Run Your First Scan

This guide walks you through triggering a scan on an active connector and reviewing the results.

Time required: 5–15 minutes (depending on data volume)

Prerequisites:

• At least one active connector (see connector setup guides)

Step 1: Select a Connector

1. Navigate to Connectors in the Customer Dashboard sidebar.
2. Locate the connector you want to scan — it should show Active status.
3. Click the Scan button on the connector row, or navigate to Scans in the sidebar.

Step 2: Configure the Scan

In the scan configuration dialog:

1. Scan Type: Select Full Scan for the initial run.
   • Full Scan processes all files in the connector scope
   • Incremental Scan (for subsequent runs) only processes new or modified files
2. File Filters (optional): Restrict to specific file types (e.g., .csv, .json).
3. Prefix Filters (optional): Limit to specific paths (e.g., data/exports/).
4. LLM Assist (optional): Enable for AI-powered false positive reduction on borderline findings.
5. Click Start Scan.

Step 3: Monitor Progress

The Scan Monitor displays real-time progress:

• Files Total — Total files discovered in scope
• Files Processed — Files completed by workers
• Findings — Sensitive data matches detected so far
• Workers — Number of active parallel workers
• Elapsed Time — How long the scan has been running
• Estimated Remaining — Projected time to completion

Step 4: Review Findings

When the scan completes (status changes to Completed):

1. Navigate to Data Catalog to see the full inventory of files with findings.
2. Sort by Risk Score to prioritize the highest-risk files.
3. Click any file to view its findings:
   • PII Category — What type of sensitive data was detected (SSN, email, etc.)
   • Confidence — How certain the detection is (0.0 to 1.0)
   • Classifier — Which detection method identified it (regex, ML, proximity)
   • Location — Where in the file the data was found

Step 5: Take Action

Based on your findings, you can:

• Investigate — Navigate to the Investigation page for deeper analysis by category or severity
• Create Policies — Set up governance rules to automatically handle future findings
• Schedule Scans — Configure recurring scans to maintain ongoing visibility
• Export Results — Download findings as CSV or JSON for external reporting

What Happened During the Scan

The scan executed the following pipeline for each file:

1. Pre-Screen — A probabilistic check determined whether the file was likely to contain sensitive data
2. Download — File content was streamed from cloud storage
3. Parse — File format was detected and content extracted
4. Classify — All active classifiers ran against the content
5. Score — Each match received a confidence score
6. AI Disambiguation — (If enabled) Findings in the ambiguous range were escalated to AI for adjudication
7. Store — Final findings were written to the Data Catalog

Troubleshooting

Next Steps

• Create custom classifiers for organization-specific data patterns
• Set up governance policies to automate remediation
• Configure event-driven scanning for real-time detection