# Activation > Configures CDP activations for exporting segment/journey audiences to external destinations. Covers connection discovery, activation YAML structure, schedule options (daily/weekly/monthly/cron), column selection with masking, multi-table behavior data export (join up to 4 behavior tables via `behaviors:`), and notifications. Use when setting up destination exports for segments, configuring journey activation steps, or setting up schedules and column mappings. Always run `tdx connection list` and `tdx connection schema` first. - Skill: `treasure-data/activation` (Agent Skill, multi-file: 3 files) - Install (CLI): `npx skillmds add treasure-data/activation` - Raw SKILL.md: https://api.skillmd.com/api/skills/treasure-data/activation/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Product & Planning - Author: treasure-data (https://skillmd.com/u/treasure-data) - Updated: 2026-09-10 - Page: https://skillmd.com/skills/treasure-data/activation --- # tdx Activation - CDP Activation Configuration ## Discovery Commands (MUST run first) ```bash tdx connection list # List available connections (name + type) tdx connection show "Connection Name" # Show connection details tdx connection schema "Connection Name" # Discover connector_config fields tdx connection settings # Show connection-level credential fields tdx connection types # List all connector types ``` **CRITICAL**: Always run `tdx connection list` before writing activations. Do NOT guess connection names or connector_config fields. ## Segment Activations Activations are a **list** under `activations:` in segment YAML (vs. journey activations which use a **key map**). ```yaml activations: - name: SFMC Contact Sync connection: salesforce-marketing # Exact name from `tdx connection list` columns: - email - first_name - last_name schedule: type: daily # none | hourly | daily | weekly | monthly | cron timezone: America/Los_Angeles connector_config: # Fields from `tdx connection schema` de_name: ContactSync data_operation: upsert notification: notify_on: [onSuccess, onFailure] email_recipients: [team@company.com] ``` ## Journey Activations Activations are a **key map** (not a list). Steps reference keys via `with.activation`. ```yaml activations: welcome-email: # Key — referenced in step's with.activation name: Welcome Email Campaign connection: My SFMC Connection all_columns: true run_after_journey_refresh: true # Common for journeys connector_config: de_name: WelcomeEmails data_operation: upsert sms-reminder: name: SMS Reminder connection: Twilio SMS Connection all_columns: true run_after_journey_refresh: true connector_config: message_template: welcome_reminder ``` Journey steps reference activations by key: ```yaml steps: send-welcome: type: activation with: activation: welcome-email # Key from activations map ``` ## Column Selection ```yaml all_columns: true # Export all columns # OR specific columns: columns: - email - first_name - name: ssn # With visibility control visibility: masked # clear | masked ``` ## Schedule Options Use **either** `schedule` or `run_after_journey_refresh` (mutually exclusive). ```yaml schedule: type: daily # none | hourly | daily | weekly | monthly | cron timezone: America/Los_Angeles # none = manual trigger only # Advanced (optional): repeat_unit: week # minute | hour | day | week | month | once | none repeat_frequency: 2 # Every 2 weeks repeat_sub_frequency: [1, 3] # Monday and Wednesday (0=Sun, 1=Mon, ...) start_at: "2025-04-01T09:00:00Z" end_on: "2025-12-31" # Journey alternative (instead of schedule): run_after_journey_refresh: true ``` ## Notifications ```yaml notification: notify_on: [onSuccess, onFailure] # onSuccess | onFailure email_recipients: [team@company.com] ``` ## Behavior Data Export Join event/history from **behavior tables** alongside audience columns. Use the plural `behaviors:` list — each entry is its own mini-export (pick a table, how many rows per customer, which columns). Up to **4** behavior tables per activation. Rules the API enforces (each surfaces as a push error otherwise): - `order_by` is required on **every** entry (all strategies). `formatting` is required for `Top-N`; for `First`/`Last` it defaults to `rows` when omitted. - `behavior_table` must be the exact behavior source name from `tdx sg fields` (e.g. `behavior_purchase_history`). - The `order_by` `key` must also appear in that entry's `columns` (else: `'' is not included in export columns`). - Each behavior table used here must also be referenced by a `Behavior` condition in the segment `rule:` (else: `is not included in Segment rules`). ```yaml behaviors: # Up to 4 entries - behavior_table: behavior_purchase_history # Source name from `tdx sg fields` join_strategy: Top-N # All | First | Last | Top-N join_row: 10 # Row limit for Top-N (defaults to 1 for First/Last; unset for All) formatting: rows # rows | cols — required for Top-N order_by: # Required on every entry - key: timestamp order: descending # ascending | descending — applied before the join columns: # Plain strings, or `- name: col` with `visibility: masked` - timestamp # order_by key must be among the columns - product_name - order_total - behavior_table: behavior_page_views # Combine multiple tables in one export join_strategy: Top-N join_row: 5 formatting: rows order_by: - key: timestamp order: descending columns: - timestamp - url ``` **Prefer `behaviors:` (plural).** The singular `behavior:` (a single object, no list) is deprecated backward-compat and will be removed — emit `behaviors:` for new configs. Setting both `behavior` and `behaviors` is an error. ## Common Issues | Issue | Solution | |-------|----------| | Connection name wrong | `tdx connection list` to verify exact name | | connector_config fields unknown | `tdx connection schema "Connection Name"` | | Activation not running | Check schedule type and parent segment refresh status | ## Related Skills - **connector-config** - `connector_config` fields per connector type (SFMC, S3, BigQuery, etc.) - **segment** - Segment rule syntax - **journey** - Journey structure and activation steps