Skip to content

CI/CD Getting Started

Automate data validation in your development workflow. Catch data issues before they reach production with continuous integration and delivery built specifically for dbt projects.

What you'll achieve

Set up automated workflows that:

  • Maintain current baselines - Auto-update comparison baselines on every merge to main
  • Validate every PR/MR - Run data validation checks automatically when changes are proposed
  • Prevent regressions - Catch data quality issues before they reach production
  • Save team time - Eliminate manual validation steps for every change

Note

CI/CD automation requires Recce Cloud Team plan. A free trial is available.

Understanding CI vs CD

Recce uses both continuous integration and continuous delivery to automate data validation:

Continuous Integration (CI)

  • When: Runs on every PR/MR update
  • Purpose: Validates proposed changes against baseline
  • Benefit: Catches issues before merge, with results in your PR/MR

Continuous Delivery (CD)

  • When: Runs after merge to main branch
  • Purpose: Updates your baseline Recce session with latest production state
  • Benefit: Ensures future comparisons use current baseline

Choose your platform

Recce integrates with both GitHub Actions and GitLab CI/CD.

Select your Git platform to get started:

GitHub

If your dbt project uses GitHub:

  1. Setup CI - Auto-validate changes in every PR
  2. Setup CD - Auto-update baseline on merge to main

GitLab

If your dbt project uses GitLab:

  1. Setup CI - Auto-validate changes in every MR
  2. Setup CD - Auto-update baseline on merge to main
  3. GitLab Personal Access Token Guide - Required for GitLab integration

Prerequisites

Before setting up, ensure you have:

  • Recce Cloud account with Team plan or free trial
  • Repository connected to Recce Cloud (setup guide)
  • dbt artifacts (manifest.json and catalog.json) from your project

Architecture overview

Both CI and CD workflows follow the same pattern:

  1. Trigger event (merge to main, or PR/MR opened/updated)
  2. Generate dbt artifacts (dbt docs generate or external source)
  3. Upload to Recce Cloud (automatic via workflow action)
  4. Validation results appear in Recce dashboard and PR/MR

Recce CI/CD architecture

Automated validation workflow for pull requests

Next steps

  1. Choose your platform (GitHub or GitLab)
  2. Start with CD setup to establish baseline updates
  3. Add CI setup to enable PR/MR validation
  4. Review best practices for environment preparation

After setting up CI/CD automation, explore these workflow guides: