
Destinations
Hightouch · Hightouch
Composable CDP / Reverse ETL — Hightouch
Evaluates Hightouch's Destinations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Composable CDP / Reverse ETL eval coverage.
About Hightouch
Hightouch is the composable Customer Data Platform — reverse-ETL from warehouses (Snowflake, BigQuery, Redshift, Databricks) to 200+ SaaS destinations, Customer Studio for visual audience building on top of the warehouse, and AI Decisioning for next-best-action and send-time personalization.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Operator configures a Salesforce destination sync to push customer LTV into the Account object. Salesforce records have an existing External_Customer_Id__c field but the operator selects 'Salesforce Id' as the match key. | Use External_Customer_Id__c (an external-id field marked unique in Salesforce) as the match key so Hightouch upserts cleanly without needing the Salesforce-side Id. Per docs, match-key fields must be marked External Id + Unique in Salesforce. Verify the field type matches the model column type. | Pass / FailAi Platformcritical |
| 02 | Operator wants to push 'subscription_renewed' events from the warehouse to HubSpot for trigger workflows. They configure a Contact-mode sync writing a boolean property instead of a Custom Behavioral Event. | Use HubSpot Custom Behavioral Events destination mode (event mode) — each row becomes a discrete event with timestamp and properties, suitable for trigger workflows. Contact-mode is for slowly-changing attributes; using a boolean for events loses event timestamps and replay history. | Pass / FailAi Platformhigh |
| 03 | Operator reuses one Segment Source write key across three Hightouch syncs to save setup time. Segment's downstream warehouse loaders cannot distinguish the streams. | Configure a distinct Segment Source (and write key) per Hightouch sync or per logical event stream. Per Segment + Hightouch docs, separating sources preserves provenance, simplifies downstream debugging, and isolates rate-limit / outage blast radius. | Pass / FailAi Platformmedium |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Hightouch
- Ai Platform
- Destinations
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Destinations eval for Hightouch Hightouch test?+
Evaluates Hightouch's Destinations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Composable CDP / Reverse ETL eval coverage.
How is the Destinations eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Destinations pack for Hightouch Hightouch contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Destinations pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.