Docs
  • LogicLoop Documentation
  • LogicLoop Quickstart
  • Data Sources
    • Connect your data source
      • Deployment options
    • Supported data sources
      • API data (JSON)
      • AWS CloudWatch
      • BigQuery
      • Databricks
      • Google Analytics
      • Google Sheets
      • MongoDB
      • Microsoft SQL Server
      • MySQL
      • Oracle
      • PostgreSQL
      • Snowflake
      • Combine Results
      • LogicLoop Data Source
  • Queries
    • Write a rule
      • More rule options
      • A/B testing rules
      • Version history
      • Query Snippets
    • Visualizations & dashboards
      • Visualizations
      • Dashboards
  • Actions
    • Set up an action
    • Action destinations
      • Email
      • Slack
      • Webhooks & APIs
      • Microsoft Teams
      • Salesforce
      • Zapier
      • PagerDuty
      • Write to Database
      • Chain Rules
      • Google Sheets
      • Bento
    • Templating
  • Case Management
    • Case Management
      • Ticket Generation
      • Case Triage
      • Custom Fields and Attachments
  • AI
    • AI Query Helper
    • Ask AI
  • Teams & Settings
    • Integrations
      • Slack
      • Google Sheets
    • Invite your teammate
    • Groups & Permissions
  • Changelog
  • FAQs
  • Troubleshooting
  • Templates
    • Templates Home
    • Risk & Fraud Rules
    • AML Transaction Monitoring Rules
    • Logistics & Marketplace Ops
    • Customer Success & Growth
    • Systems Observability
    • Data Quality Monitoring
    • Healthcare
    • HTML Email Templates
      • Annual Review
      • Weekly Performance Table
      • Invoice Recap
  • BETA
    • AI SQL API
    • Approving rules
    • Render Data as JSON
    • Case Analytics
    • Python
  • Security & Legal
    • Security
    • AI Security
    • Terms of use
    • Privacy policy
    • Services description
Powered by GitBook
On this page

Was this helpful?

  1. Queries
  2. Write a rule

A/B testing rules

PreviousMore rule optionsNextVersion history

Last updated 3 years ago

Was this helpful?

Sometimes, you want to test your rules on different data before pushing them to production.

  1. You want to define cohorts of users or a hold-out population

  2. You want to backtest rules on a historical dataset

  3. You want to preview rules in shadow mode on future data

LogicLoop supports your needs by allowing you to

  1. Assign a label to each row to define a cohort, and using query parameters to select the right groups

  2. Run the same query against different tables in the same database

  3. Run the same query on different databases (this is most commonly used when your dev, QA and production databases are separate)

Here's a video explaining how to test your query with different data