Home / Blog / Gearset

Getting started with Gearset CI/CD for Salesforce

By Sikandar Pathan2 min read

If your team still ships Salesforce changes with change sets, every release is a manual, error-prone event. Gearset turns that into a repeatable pipeline. I've rolled it out across cross-functional teams, and this is the order I'd do it in again.

1. Put your metadata in version control

Pick one repository (Azure DevOps, GitHub or Bitbucket) as the source of truth. Every change starts as a branch and ends as a pull request. This one habit makes everything else possible.

2. Start with comparisons and manual deployments

Before automating anything, let the team use Gearset to compare orgs and deploy by hand. They learn to read the differences, spot missing dependencies and trust the tool.

3. Add CI jobs per environment

Set up continuous integration jobs that validate or deploy from a branch to its matching org: integration, UAT, then production. Validation-only jobs on pull requests catch broken deployments before they merge.

4. Keep sandboxes in sync

Environments drift. A regular back-sync routine that pushes production changes down to lower sandboxes keeps everyone testing against reality.

5. Add quality gates

  • Static code analysis (like PMD rules) on every pull request.
  • Required Apex tests for production deployments.
  • Pre- and post-deployment steps written down in a runbook.

    Tools don't create DevOps culture. Clear standards, small changes and fast feedback do.

The full walkthrough, step by step with screenshots, is coming in the Gearset Course.

Sikandar Pathan

Senior Salesforce Platform Engineer writing about Salesforce DevOps, Gearset and AI at Kaitrel. More about me.

← Your first Python script to automate a boring taskHow I built an AI assistant for Salesforce with Claude and MCP →

Keep reading

More from the blog

All posts

Newsletter

Prefer email?

Get new posts on code and automation in your inbox. One email when something new is published. No spam, unsubscribe anytime.