Why PST import still shows up
PST files are a “legacy artifact,” but they remain common in real migrations because they often represent the only complete historical record for specific mail users, shared mailboxes, or decommissioned systems. Microsoft’s supported approach for bulk PST ingestion into Exchange Online is the Import service in the Microsoft Purview portal, using either network upload (AzCopy) or drive shipping. This post focuses on network upload because it’s the most common path in modern projects.
If you’re planning an archival ingestion, it’s important to understand import methods, requirements, and the role PST ingestion plays in compliance, retention, eDiscovery, and archive mailbox strategies.
Primary migration use cases
1) Google Workspace clients where Outlook was the primary interface
In some Google Workspace environments, Outlook was used heavily via IMAP or sync tooling, and users often created local PSTs for long-term storage, “cleanup,” or offline access. When you migrate server-side mail from Google to Microsoft 365, those local PST archives are typically not included—so a follow-on project is needed to ingest PST content into Exchange Online, either into the primary mailbox or (preferably) the archive mailbox.
2) Exchange on‑prem native cutover migration
In Exchange on-prem cutovers, PSTs often surface as a deliberate strategy to reduce mailbox sizes, preserve historical content, or capture mailboxes that won’t be migrated “live.” After Exchange Online is authoritative, Purview Import provides a supportable way to rehydrate that historical content into the correct mailbox or archive.
3) Additional supported scenarios
PST import is also applicable for broader archival and compliance scenarios, including ingesting legacy archives, consolidating distributed PST files, and preparing data for retention, content search, and eDiscovery workflows.
How Purview PST import works (at a high level)
- Create an Import job and generate a SAS upload URL
- Upload PSTs to Azure storage using AzCopy
- Create a CSV mapping file for mailbox targeting
- Validate and analyze uploaded PSTs
- Execute import with optional filtering
- Review reports for completion and errors
Step-by-step: Network upload import
Step 0 — Preconditions
- Decide destination: primary vs archive mailbox
- Enable archive mailboxes if needed
- Plan batching for manageable ingestion runs
Step 1 — Create import job and SAS URL
Create a new import job using the network upload option and generate the SAS URL. Treat this URL securely—it grants temporary upload access into your tenant’s ingestion storage.
Step 2 — Upload PSTs using AzCopy
azcopy copy "D:\PST" "https://<your-sas-url>" --recursive=trueAzCopy is optimized for bulk upload and is the supported method for PST ingestion transfers.
Step 3 — Optional validation with Storage Explorer
Storage Explorer can be used to visually confirm uploaded files and folder structure, but should not replace AzCopy as the primary transfer method.
The mapping file: where imports succeed or fail
The mapping CSV controls how PST content is ingested into Exchange Online, including mailbox targeting, archive placement, and folder structure.
Core fields
- Workload: Exchange
- FilePath: Storage container path
- Name: PST filename
- Mailbox: Target email address
- IsArchive: TRUE or FALSE
- TargetRootFolder: Optional destination folder
Practical recommendations
- Default to archive mailbox for historical data
- Always define a target root folder
- Normalize PST filenames
- Batch import jobs to isolate issues
Operational tips and pitfalls
Duplicate handling
Duplicate behavior depends on PST origin and import target. Re-importing the same PST into the same target folder is typically safe, but overlapping PSTs can introduce duplicates.
Metadata preservation
Message metadata such as timestamps, recipients, and structure is preserved during import.
Filtering during import
Filtering allows organizations to reduce dataset size or align imported data with retention strategy.
Performance considerations
Upload speed is dependent on bandwidth and file size. Large imports should be staged and monitored carefully.
Verification and reporting
- Confirm upload completeness
- Wait for analysis phase
- Execute import
- Review reports for errors and completion