Overview
Identity-enriched logs are Beeswax Bid and Win Logs that have been enhanced with identifiers from the FreeWheel Identity Network. These logs start with existing Beeswax transactional data and are then enriched by adding identifiers from FreeWheel’s Core Graph, such as universal IDs and hashed emails. The enrichment can help improve the accuracy of measurement and attribution by providing better identity data for individuals and households. This process occurs within the FreeWheel Data Clean Room, Powered by Snowflake, ensuring that the sensitive information remains protected. This setup enhances interoperability, scale, and precision in measurement results.
List of Supported IDs
| ID Name | ID Fidelity | Translation Required? |
|---|---|---|
| Experian Living Unit ID (LUID) | Household | Yes |
| Blockgraph ID | Household | No |
Hashed Emails (HEMs) For more information about Hashed Emails, click here. | Person | Yes |
| LiveRamp RampIDs (directly emitted to LiveRamp) | Person | Yes |
- Some universal IDs can be unique to each customer, and may require translation to a customer's version of that ID.
- ID Fidelity refers to the granularity of an identifier. This can either be at the person or Household level.
The FreeWheel Data Clean Room, Powered by Snowflake
The FreeWheel Data Clean Room offers two approaches to leverage Identity-enriched logs: Templates and Constrained Querying, detailed below:
What are Templates?
Templates are pre-approved queries that must be submitted for FreeWheel’s review before use. Upon approval, they become available within the clean room for usage. Templates can be accessed through the application’s UI, be automated and scheduled for execution at specific times. Within the UI, it is possible to preview the query, add dimensions to group by and apply filters.
- To use templates: Choose the query template to run, insert the dimensions, and execute them either ad-hoc or as scheduled workflows. Templates facilitate complex SQL operations that may not be supported under constrained querying.
- Data Controls: Aggregation is required with a minimum of 75 IDs grouped per row of output.
What is Constrained Querying?
With constrained querying, it is possible to write custom SQL queries instead of relying on templates. This is accomplished by applying the data controls on the outputs instead of the query itself. While this provides greater flexibility for querying, not all Snowflake SQL functions are supported in this approach. If any output row does not meet FreeWheel’s privacy requirements, it will not return any results and will display “null” in all columns.
- To use constrained querying: SQL queries must adhere to the projection and aggregation policy requirements. This approach allows for more flexibility in ad-hoc analytics or exploratory work.
- Data Controls: Projection Policies prevent sensitive columns from being included in the output. Aggregation policies require a minimum of 75 IDs per row of output. Refer to the Constrained Querying tab in the application UI for more details.
Identity-Enriched Fields
The identity-enriched logs are based on the standard Beeswax Bid and Win Logs. They contain most of the same fields, however specific fields have been omitted to ensure privacy compliance requirements. In addition to this, the following new fields can be found in these logs related to identity enrichment:
The <YOURCUSTOMID> part of the field will be populated with the ID type selected for enrichment.
| Field | Description | Data Type |
|---|---|---|
| YOURCUSTOMID_PERSON | Synthetic ID that will be present if the impression has person level matches to <YOURCUSTOM ID>. If not, will be blank. Can be used for aggregate person level reporting in constrained querying. | VARCHAR |
| YOURCUSTOMID_PERSON_IDS | Person level <YOURCUSTOM ID> matches, derived from the FreeWheel Identity Network's translation of signals found on the bid request. | Array |
| YOURCUSTOMID_HOUSEHOLD | Synthetic ID that will be present if the impression has household level matches to <YOURCUSTOM ID>. If not, it will be blank. Can be used for aggregate household level reporting in constrained querying. | VARCHAR |
| YOURCUSTOMID_HOUSEHOLD_IDS | Household level <YOURCUSTOM ID> matches, derived from the Freewheel Identity Network's translation of signals found on the bid request. A household represents a collection of person level IDs grouped to represent one entity. | Array |
Executing Templates
Please reach out to your Beeswax Account Representative to upload templates.
To submit and run templated queries, follow these instructions:
- Navigate to the Templates page where three tabs are available:
- Overview Tab: Introduces templates and where to submit queries.
- Settings Tab: Adjusts settings for templates with join conditions. See below for more details.
- Task Configuration Tab: Configure and edit any scheduled templated queries. More information is available in the Ingesting and Scheduling Templates section.
- Overview Tab: Introduces templates and where to submit queries.
- Selecting Templates:
The Select Your Template drop-down menu on the Overview tab allows for the selection of pre-approved queries. If the desired queries are not listed, FreeWheel can be contacted to request new templates. There are two types of templates available: Templated Queries without a Join Clause and Templated Queries with a Join Clause. - Templated Queries without a Join Clause:
After selecting a template, a preview of the template will be displayed. Dimensions for grouping can be chosen from the Select Group By Dimensions drop-down menu. Optionally, filters can be applied using a where statement, adhering to the syntax requirements of double quotes around the clause and double dollar signs around field values. Finally, submit the query by clicking the Submit Query button, with approval typically occurring within seconds. - Templated Queries with a Join Clause:
For templates queries that join to your data, there are additional parameters that can modified: consumer_schema, consumer_mapping_table, and consumer_internal_join_field. These parameters can updated in the Settings Tab:
- consumer_schema: Schema containing the table or view to join with.
- consumer_mapping_table: Table or view name to join with.
- consumer_internal_join_key: Column name to use as the join key.
- After updating the settings, return to the Templates page to view the updated template preview with the populated join table and join key.
Ingesting and Scheduling Templates
Submit Query for Execution:
- Select a query template.
- Add dimensions and an optional filter statement.
- Submit the query for approval.
Ingest Results:
- Once the query is approved, click the Ingest Results button located beneath the Submit Query button.
- Edit the following inputs:
- Target Database: Default is pre-filled (e.g., BW_ENRICHED_LOGS_DCR_CONFIG_DB).
- Target Schema: Default is pre-filled (e.g., MYDATA).
- Target Table: Specify the table name where results will be exported.
- Click the Execute Ingestion button.
- Confirm ingestion.
Schedule Ingestion Tasks:
- To schedule a recurring ingestion task, click the Schedule Ingestion Tasks button.
- Enter the following inputs:
- Scheduled Ingestion Task Name: Name for the task.
- Scheduled Ingestion Warehouse: Warehouse to use for the task.
- CRON Schedule: Frequency of the task.
- Click the Create and Schedule Task button.
- Confirm task creation and scheduling.
Configure Scheduled Tasks:
- Go to the Task Configuration tab on the Templates page.
- View and manage scheduled tasks.
- Options available: Suspend, resume, or execute tasks in the "Configure Tasks" section.