No-Code Workflows

Automations

Complete the Load → Enrich → Write-Back cycle to external systems

What are Automations?

Automations complete the data lifecycle by automatically writing enriched data to external systems. After George AI extracts structured data from your documents, automations push that data to Shopware, webhooks, APIs, or other platforms—without code.

Configure once, then run manually, on schedule, or automatically whenever new data is enriched.

No-Code Setup
5 Min
Average configuration time
Processing
Real-Time
Or scheduled/manual
Connectors
1+ Active
Shopware 6, more coming

How It Works

  • Create Library
    Organize your data
  • Upload/Crawl Files
    Add documents
  • Enrich Fields
    AI extracts data in lists
  • Write Data
    Automations to external systems
Tip: Start with manual execution to test field mapping, then enable scheduling or real-time triggers.

Supported Connectors

Shopware 6
Active

Create or update products from enriched list data

Available Actions:

  • • Product Upsert (create or update)
  • • Field mapping for name, description, price, stock
  • • Product number matching for idempotent updates

Coming Soon
Planned

Future connectors in development

Planned Connectors:

  • • Webhooks (HTTP POST)
  • • REST APIs (custom endpoints)
  • • Google Sheets
  • • CSV Export with automation

Configuration Options

Field Mapping

Map enrichment fields from your List to target system fields:

List Field Maps To Example
Product Name Shopware: name "Premium T-Shirt"
Description Shopware: description "High-quality cotton..."
Price Shopware: price 29.99
Stock Shopware: stock 100
Type Validation: George AI validates data types match (text→text, number→number)
Filters

Control which list items are processed:

  • All Items - Process entire list
  • New Items Only - Skip previously processed items
  • Failed Items - Retry only items that failed last time
  • Custom Query - Advanced filtering (coming soon)
Best Practice: Use "New Items Only" for scheduled runs to avoid duplicate processing
Triggers

Manual

Click "Run Now" button

Best for: Testing, one-time imports

Scheduled

CRON expressions (e.g., daily at 2 AM)

Best for: Batch updates, reports

Real-Time

Execute when enrichment completes

Best for: Live data sync

CRON Examples:
0 2 * * *    # Daily at 2:00 AM
0 */6 * * *  # Every 6 hours
0 0 * * 1    # Every Monday at midnight
Batch Settings
  • Batch Size - Number of items per batch (default: 100)
  • Concurrency - Parallel API calls (default: 5)
  • Error Handling - Continue on error vs stop immediately
  • Retry Policy - Auto-retry failed items (0-3 retries)
Rate Limiting: Reduce concurrency if target API has rate limits

Real-World Examples

E-commerce: PDF Product Catalogs → Shopware

Import product data from supplier PDFs directly into your Shopware store.

Step Action
1. Upload Drop 500 product PDFs into Library
2. Enrich Extract: Product Name, SKU, Price, Description, Stock
3. Automate Map fields to Shopware Product Upsert connector
4. Execute Run automation → 500 products created/updated in Shopware
Time Saved: Manual entry: ~50 hours → Automation: 10 minutes

Content Management: Document Metadata → CMS

Extract metadata from Word docs and sync to your content management system.

  • Upload: 1,000 Word documents from multiple authors
  • Enrich: Extract title, author, keywords, summary, category
  • Automate: Push metadata to CMS via webhook
  • Schedule: Daily at 3 AM to process new documents

Monitoring & Troubleshooting

Batch History
View All Runs
Timestamp, duration, counts
Per-Item Status
Success/Warning/Failed
Click to see error logs
Quick Actions
Stop/Retry/Clear
Manage active batches
Common Errors & Solutions
Error Cause Solution
Authentication Failed Invalid API credentials Re-enter credentials in connector settings
Required Field Missing Enrichment field not populated Check enrichment status, retry extraction
Rate Limit Exceeded Too many API calls Reduce concurrency in batch settings
Product Number Conflict Duplicate SKU in target system Review field mapping, check for duplicates
Retry Mechanisms
  • Automatic Retry - Configure 0-3 retries with exponential backoff
  • Manual Retry - Click "Retry Failed" button to reprocess failed items
  • Selective Retry - Choose specific items to retry from batch history
  • Clear and Re-run - Reset batch status and start fresh

Best Practices

Start Small

Test with 5-10 items before running full batch

Use Filters

Control scope to avoid processing unnecessary items

Monitor First Runs

Watch the first few executions for errors

Off-Peak Scheduling

Run large batches during low-traffic hours

Related Documentation

Learn more about the complete workflow:

George-Cloud