Overview
If your organization uses DWA, you can access Helix data directly using SQL. Although this isn't necessary for most users, it enables advanced users to write powerful queries for specialized purposes.
Connect to your data warehouse as usual. The data is stored across five tables within the helix schema.
ERD (Entity Relationship Diagram)
Note: DWA uses Amazon Redshift, which does not enforce table constraints such as uniqueness of every row or primary key. The "PK" (private key) and "FK" (foreign key) notations in this diagram are solely for ease of use.
The grrecords table contains the same fields you defined during setup in addition to a system-generated ID string and timestamps. The attributes, tags, and mappings tables use that ID (gr_id) to associate any number of rows with a particular Golden Record. There is not a limit for how many tags, attributes, and data set records a Golden Record may connect to.
Data dictionary
| Table name | Column name | Data type | Notes |
| grrecords, attributes, tags, mappings |
gr_id | varchar(100) | Represents a single contact |
| grrecords, attributes, tags, mappings |
helix_processedon | timestamp | Time when Helix processed this record |
| ALL | processed_date | timestamp | Time when this record was pushed from Helix into DWA |
| grrecords | CUSTOM-FIELD-NAMES | varchar(500) | The field's API name |
| attributes | attribute | varchar(100) | The attribute's API name |
| attributes | value | varchar(100) | For use with Audience Builder conditions |
| tags | tag_name | varchar(100) | The tag's API name |
| mappings | helix_recordid | varchar(200) | Formed from the data set's name and the record's ID |
| integrationlogs | type | varchar(100) | Name of the table changed |
| integrationlogs | changed | int8 | Number of records updated in DWA during the push at the referenced time |