversion.fb_pixel_custom Custom pixel events defined by the advertiser. This will group together individual offsite_conversion.custom% custom conversion events.
offsite_conversion.fb_pixel_lead The number of "lead" events tracked by the pixel or Conversions API on your website and attributed to your ads. Off-Facebook leads, in short.
onsite_conversion.lead_grouped The number of leads submitted on Meta technologies (including forms and Messenger) and attributed to your ads. On-Facebook leads, in short.
offsite_conversion.fb_pixel_purchase The number of "purchase" events tracked by the pixel or Conversions API on your website and attributed to your ads. Off-Facebook purchases, in short.
onsite_conversion.purchase The number of purchases made within Meta technologies (such as Pages or Messenger) and attributed to your ads. On-Facebook purchases, in short.

These metrics will also be summed together into conversions and conversions_value fields in each *_report end model.

However, you may choose your own action_types to consider as conversions. To do so, provide each action type to the below facebook_ads__conversion_action_types variable. For each action type, provide either an exact name OR a consistent pattern of naming convention. You may also provide an optional where_sql argument for each action type, in case you would like to dynamically choose conversion actions based on other columns (ie source_relation if you are running the package on multiple advertisers' datasets).

# dbt_project.yml
vars:
  facebook_ads__conversion_action_types: # case-insensitive
    - name: exact_conversion_action_type_name # will grab `basic_ad_actions` and `basic_ad_action_values` records where action_type = 'exact_conversion_action_type_name'
    - pattern: onsite_conversion% # will grab all `onsite_conversion%` records
    - name: offsite_conversion.custom.my_custom_conversion_123
    - name: very_specific_conversion_action
      where_sql: source_relation = 'specific advertiser source' # will grab `basic_ad_actions` and `basic_ad_action_values` records where (action_type = very_specific_conversion_action and {{ where_sql }})
    - pattern: subscribe%
      where_sql: source_relation = 'advertiser who only cares about subscriptions' # will grab `basic_ad_actions` and `basic_ad_action_values` records where (action_type like 'subscribe%' and {{ where_sql }})

Note: Please ensure to exercise due diligence when adding or removing conversion action types. The action types added by default have been heavily vetted by our friends at Seer Interactive and the Fivetran team maintaining this package for accuracy. There are many ways to accidentally double-count conversion values, as some action types are hierarchical/aggregates or overlap with others. Reference the action type descriptions in the Meta API docs to ensure you select action types that appropriately and accurately fit your use case.

Change the build schema

By default, this package builds the Facebook Ads staging models (8 views, 8 tables) within a schema titled (<target_schema> + _facebook_ads_source) and your Facebook Ads modeling models (6 tables, 2 intermediate views) within a schema titled (<target_schema> + _facebook_ads) in your destination. If this is not where you would like your Facebook Ads data to be written to, add the following configuration to your root dbt_project.yml file:

models:
    facebook_ads_source:
      +schema: my_new_schema_name # leave blank for just the target_schema
    facebook_ads:
      +schema: my_new_schema_name # leave blank for just the target_schema

Change the source table references

If an individual source table has a different name than the package expects, add the table name as it appears in your destination to the respective variable. This is not available when running the package on multiple unioned connectors.

IMPORTANT: See this project's dbt_project.yml variable declarations to see the expected names.

vars:
    facebook_ads_<default_source_table_name>_identifier: your_table_name

(Optional) Step 5: Orchestrate your models with Fivetran Transformations for dbt Core™

Expand for more details

Fivetran offers the ability for you to orchestrate your dbt project through Fivetran Transformations for dbt Core™. Learn how to set up your project for orchestration through Fivetran in our Transformations for dbt Core setup guides.

Does this package have dependencies?

This dbt package is dependent on the following dbt packages. These dependencies are installed by default within this package. For more information on the following packages, refer to the dbt hub site.

IMPORTANT: If you have any of these dependent packages in your own packages.yml file, we highly recommend that you remove them from your root packages.yml to avoid package version conflicts.

packages:
    - package: fivetran/facebook_ads_source
      version: [">=0.8.0", "<0.9.0"]

    - package: fivetran/fivetran_utils
      version: [">=0.4.0", "<0.5.0"]

    - package: dbt-labs/dbt_utils
      version: [">=1.0.0", "<2.0.0"]

    - package: dbt-labs/spark_utils
      version: [">=0.3.0", "<0.4.0"]

How is this package maintained and can I contribute?

Package Maintenance

The Fivetran team maintaining this package only maintains the latest version of the package. We highly recommend you stay consistent with the latest version of the package and refer to the CHANGELOG, DECISIONLOG and release notes for more information on changes across versions.

Contributions

A small team of analytics engineers at Fivetran develops these dbt packages. However, the packages are made better by community contributions.

We highly encourage and welcome contributions to this package. Check out this dbt Discourse article on the best workflow for contributing to a package.

Contributors

We thank everyone who has taken the time to contribute. Each PR, bug report, and feature request has made this package better and is truly appreciated.

A special thank you to Seer Interactive, who we closely collaborated with to introduce native conversion support to our Ad packages.

Are there any resources available?

  • If you have questions or want to reach out for help, see the GitHub Issue section to find the right avenue of support for you.
  • If you would like to provide feedback to the dbt package team at Fivetran or would like to request a new dbt package, fill out our Feedback Form.