SitecoreAI Pathway Guide: End-to-End Migration Walkthrough

Content migration is one of the most complex phases of any Sitecore modernization effort. SitecoreAI Pathway introduces AI-assisted tooling that helps analyze legacy XM/XP implementations and rebuild content within SitecoreAI. While the platform provides intelligent template and component mapping, successful migrations still require careful preparation, validation, and execution.

This guide provides a step-by-step walkthrough of the complete Pathway migration process, including target and source extraction, AI-assisted grouping and mapping, and final migration execution. It focuses on the practical implementation details, file preparation, and validation steps developers need to successfully migrate content.

For a conceptual overview of Pathway’s architecture and value, refer to this article.

Step 1: Migration Creation and Configuration

Before starting a Pathway migration, you must ensure you have met all prerequisite requirements.

Prerequisites

Source is Sitecore XM/XP
Sitecore AI target site has been created
Templates, renderings, page designs, and partial designs have all been created in the target
Pathway App has been installed (marketplace app available to Sitecore 360 subscribers)

Once these are met, you can begin your migration. Open the Pathway app in SitecoreAI and click on “New Migration.” On the next screen, add the name and description for the migration. These should clearly identify the site and language you are migrating. Click “Next” to proceed.

On the next screen, enter the target environment and site details in the first column.

Pathway migration setup screen showing target environment and site details

Click “Initialize Migration.” A success message should appear and an “Upload URL” will be generated in the middle column. This will be used later during Source Data Extraction.

Step 2: Target Data Extraction

Navigate to the Pathway Dashboard and click on “Show Info” at the top right of the page. Scroll down until you see two download buttons. Download the zip package next to “Analyze SitecoreAI project.” After extracting it, you will see the following files:

Extracted files from the Analyze SitecoreAI project package

Follow the included instructions. The exporter looks for GUID-based folders with the standard serialization structure: {GUID}/en/<version>/xml.

To achieve this structure, create Sitecore packages containing:

Templates (page and component)
Renderings
Page designs and partial designs

After creating the packages, extract them and organize the files according to the instructions. Ensure the extracted files are placed in the same directory as the export-structure.ps1 script.

Components

Extract the serialization package zip file and locate the items folder
Create a new Components folder in the same directory as the export script
Copy the items folder into the new Components folder

Page and Partial Designs

Extract the serialization package zip file
Locate items/master/sitecore/content/<project>/<site>/Presentation
Create a Presentation folder beside the export script
Copy the extracted Presentation folder contents into the new folder

Renderings

Extract the serialization package zip file
Locate items/master/sitecore/layout/Renderings
Create a Renderings folder beside the export script
Copy the entire contents into the new Renderings folder

After preparing the data, execute the export-structure.ps1 script. The console will display logs identifying successes and failures as the script runs.

PowerShell console output from export-structure script

The generated xmc-structure.json file contains several top-level arrays describing pages, renderings, page design mappings, and page design structures.

Pages: One entry per page template with simplified field metadata
Renderings: Rendering definitions with datasource template and parameters
PageDesignMappings: Pairs of templateId and pageDesignId
PageDesigns: Page designs with partial designs and parsed renderings

The JSON output is also written to the console. Because it may be extremely large, you may want to remove the console output section from the script.

Script line showing JSON output that can be removed from console logging

Once validated, upload the xmc-structure.json file back into Pathway. A successful upload will display a new status labeled Uploaded – SITECOREAI STRUCTURE.

Uploaded SitecoreAI structure status in Pathway

Step 3: Source Data Extraction

Return to the Pathway Dashboard and download the package labeled “Scan your XM instance.” After extracting the files, you will see the following:

Extracted files from the Scan your XM instance package

Add ExtractorHandler.ashx to your source CM server under wwwroot/sitecore/admin.

ExtractorHandler file placed in the Sitecore admin directory

To verify installation, visit:

https://your-sitecoreinstance/sitecore/admin/ExtractorHandler.ashx?action=getallsitenames

You should receive a list of sites from your Sitecore instance.

List of Sitecore sites returned by ExtractorHandler

Next, update the setting Sitecore.Services.SecurityPolicy in:

.\App_Config\Sitecore\Services.Client\Sitecore.Services.Client.config

Set the value to:

Sitecore.Services.Infrastructure.Web.Http.Security.ServicesOnPolicy

This grants access to Entity and Item Services and can be reverted after extraction.

Developer Tip

To inspect extractor JSON output, add a custom log function in ExtractorHandler.ashx and log the pageObject at the end of the ParseHTML function.

Step 4: Page Grouping

After both source and target data have been extracted, Pathway can begin AI analysis. Click the Grouping button to allow Pathway to automatically group pages from the source site.

Once grouping is complete, select Template Mapping to match page groups with target page templates. You can review the mapping details to see the reasoning behind the AI decisions.

Pathway page grouping and template mapping interface

Assigned template from the target site
Template item ID
AI reasoning behind the mapping decision

Step 5: Component Mapping

Click Component Mapping to map components from the source site to component templates in the target environment.

Target placeholder
Target component
AI explanation for the mapping

Step 6: Migration Execution

Click the Migration button to begin the migration process. Once completed, Pathway will display successfully migrated items along with any failures.

Key Takeaways and Considerations

Based on hands-on implementation and testing, SitecoreAI Pathway significantly improves migration workflow structure and speed. However, some limitations and operational considerations remain.

Strengths

Significantly faster migrations compared to traditional manual approaches
Structured workflow ensures consistency and repeatability
AI-assisted grouping and mapping reduce manual alignment work

Current Limitations

Inherited fields are not parsed during target export, which can affect component mappings
Limited transparency into AI decision-making processes
No ability to manually override AI-generated mappings

Operational Gaps

Troubleshooting failed migrations
Supplementing migrations with custom scripts
Validating migrated data and content
Post-migration stabilization and cleanup

Areas for Future Improvement

Manual editing of AI-generated mappings
Improved logging and diagnostics
Greater transparency in AI decision-making

Conclusion

SitecoreAI Pathway represents a major step toward modernizing Sitecore content migrations. By introducing a structured, AI-assisted workflow, it transforms migration from a highly manual and unpredictable process into a more repeatable and scalable one.

Although the platform continues to evolve, it already delivers substantial value by accelerating migration timelines and reducing manual mapping effort. As visibility, control, and customization capabilities improve, Pathway is positioned to become a core tool for Sitecore modernization initiatives.