In Parent Segments, you can execute a list of enrichment actions by adding specific td_ columns to the behavior tables in the TD Console, as shown in this example:
Attribute Enrichment Options are enabled by default. However, if you don't have the required fields (td_client_id, td_ip, td_host, td_path, td_title, td_description) in your parent segment, we recommend disabling these options. Disabling these options reduces some tasks in the Audience Workflow build, which will help improve workflow performance.
For these types of enrichment, add the source column to the Attribute table definition when defining your master segment.
| Enrichment type | Description | Source column | Added output columns | Column description |
|---|---|---|---|---|
| Last Access Date | Adds the most recent customer access date. | td_client_id | td_last_access_date | — |
| Last IP Information Enrichment | Adds customer information by IP using TD_IP_TO_* functions. | td_ip | td_country_name, td_ip_city_name, td_ip_city_latitude, td_ip_city_longitude, td_ip_city_metro_code, td_ip_city_time_zone, td_ip_city_postal_code, td_ip_least_specific_subdivision_name, td_ip_most_specific_subdivision_name, td_ip_subdivision_names, td_ip_connection_type, td_ip_domain | See the Treasure Data IP reference. |
| Content Affinity Engine | Adds words of interest and interest categories based on content title and description. Uses OpenRTB’s category list. | td_host, td_path, td_title, td_description | td_interest_words, td_affinity_categories, td_affinity_category_scores, td_affinity_sub_categories, td_affinity_sub_category_scores | TF-IDF keywords and category scores. |
For these types of enrichment, add the source column to the Behavior table definition when defining your master segment.
| Enrichment type | Description | Source column | Added output columns | Column description |
|---|---|---|---|---|
| User Agent Parsing | Master Segments can automatically parse the user_agent string. | td_user_agent | td_user_agent_category | Category extracted from td_user_agent, for example element_at(td_parse_agent(td_user_agent), 'category'). |
td_user_agent_name | Name extracted from td_user_agent. | |||
td_user_agent_os | OS extracted from td_user_agent. | |||
td_user_agent_os_version | OS version extracted from td_user_agent. | |||
td_user_agent_vendor | Vendor extracted from td_user_agent. | |||
td_user_agent_version | Version extracted from td_user_agent. | |||
utm_ Parameters Parsing | Automatically extracts utm_campaign, utm_medium, and utm_source URL parameters used for online campaign attribution. | td_url | td_utm_campaign | Campaign name. |
td_utm_medium | Medium (for example, email, CPC ads). | |||
td_utm_source | Source (for example, Google, Facebook). | |||
| IP Information Enrichment | Adds additional customer information by IP using TD_IP_TO_* functions. | td_ip | td_country_name, td_ip_city_name, td_ip_city_latitude, td_ip_city_longitude, td_ip_city_metro_code, td_ip_city_time_zone, td_ip_city_postal_code, td_ip_least_specific_subdivision_name, td_ip_most_specific_subdivision_name, td_ip_subdivision_names, td_ip_connection_type, td_ip_domain | See Attribute Enrichment. |
| Web Crawler Access Exclusion | Removes access data generated by web crawlers so profiles contain only real human activity. | td_user_agent | — | The behaviors are removed; no additional columns are added. |
Based on the content's title and description, Treasure Data adds words of interest and estimates to profile interest categories. OpenRTB's list of categories is used as a master category list.
