Intelligent document processing solutions - feature comparison matrix

Review the capabilities available in Automation 360 IQ Bot and Document Automation to understand the feature parity.

Automation 360 IQ Bot Document Automation
Administration and setup
Installation Separate installer Installed as part of the Control Room
Database administration Separate database; requires additional effort for backups and maintenance Uses the Control Room database; no additional backup or maintenance needed
Server configuration Requires additional SSL configuration, RabbitMQ administration, HA and DR Works out-of-the-box
Learning instance configuration
Alias management (keywords) Yes, parity Users can view, add, or remove field aliases for the learning instance
Confidence threshold Does not work out-of-the-box; server-level configuration Helps to catch false positive outcomes; field-level configuration
Field order and field name customization Not supported Helps to improve validation efficiency
Field configuration Must be done for every group Needs to be done only one time for a learning instance
Ability to add and remove fields from the learning instance No Yes
Search for form or table fields No In create/edit learning instance search by field name, label, or data type.
Extraction capabilities
Pre-trained models Pre-configured domains that require additional training Yes, out-of-the-box
Handling document variations Classifier relies on document layout and training. Documents uploaded with a new layout are sent to a new document group, which requires training. Some docs could not be classified. Layouts are identified using an object detection model to find key-value pairs. It is not necessary to configure all the fields in a learning instance before uploading documents.
Training process Each group needs to be manually trained before it can process documents Learning instance extraction improves based on changes users make in the Validator
Output format CSV only CSV or JSON
Process invoice-like documents with user-defined document type Custom domains can be created exclusively through APIs for processing custom document types Supported out-of-the-box
Regular expression-based extraction Not supported Use regex pattern extraction for values that follow a specific pattern, such as a VAT number or a PO number
Standardize the format of extracted numbers and dates No Configure your learning instances to convert extracted dates and numbers to standard formats
Unstructured document support No Yes, available with GenAI integration
Extraction from Fixed Forms No Yes, supported via Microsoft Standard Forms
Google Document AI
Google Document AI support No
  • Pre-trained models for invoices, Utility bills, Receipts
  • Google CDE support
  • Available through BYOL and OEM licenses
Human review and Validation
Validation queue management No Use filters and search to find specific learning instances or documents
Auto-fill for tables No Use the Auto fill button in the validation interface to expedite the extraction of values from tables
Document-level Validation rules No Document Automation enables you create complex validation rules with different fields, multiple conditions and actions
End-to-end automation
Task bots Must be manually created Document Automation automatically creates Task Bots for a learning instance
See status and history of uploaded documents No Yes
Customizable document processing workflow No Yes
Ability to split documents No Supported by the Advanced Classifier package
Monitoring and reporting
Custom dashboards No Available with Bot Insight integration
Licensing No License consumption is tracked properly
Security
Retention policy No Supported for 90 days by default
Secret management No Can use Credential Vault
Audit log No Yes