DATEV Integration

The DATEV Integration is the central link between ELO document management and DATEV financial accounting. It consists of several components that can be activated individually or together:

Supported Providers

1. Set Up Connection

1. Navigate to Integrations → DATEV → Configuration
2. Enable the integration using the master switch
3. Select the provider: Cloud or On-Premise
4. Cloud: Enter OAuth Client ID, Client Secret, and Scopes. Select Sandbox or Production.
5. On-Premise: Enter Host, Port, and Windows credentials (DOMAIN\user). Enable TLS if needed.
6. Verify the connection via Connection Status

2. Enable Shadow-DB

1. Enable Features.ShadowDB.Enabled
2. Optional: Restrict synchronization to specific clients (ClientIDs)
3. Enable AutoSyncEnabled for automatic background synchronization
4. Run an initial manual sync via Master Data

3. Configure Booking Proposals

1. Enable Features.BookingProposal.Enabled
2. Create a new job under Booking Proposals
3. Configure search criteria (ELO mask, ObjKey filters, invoice type)
4. Adjust field mapping (which ELO ObjKeys map to which DATEV fields)
5. Define the processed marker to prevent duplicate processing
6. Choose the execution mode: manual, scheduled (Cron), or continuous

4. Classification & Posting

1. Enable the desired recognition features under Classification
2. Adjust field weights (e.g., weight IBAN higher than city name)
3. Enable the learning-based posting system under Auto-Posting
4. Test the results using Test Classification

5. Optional: Additional Features

• Payment Tracking: Enable duplicate detection and ELO marking in the configuration
• Dunning: Configure recipients and channels for dunning level notifications
• Web Enrichment: Add an API key for SerpAPI or Bing
• Learning Data: Review and manage learned posting patterns under Learning Data

Shadow-DB & Master Data

• Regular synchronization – Enable auto-sync (e.g., every 15 minutes) to keep creditors, debtors, and accounts up to date. Without current master data, classification accuracy drops significantly.
• Selective client filtering – Restrict ClientIDs to the clients you actually use. This reduces sync duration and storage consumption significantly for large DATEV installations.
• Full sync interval – Set FullSyncInterval to 24h. Delta syncs are fast, but a daily full sync ensures consistency.

Booking Proposals

• Validate before submission – Enable ValidateFirst in the job configuration to catch invalid bookings before they are transferred.
• Use processed markers – The ProcessedMarker prevents duplicate processing. Ensure the ObjKey field exists and is not written to by other processes.
• Test before production – Use the Test Classification tool to validate classification and posting suggestions on sample invoices before enabling AutoSend.

Security

• Do not disable TLS – Only disable SkipCertVerify temporarily for debugging. In production, TLS must be active.
• Rotate credentials – Change OAuth client secrets and Windows passwords regularly according to your security policy.

Performance

• Adjust parallelism – MaxParallelClients (default: 6) can be increased for many clients, but puts more load on the DATEV server.
• Optimize page size – PageSize (default: 500) is a good trade-off. Values above 2000 may cause timeouts.

Workflow: Incoming Invoice → DATEV

1. Invoice receipt – An incoming invoice is stored in ELO via email import, scan, or manual upload.
2. Document classification – The Document Classifier analyzes the document via OCR and identifies the creditor using Shadow-DB master data (name, IBAN, VAT ID, match code).
3. Auto-posting – Based on learning data and rules, the GL account, cost center, and tax key are suggested. With sufficient confidence, the fields are filled automatically.
4. Create booking proposal – A booking proposal job collects classified documents, creates DATEV-compliant booking entries, and validates them against master data.
5. Review & approval – In the Admin UI, proposals can be reviewed and corrected before transfer. With AutoSend=false, a manual approval step is required.
6. DATEV transfer – Validated proposals are submitted to DATEVconnect via API. The status is logged, and the document is marked as exported in ELO.
7. Payment reconciliation – After payment, the status is detected from open items and optionally updated on the document in ELO via ELO-Marking.

Workflow: Dunning Notifications

1. Shadow-DB syncs open items including dunning level changes
2. System detects new/increased dunning levels since last sync
3. Notification is sent via email, Teams, or Slack to configured recipients

Workflow: Vendor Master Data Enrichment

1. New vendor is detected during classification but not found in master data
2. Web enrichment searches public sources for company data (address, VAT ID, IBAN)
3. Found data is provided as a suggestion for master data completion