TATreasure AI

The Performance tab on a one-off Engage Studio campaign shows unique opens, clicks, bounces and complaints so marketers can review results without a separate BI tool or query.

List Campaigns let marketers in Engage Studio send one email to a contact list built from a CSV file, skipping segment setup, using up to 20 list columns as merge tags.

The Audience Studio UI now supports configuring multi-table Behavior Activations with up to four mapped behavior tables, for batch and journey activations, without the API.

AI Studio's command approval selector now offers Ask, which requests approval before tdx commands, or Auto, which runs them without the approval card, set per chat.

Personalization Studio audiences are now built directly in its rule editor from batch and real-time attributes, instead of importing batch segments from Audience Studio.

MCP Endpoints now support OAuth, letting each person sign in individually from their MCP client so chats and Actions run under their own permissions and credits instead of a shared connection token.

In beta, administrators with MasterSegmentConfig access can delete all Real-Time profiles for a parent segment to reset from an empty state; beta access requires contacting a representative.

Treasure AI Studio is available as a native Android app on Google Play, syncing chats, work folders, skills, and data access with the Web and Desktop apps via server-side execution.

AI Studio agents now show a question card asking approval before running a tdx command, including read-only queries, except on action-schedule runs.

Treasure AI Voice for iOS accepts deep links from other apps and saves their query parameters as external metadata on the resulting recording, exported to Treasure Data CDP.

Real-Time Profile Lookup lets you investigate a profile's current identity graph via Treasure Workflow, returning resolved IDs and persisted attributes for debugging.

The dashboard shows 18 months of monthly resource usage across an organization's Treasure AI accounts, letting owners and admins compare usage by account, track totals against plan allowances, and view daily breakdowns.

New Adform Export connector sends audience definitions and members to an Adform-managed S3 bucket for DSP targeting; currently in beta.

Customer.io Export connector syncs Treasure Data CDP profiles and segments to Customer.io after each workflow run, supporting create, update and removal.

Personalization Studio campaigns can now be paused and resumed from the campaign card, stopping content delivery without deleting the configuration.

Real-Time now records identity decisions to validation_failures and id_changes queryable tables, tracking stitching outcomes and evictions.

Airship Import Integration adds Channels as a data source, bringing device-level opt-in state, tags and attributes into Treasure AI.

Engage Studio adds link shortening and click tracking for SMS campaigns, using a short domain and async click-event recording with bot detection.

Composable Audience Studio now tags its Snowflake queries with structured metadata, letting teams attribute warehouse costs and monitor usage per audience.

Hive query engine now supports DROP TABLE, completing the basic table lifecycle alongside the CREATE TABLE support added in July, without switching to Trino.

Account admins can self-manage usage overage notification settings: alert thresholds, recipients, and weekly or monthly report delivery.

Treasure Code CLI gains commands to view and manage Personalization Studio campaigns, plus a saved session context for repeated commands.

Saved AI Studio actions can now run automatically on a repeating schedule or cron expression, with run history and an enable switch.

Batch Journey A/B Test steps can now reassign a fresh variant when a profile re-enters, avoiding stale sticky assignments across recurring runs.

BigQuery access to Complete CDP tables through Open Catalog Access is available as an experimental integration.

Always-on campaigns can now be pulled, edited locally as YAML, and pushed back via Treasure Code, matching CLI workflows for templates.

AI Studio chats can now read, write and move files across all of a user's work folders instead of only the current one.

Each AI Studio chat now runs in its own isolated environment; files the AI writes outside a work folder are deleted with the chat.

Behavior Activation now supports up to four linked behavior tables per activation via the Audience API, up from a single table; Console support pending.

Meta Dataset Quality Import connector now tracks quality scores for offline conversion events alongside existing web event metrics.

Salesforce Marketing Cloud V2 export connector now supports composite primary keys of up to four fields when creating Data Extensions.

New export connector uses Google's Data Manager API to keep sending enhanced conversion events to Google Ads after Google retired the legacy API.

LINE OA becomes a supported delivery channel for Always-on Campaigns in Engage Studio, on parity with email and push, supporting text, Imagemap and Flex messages.

Batch Export connector adds an Audience Export target that syncs Treasure AI segments to Batch audience lists for campaign targeting.

Yahoo! JAPAN Ads connector now migrates LINE Ads Custom Audiences with automatic identifier mapping, ahead of LINE Ads API's October 2026 sunset.

AI Agent Foundry's Nova Canvas image generation is being retired by AWS, ending no later than end of September 2026, with no replacement model planned.

Treasure AI Studio ships as a native Windows desktop app, bringing the existing web and Mac experience to Windows for the first time.

Hive query engine in Data Workbench adds production-ready UPDATE statement support, using the same WHERE-clause and partition guidance as DELETE.

Trino query engine in Data Workbench gains an experimental UPDATE statement, alongside Hive's production-ready implementation.

AI Studio adds a file browser and manager for work folders: create, rename, move, and delete files and folders without opening a chat.

Engage Studio adds a dedicated Pre-header field for email preview text, with Liquid personalization and character limits, in both editors.

Hive query engine in Data Workbench now supports CREATE TABLE statements, including comments, retention and partitioning, without switching to Trino.

Account admins can set an org-wide custom instruction added to every agent's system prompt in AI Studio, applied automatically to new chats.

Custom skills built in an AI Studio project can now be shared account-wide as a private plugin, installed by teammates without a public GitHub repo.

AI Studio can now prompt an in-chat Connection Required card so users authorize Databricks or Snowflake access without leaving the conversation.

AI Studio is now available in Korea (ap02), keeping LLM inference within Korea via Amazon Bedrock Seoul; only the Balanced model tier is offered there.

HTML Email Hosting serves a launched email campaign as a web page for a 'View in Browser' fallback; in beta, requires Email Domain and support enablement.

Iterable Export connector adds a truncate mode that replaces a list's membership in one job instead of separate add and remove jobs.

Policy-based permissions now extend to Composable Audience Studio parent segments and folders, letting admins grant full, limited or no access per segment.

Account admins can restrict AI Studio to an approved set of users through the Treasure AI REST API; the restricted mode is enabled on request by the Customer Success Manager.

Treasure AI Voice adds personal Summary Templates that control how call and meeting summaries, action items and titles are generated.

Real-time personalizations and journeys add an opt-in option to include profiles with no batch data yet, so new users aren't wrongly excluded.

New import connector pulls Batch campaign engagement data - sends, opens, reachability - into Treasure AI tables on a schedule for attribution analysis.

On-Demand Journey Execution lets marketers trigger a journey immediately using the latest segment data, without waiting for a full segment rebuild.

AI Studio adds a Credit Usage dashboard and a Credit Policy letting administrators monitor and cap per-user daily AI credit consumption.

Composable Audience Studio now emits premium audit log events for segments, folders, syndications, templates and bulk operations, matching Audience Studio parity.

New connector sends audience data from Treasure AI into Selligent (Marigold Engage) Lists and Segments, replacing custom workflow pipelines.

AI Studio agents can now search the web in real time during a chat, pulling in current information automatically when needed.

Engage Studio adds LINE Flex Message support, letting marketers paste Flex Message JSON to build rich layouts like carousels and product cards.

New connector sends offline and server-side conversion events from Treasure AI to Yelp Ads for attribution and ROAS measurement.

New connector sends conversion data to Google Campaign Manager 360 tied to Floodlight activities, for cross-channel attribution reporting.

Workflow pools let administrators cap how many workflow attempts in a chosen group run concurrently, preventing one group from starving others.

Workflow Trigger automatically starts a downstream workflow attempt when an upstream workflow succeeds, replacing manual call/require dependencies.

API key expiration policies are now GA, letting admins enforce automatic key rotation with system-wide or per-user expiration lifespans of 1-365 days.

Hive query engine in Data Workbench now supports DELETE FROM ... WHERE, previously available only in Trino, for large-scale row removal.

Engage Studio can log email engagement events using non-PII identifiers instead of raw email addresses, preserving traceability while limiting PII at rest.

Real-Time journeys add a 'Never allow re-entry' option so a profile can only enter the journey once, even if it re-qualifies later.

Salesforce Marketing Cloud export connector adds row-based data retention, letting each exported row expire independently of the Data Extension's default policy.

Real-Time ID stitching now caps each profile at 200 unique IDs, automatically ejecting the oldest to prevent over-stitching and protect performance.

Six AI-powered debugging skills installable via Treasure Code generate diagnostic queries for Real-Time activation, identity and journey issues.

Composable Audience Studio extends to Google BigQuery with the same zero-copy, federated segmentation and activation architecture used for Snowflake and Databricks.

Salesforce Import and Export connectors gain OAuth2 Client Credentials login via External Client Apps ahead of Salesforce's Connected App deprecation.

Treasure AI announced it will remove weak TLS 1.2 cipher suites platform-wide, continuing hardening after retiring TLS 1.0/1.1.

New export connector sends conversion data to Yahoo Japan's updated Conversion API endpoint, replacing the retired v1 connector.

New Databricks Workflow Operator lets Treasure Workflow submit and manage SQL and ETL processing directly on a customer's Databricks environment.

New Snowflake Workflow Operator lets Treasure Workflow run SQL directly on a customer's Snowflake warehouse for ETL and data prep steps.

AI Foundry agents now execute with the permissions of the end user running them rather than the agent's original creator, improving access governance.

Klaviyo Batch Export connector adds Klaviyo's Subscription (Consent) API support to sync true customer consent status in bulk, not just list membership.

New import connector pulls LINE Official Account insight data - delivery, followers, demographics, interactions - into Treasure Data for reporting.

Campaign editor support for FCM mobile push in Engage Studio: message type, tap behavior, image validation and delivery error logs; enabled on request by the Treasure account team.

Snowflake Import connector adds bulk unload via Amazon S3 for higher-volume data movement into Treasure Data.

Treasure Data will drop support for TLS 1.0/1.1 on customer-facing endpoints effective February 28, 2026, requiring TLS 1.2 or higher.

AI Agent Foundry lets organizations configure how long chat data is retained, to match each company's data retention preferences.