Skip to content

Data Focus Release Notes

Software Version: 3.5.0

Publication Date: July 2026

What's New

  • Introduced Unstructured Data Discovery, a new agentless module for detecting, classifying, and reporting personal and sensitive data across file systems and cloud storage. Highlights include:

    • Broad data source support: SMB/CIFS, SFTP, Amazon S3 and S3-compatible storage, Azure Blob Storage, ADLS Gen2, Google Cloud Storage, Google Drive, SharePoint, Dropbox, Box, and Gmail.
    • Advanced file analysis across PDF, Office, image, e-mail, and archive formats, with OCR support for scanned documents.
    • Smart scan management with incremental scanning, scheduling, blackout time, sampling, scan groups, and QoS profiles.
    • Detailed scan tracking, including progress, processed/failed files, folder breakdowns, DLQ records, and system logs.
    • Document categorization and risk scoring across business categories such as Financial, HR, Legal, and Medical.
    • Finding search and an in-app file viewer, so files can be reviewed without downloading them.
    • A review workflow for marking findings as sensitive, safe, or false positive.
    • Masked copy generation, producing a secure PDF copy with sensitive data masked, without touching the original file.
    • Similarity and duplicate file detection based on content, not just file name or metadata.
    • Metadata tagging that writes classification results into PDF, Office, and file-system metadata.
    • Scan and compliance reporting, plus version comparison between scans.
    • Secure, zero-trace scanning: source files are accessed read-only and are never modified. Sensitive content is processed entirely in memory, in a streaming, constant-memory fashion, without being written to disk — so files of any size can be scanned safely and without leaving a trace.
    • Resilient, scalable scan infrastructure: scan jobs are distributed across multiple worker processes; if a worker stops unexpectedly, its task is automatically taken over so the scan completes without data loss, and capacity can be scaled horizontally as data volume grows.
    • Resilience against malicious content: service-exhausting content such as nested or highly compressed (zip-bomb style) archives is automatically detected and safely contained — such files are flagged as failed while the scanning service keeps running without interruption.
    • Security and traceability: masked data storage, encrypted credential management, SSO, mTLS, and audit log support provide a secure and traceable discovery process. Sensitive data is never written to disk and is processed in memory, source files are accessed read-only, and all sensitive operations — including PII reveal and content access — are recorded in the audit log.

    See Unstructured Data Discovery for the full user guide.

  • Added view information display in Metadata Explorer results — discovered database views are now clearly marked as views in the interface, allowing users to distinguish tables from views directly.

  • Standardized the Qualified Name structure across all data assets (Datasource, Catalog, Schema, Table, and Column). The datasource name is now included as the starting point of every Schema, Table, and Column identifier, so objects with the same name can be safely distinguished across different datasources. Object names containing a dot (.) character are now handled correctly, and every screen and process that relies on Qualified Name — search, metadata display, discovery results, and asset relationships — has been updated to use the new structure.

  • Discovered database objects are now grouped and displayed separately by entity type — Tables, Views, and Stored Procedures — making it easier to review and manage different object types.

  • Added Metadata Checkpoint support: if a metadata scan is unexpectedly interrupted or restarted, it now resumes from where it left off instead of repeating already-completed steps, reducing overall scan time after an interruption.

  • Added a Track Database Changes checkbox to the data source creation screen. When a database change is detected, metadata is automatically re-scanned, and a task is assigned to the user asking whether a discovery scan should be started, with Seen and Start Scan actions. This flow works with information coming from DAM and DataTouch.

  • Added a Transfer Governance to New Catalog option to the Metadata Explorer scan creation screen. When enabled, Classification, Term, Label, Owner (user and group), and Domain information from the existing catalog is automatically transferred to the newly generated catalog during or after the scan — automating a capability that was previously available only from the Assets Transfer page.

  • Added AI-suggested descriptions for Classifications — users only need to enter a title, and the description is automatically generated by AI.

  • Added an AI Review badge/flag for columns that have been reviewed and approved by AI, making it visible in the interface which columns went through AI-assisted review.

  • Added a Review with AI button to the Discovery Column Filter tab. Users can optionally include SUSPECT columns (Include SUSPECT columns) and click Start Analysis to get AI suggestions, shown under Suggested and Uncertain tabs; suggestions can be applied with Apply selected or Apply all, or re-run with Re-analyze.

  • Added Risk Labels, used to assign a risk level to classifications. Risk labels are managed from Labelling > Risk Labels, where users can view, create, edit, and delete labels. A risk label can be selected — or created on the fly — from the Risk Level field when creating or editing a Classification, and is displayed as a colored badge in both list and hierarchical classification views.

  • Added Risk Score calculation for structured data assets, based on the classifications detected during discovery. Each classification's hit ratio is multiplied by its label weight to produce a score between 0–100, which can be used for reporting, prioritization, and risk analysis.

  • Added Structured Search and Advanced Search screens, allowing users to search and filter across all discovery results without first selecting an asset.

  • Added a Risk Label filter to Advanced Search, which can be combined with other advanced search filters to narrow results by assigned risk label.

Enhancements

  • Added the ability to export Monitoring and Audit Log records for Discovery View Log.

  • Added suggestions in the schema/table Include/Exclude fields based on previously performed metadata scans.

  • Added a Sample Text and Test feature to the Rule screen, allowing users to instantly test their rules against sample text.

  • Reorganized the sidebar into two new menu groups, Structured Data and Unstructured Data, grouping related pages accordingly.

Integrations

  • Added the Identigro integration, allowing users to perform user management operations directly from Data Focus without accessing the Keycloak screen.

  • Added the Integro integration, which notifies users about system operations through their preferred format and channel (e.g., Slack, e-mail). Related events are created from the Integration page; an event and its associated templates must be created before an integration can be set up.