Skip to main content
Growth Method users can create new versions of campaigns while maintaining a complete version history and lineage of all previous versions. The versioning system allows you to iterate on experiments while maintaining a complete history of all versions. When you create a new version of an experiment, the system tracks the entire version lineage, making it easy to compare approaches, learn from previous attempts, and refine your growth experiments over time.

​
How versioning works

Auto-numbering: Versions automatically increment based on family size
  1. The version number is automatically incremented
  2. The campaign name is preserved
  3. The campaign hypothesis is preserved
  4. The user that creates the new version becomes the new campaign owner
  5. A version display (v2, v3 etc) is added to the campaign title
  6. A new campaign completion target data is set (typically today’s date + team duration target)
All other fields (design, results and campaign dates) are set to blank/null ready for the new campaign.

​
The versioning tooltip

Hovering over a version number on a campaign displays a tooltip for easy switching between campaign versions. The tooltip includes:
  • Version number
  • Campaign name
  • Avatar of the campaign owner
  • Links to the campaign version

​
Version Numbering

  • The original experiment is considered version 1 (v1)
  • Each copy creates a new version with an incremented number (v2, v3, v4, etc.)
  • All versions in a family share the same master ID (the ID of the original v1 experiment)
  • Version numbers increment globally across the entire version family, regardless of which version you copy from

​
Master Experiment

The master experiment (v1) is the original experiment that serves as the root of the version tree. All subsequent versions reference this master, creating a lineage that connects all related experiments. All versions (v2, v3, v4) reference v1 as their master, regardless of which version they were copied from.

​
Creating a New Version

​
Method 1: From the Actions Menu

  1. Open an experiment you want to iterate on
  2. Click the three-dot menu (⋮) in the top-right corner
  3. Select “New version”
  4. A new version will be created and you’ll be redirected to edit it

​
Method 2: After Completing an Experiment

When you complete an experiment, you’ll see a modal with two options:
  • “Refine this idea” - Creates a new version based on the completed experiment
  • “Create new idea” - Creates a brand new idea (not a version)

​
Method 3: From the Experiment Table

Click the actions menu on any experiment row and select “New version”.

​
What Happens When You Create a Version

​
Fields That Are Preserved

  • Name - Keeps the same experiment name
  • Description - Smart handling (see below)
  • Team - Remains in the same team
  • Original Submitter - The submitted_by_id from v1 is preserved across all versions
  • Revision History - The new experiment ID is appended to the revisions array

​
Smart Description Handling

The description field has intelligent behavior based on the experiment stage:
  • If the original is complete: The description becomes the test results from the original
  • If the original is incomplete: The description is copied as-is
This ensures you’re building on learnings when iterating on completed experiments.

​
Fields That Are Reset

  • Stage - Reset to “In Design”
  • Test Dates - All date fields cleared (date_test_indesign, date_test_complete, date_test_inprogress, date_test_analysing)
  • Test Results - Cleared (results move to description if original was complete)
  • Starred - Reset to unstarred (0)
  • Is Template - Reset to non-template (0)

​
Fields That Are Updated

  • Experiment Owner - Set to the user who created the version
  • Completion Target - Set to today + your team’s experiment duration target
  • Version Number - Automatically incremented based on the version family size

​
Viewing Version History

​
Version Badge

Experiments with versions display a version badge showing their version number:
  • v2, v3, v4, etc. (v1 experiments don’t show a badge by default)

​
Version Bubble & Tooltip

Click or hover over the version badge to see the complete version history: The version dropdown shows:
  • All versions in the family
  • Version number (v1, v2, v3, etc.)
  • Experiment owner’s avatar
  • Experiment name
  • “Current” badge on the version you’re viewing
  • “[Deleted]” indicator for soft-deleted experiments
Simply click on any version in the dropdown to navigate to that version. The link uses wire:navigate for fast client-side navigation.

​
Deleted Versions Remain Visible

When a version is deleted, it’s soft-deleted rather than permanently removed. This means:
  • The version still appears in the version history dropdown
  • It’s marked with “[Deleted]” indicator
  • It cannot be accessed or restored (no clickable link)
  • The avatar and version number remain visible

​
Why Keep Deleted Versions?

Maintaining deleted versions in the history:
  • Prevents version number gaps (v1, v2, [deleted v3], v4)
  • Provides context about the experiment’s evolution
  • Maintains data integrity for reporting and analytics
  • Shows the complete story of experimentation attempts

​
Search & Reporting

​
Search Integration

Versions are included in the search index:
  • The version number is searchable (search for “v2” to find all version 2 experiments)
  • Search results display the version badge
  • Original submitter is preserved for proper attribution

​
Team Reports

Team reports include version information:
  • Version badges appear in report tables
  • Completed experiments can be filtered to include/exclude versions
  • Attribution goes to both the original submitter and current experiment owner

​
Best Practices

​
When to Create a New Version

  • ✅ You want to iterate on a completed experiment with new learnings
  • ✅ You’re testing a variation of an existing experiment
  • ✅ You want to maintain a clear lineage between related experiments
  • ✅ The core hypothesis is similar but the approach has changed

​
When to Create a New Experiment

  • ❌ The hypothesis is fundamentally different
  • ❌ You’re testing a completely different growth lever
  • ❌ The experiment targets a different user segment or product area
  • ❌ There’s no meaningful connection to the original experiment

​
Naming Conventions

Since version numbers are automatically tracked, you can:
  • Keep the same name across versions for easy grouping
  • Add descriptive suffixes if variations are significant (“Hero CTA Test - Mobile Focus”)
  • Use the description field to explain what changed between versions

​
Leveraging Descriptions

When iterating on complete experiments:
  • Review the description (which contains the previous results)
  • Build your new hypothesis on top of previous learnings
  • Update the description to explain what you’re changing and why