Batch Certification

Batch Certification enables organizations to automatically certify multiple documents simultaneously, ideal for processing generated content at scale. This guide covers batch submission, monitoring, and result retrieval.

Batch Overview

Batch certification is optimized for:

• Marketing campaigns (certifying 50+ product guides simultaneously)
• Quarterly content refreshes (re-certifying existing materials)
• API-driven workflows (bulk processing from automated systems)
• Compliance reporting (certifying entire document sets for audits)

Batch Limits:

• Up to 500 documents per batch
• 10 MB total batch size
• 60-minute processing window

Creating a Batch

Via Dashboard

1. Navigate to Certification → New Batch
2. Click Add Documents → Upload files or paste content
3. Configure batch settings:
   • Batch Name: Identifier for tracking (e.g., “Q1-2024-Marketing”)
   • Auto-Revoke On Change: Automatically revoke certificates if facts drift
   • Expiry Duration: 30, 60, 90 days, or 1 year
4. Click Start Batch Processing

Via API

from truthvouch import TruthVouch

client = TruthVouch(api_key="your-api-key")

# Prepare documents
documents = [
    {
        "id": "doc_001",
        "content": "Product X specification sheet...",
        "title": "Product X Specs",
        "tags": ["product", "marketing"]
    },
    {
        "id": "doc_002",
        "content": "Product Y pricing guide...",
        "title": "Product Y Pricing",
        "tags": ["pricing"]
    },
]

# Submit batch
batch = client.certification.batch_submit(
    documents=documents,
    batch_name="q1-2024-certification",
    auto_revoke_on_drift=True,
    expiry_days=90
)

print(f"Batch ID: {batch.id}")
print(f"Status: {batch.status}")
print(f"Documents: {batch.document_count}")

Monitoring Progress

Automatic Batch Status

Monitor batch processing status automatically:

batch = client.certification.get_batch(batch_id="batch-123")

print(f"Status: {batch.status}")  # queued, processing, completed, failed
print(f"Progress: {batch.processed_count}/{batch.document_count}")
print(f"Estimated Time: {batch.estimated_time_remaining}s")

Status Stages:

• Queued: Waiting to start processing
• Processing: Documents being verified
• Completed: All documents processed (may have individual failures)
• Failed: Entire batch failed (unusual)

Dashboard Monitoring

1. Navigate to Certification → Batches
2. Click batch name to view details:
   • Real-time progress bar
   • List of documents with individual status
   • Estimated completion time
3. View results as they complete (streaming)

Handling Results

Retrieve Batch Results

Once processing completes:

# Get batch with results
batch = client.certification.get_batch(batch_id="batch-123")

# Iterate through certificates
for cert in batch.certificates:
    print(f"Document: {cert.document_id}")
    print(f"Certificate ID: {cert.id}")
    print(f"Trust Score: {cert.trust_score}/100")
    print(f"Status: {cert.status}")  # issued, failed, requires_review
    if cert.status == "failed":
        print(f"Error: {cert.error_message}")

Export Results

Export batch results in multiple formats:

CSV Export:

csv_data = client.certification.export_batch_csv(batch_id="batch-123")
with open("batch-results.csv", "w") as f:
    f.write(csv_data)

JSON Export:

json_data = client.certification.export_batch_json(batch_id="batch-123")

Spreadsheet Export:

• Dashboard: Click Export → Excel
• Auto-downloads with score, claim count, status, certificate ID

Partial Failures

If some documents fail while others succeed:

In Dashboard:

• Failed documents shown in red
• Review error messages for each
• Optionally re-submit failed documents as new batch

Via API:

results = client.certification.get_batch_results(batch_id="batch-123")

failed = [r for r in results if r.status == "failed"]
if failed:
    # Re-submit failures
    retry_docs = [
        {"id": f.document_id, "content": f.original_content}
        for f in failed
    ]
    retry_batch = client.certification.batch_submit(
        documents=retry_docs,
        batch_name=f"retry-{batch_id}"
    )

Common Batch Workflows

Pre-Launch Campaign Certification

# Marketing prepares 50 documents for new product launch
batch = client.certification.batch_submit(
    documents=marketing_docs,
    batch_name="launch-2024-q2",
    auto_revoke_on_drift=True,
    expiry_days=90
)

# Wait for completion
import time
while batch.status != "completed":
    time.sleep(10)
    batch = client.certification.get_batch(batch_id=batch.id)

# Check all passed
passing = [c for c in batch.certificates if c.trust_score >= 75]
if len(passing) == len(batch.certificates):
    print("All content certified, safe to launch")
else:
    print(f"Review {len(batch.certificates) - len(passing)} documents")

Quarterly Content Refresh

# Re-verify all content to catch any drift
existing_docs = client.certification.get_certificates_by_tag("active")

batch_docs = [
    {
        "id": cert.document_id,
        "content": cert.original_content,  # content stored in certificate
        "tags": cert.tags
    }
    for cert in existing_docs
]

refresh_batch = client.certification.batch_submit(
    documents=batch_docs,
    batch_name="q1-2024-refresh"
)

# Review new scores for drift
results = client.certification.get_batch_results(refresh_batch.id)
drifted = [r for r in results if r.drift_detected]
print(f"Found {len(drifted)} documents with drift")

Batch Notifications

Configure notifications for batch events:

Dashboard Settings:

• Navigate to Settings → Notifications
• Enable “Batch completion alerts”
• Choose: email, Slack, or both

API Webhooks:

client.webhooks.subscribe(
    event="certification.batch.completed",
    url="https://example.com/hooks/batch-complete",
    batch_name_pattern="q*-*"  # optional filter
)

Next Steps

• Submitting: Learn about single vs batch submission
• Results: Understand and act on batch results
• Automation: Build batch processing into your pipeline