Operations Observability Alerts
Hightouch · Hightouch
Composable CDP / Reverse ETL — Hightouch
Evaluates Hightouch's Observability, Run History & Alerts 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 | Sync to Salesforce fails. Alert reaches Slack as 'Sync 1234 failed' with no destination, no error message, no rejected-row count. | Alert payload must include sync name, destination, run id (link to UI), error category, and rejected-row delta vs prior run per docs. On-call should be able to triage from the alert alone. Configure alert templating in workspace settings; verify rendering before pushing to production. | Pass / FailAi Platformhigh |
| 02 | Sync silently rejects 0.05% of rows each run; over a quarter, 4M rows are missing from the destination. No alert fires. | Configure a rejected-row threshold alert (absolute count or % of run) per sync, tuned to the sync's volume. Per docs, alerts can fire on warning status or rejected-row deltas. Treat slow-bleed rejections as a real failure mode requiring inspection. | Pass / FailAi Platformhigh |
| 03 | Auditor asks for the sync run history of all Salesforce-bound syncs in Q1. Operator finds run history is rolled up at 30 days and detailed run logs are no longer available. | Confirm Hightouch's run history retention contract [REQUIRES-VERIFICATION] and, for compliance use cases, export run metadata to the customer warehouse or a SIEM continuously so a longer audit window is available. Document the export schedule and validate the schema. | Pass / FailAi Platformhigh |
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
- Operations Observability Alerts
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 Operations Observability Alerts eval for Hightouch Hightouch test?+
Evaluates Hightouch's Observability, Run History & Alerts 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 Operations Observability Alerts 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 Operations Observability Alerts 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 Operations Observability Alerts 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.