Overview
Genpact Deductions Recovery is an agentic solution that provides capabilities to extract and collect data from different types of documents like invoices and PODs, and process deduction cases. The system collects the data from the ERP systems along with carrier and customer portals. The extraction engine in the system extracts data from other supporting documents such as backup (debit-memo), and Proof of Delivery (POD) files related to each case.
Genpact Deductions Recovery validates the extracted data, and creates a case for processing. The system automates the extraction, validation and processing of deduction cases and provides resolution per line item in the case. Thus, reducing manual processing and error occurrence.
With Genpact Deductions Recovery, the data capture and deduction case processing becomes faster, more accurate, and more secure, allowing the customer to focus on strategic decision-making.
Key features and benefits
Genpact Deductions Recovery offers the following key features and benefits to customers:
- Extraction and collection of accurate data from different sources and formats (backup and POD files).
- An intuitive UI to validate and update incorrect/missing data, and create and process a deduction case.
- Automated resolutions.
- Data insights with analytics dashboard for real-time updates.
Technical specifications
- DocIntel and Open AI for extraction of data.
- Deduction management system for getting the documents, opening a deduction processing case, validating the case documents, processing the case, deciding if the deduction case is invalid or partial valid, and submitting the case for deduction (creating billback).
How Genpact Deductions Recovery works
Genpact Deductions Recovery supports collection of data from different kinds of documents, validating the data, and processing the data for deduction cases.
The system supports multiple file types which are processed differently.
- Structured files are processed through ADF and stored in the database.
- Unstructured files (PDFs), such as POD documents, are extracted and passed through the Eyeball interface, enabling the end user (Analyst) to review and verify the extraction results.
Genpact Deductions Recovery can read and process metadata from ERP, Carrier and Retailer portals, and match the data to the system configuration. The system applies the required business rules to the incoming data before passing the validated data to the deduction portal.
If any file that is mandatory for case processing is missing, the case is flagged as an exception, and the Analyst manually uploads the missing files.
The extracted data is displayed on UI (the portal), enabling the Analyst to perform eyeballing, side-by-side comparison with the source file.
Once all exceptions are cleared and the eyeball is complete, the case moves to a four-way matching process for validation. The system populates the case with a validation result and, if applicable, adds an invalid reason for lines marked as invalid or partially valid. After validation, the Analyst confirms the case status as valid, invalid, or partially valid, in the portal.
The system then applies the following automated workflows:
- Deductions, which are confirmed as valid by the Analyst and for which forms are submitted, are automatically closed.
- Invalid and partially valid deductions are redirected to the billback process. The system automatically generates a billback template email in the portal, and the Analyst sends the email request for billback (payback) after updating the case details in the email body.
The automation streamlines the deduction validation process, reduces manual effort, and minimizes errors, significantly increasing the volume of processed deductions and billback requests.
Following are the components of Genpact Deductions Recovery, and the steps in the process of how Genpact Deductions Recovery works:
| Components | Description |
|---|---|
| Saga | Based on the defined process, the system handles the lines in queue and redirects to the next step. |
| Extraction | The system extracts the data from the DocIntel and OpenAI and translates it to the system language, JSON. |
| Validation | Based on the validation rules defined in YAML files, the system validates the collected data. In case of failed validation, the system throws an error.
|
| Eyeball | Based on the eyeball rules defined in YAML files, the system displays the extraction results in the UI where an Analyst validates these results side by side with the original document (image), and makes correction (add or remove data), if required, before the system processes the data.
|
| Deduction Management | System creates a deduction case after validating the collected data and enables an Analyst to confirm or update the validation result prior to sending a billback request, if it is identified as invalid or partially valid or closing the deduction case, if it is identified as valid. |
