From Wikipedia, the free encyclopedia - View original article
Invoice processing involves the handling of incoming invoices from arrival to post. Invoices have many variations and types. In general, invoices are grouped into two types: 1. Invoices associated with a request 2. Invoices that do not have an associated request.
Most organizations have clear instructions regarding processing incoming invoices. Different sets of instructions are commonly found in most organizations regarding the handling of purchase order invoices or non-purchase order invoices. The main department that processes invoices is known as accounts payable department. The process involving a supplier invoice is also known as purchase-to-pay.
A typical process involving paying a supplier invoice begins at the arrival of invoice at the door of the organization regardless of the methods of arrival such as via email, postal mail, facsimile, etc. Once the invoice arrives, the accounts payable clerk must ensure that the document is indeed an invoice. Then the clerk classifies and sorts the invoice into various categories. The definition of various categories is not always the same across different organizations but is normally defined by each organization.
Once its category defined, the invoice is forwarded to the responsible person for that particular invoice. This is normally the person who has placed that order. If there has been a purchase order involved when placing that order, the invoice must then be matched against the purchase order to ensure that the amount invoiced is correctly stated on the invoice. If the amount is right and the goods have arrived, the responsible person will have to approve the invoice by signing off on it. If the amount invoiced exceeds a certain amount that is limited by the organization, the superior of that person may have to approve the invoice as well. This, of course, differs from organization to organization.
A manual invoice process can sometimes exceeds 15 steps before the final posting is done.
Technology has long enabled the automation of invoice processing from arrival to post. This means that at arrival of the invoice, the same accounts payable clerk will only need to scan the invoice into an automation software. The automation software then converts the invoice's scanned image into a text researchable document. The different fields on an invoice can also be defined into the software so that it remembers which fields it should capture and register into the ERP systems, for instance, the amount purchase, the quantity, the supplier name, the supplier code, and so on. The benefits of an automatic processing work flow may include reduced human error, on-demand reports, and data resilience. Automated invoice processing system vendors include Open Text, Contact Telecoms, Basware, Automation Centre and Read-soft. Most automation software today integrates into common organizational ERP systems such as those offered by SAP, Microsoft and Oracle.
In an automatic process, once the data is extracted or captured from the invoice the data is sent into the system for automatic matching against the purchase order. The responsible person will then receive an email alert so that he or she can approve the invoice. If there are other people involved in the approval work flow, email alerts to them will also be automatically generated.
Pay Stream Advisors, an industry analyst, has defined accounts payable automation as the solutions and services which automate the people and paper-based processing invoices for approval and posting into the accounting system. More information about automating invoice processing with technology can be obtained from the common vendors' websites such as ABBYYACOM, Documation, and Version One Ltd. According to the Document Imaging Report, issue no. 8-17-07, there are around 3757 invoice processing applications in the world.
In the 2008 report, "Optical Character Recognition: Invoice Imaging and Data Capture for Accounts Payable," Pay Stream advisors writes:
<block-quote>Intelligent Data Capture (IDC) or learning systems enable end users to extract content from invoices without the system having to learn the layout of the invoice. Some intelligent engines are able to correctly sort batches on the fly, locate data fields such as invoice and PO number, as well as line item information, and then extract the desired content from those data fields.
Intelligent solutions do not require the coding of rules or design form templates. Rather the system learns by reviewing a relatively small number of invoice samples. This helps the system scale to large invoice volumes and widely varying document layouts without requiring a human operator to specify a template for each one, or explicitly create and tune an extensive library of keywords.