A large, multinational retail company operates in a variety of different channels, from physical stores to ecommerce, spread across the world. Every day, the group receives thousands of invoices for goods being distributed in their supermarkets, groceries, drugstores and other channels.
The challenge
While most of these invoices are received via Electronic Data Interchange (EDI), which is a structured format and therefore is automatically processed, close to 100,000 each year are received in paper or PDF formats. Processing paper or PDF invoices is very time-consuming, especially when these documents include a long list of items and lines for large categories of products being ordered. All information, including the line-items, needs to be booked manually.
How can they automatically extract the information from these paper and PDF invoices in order to achieve considerable time and cost savings?
The requirements: fit in IT landscape and prevent fragmented IT landscape
One of the first challenges or requirements is that any new technology perfectly fits in their existing IT landscape. This means that the solution should be able to cross-check data with other sources of master data and export to multiple back-end systems. Also, the solution should be ready for connecting to new systems such as SAP in the future.
Another challenge: they preferred a solution from one of their current software vendors to prevent a fragmented IT landscape with dozens of different software vendors.
The solution
Faced with this brief, we came up with three possible scenarios as a growth path. It’s a best practice to implement this type of solution as an iterative project, so we can adapt to the different challenges we come across. The main installation and initial setup is not the most complex part of such a project. In this case, the complexity revolved around the line-item recognition and contextual analysis of item descriptions in order to automatically assign tax codes and ledger accounts per line.
The cloud solution would be OpenText Core Capture. The on-premise/private cloud solution would be OpenText Intelligent Capture for Invoices.
Our recommendation would be to start processing these invoices with OpenText Intelligent Capture. Intelligent Capture is a world-class input management platform, specialized in automatically extracting data from paper and PDF documents. The solution is configurable and future-proof.
Since all invoices contain the same type of information (invoice number, date, vendor details, amount, etc.) in a semi-structured format, these documents are highly suitable for automated recognition. Next to invoices, Intelligent Capture can also process contracts, e-mails, inbound mail and many other kinds of documents.
For advanced line-matching recognition, a line-item description search engine can automatically assign tax codes and general ledger accounts to each invoice line.
This article was co-authored by Wouter Boot and Dennis van Aelst.
Updated