Companies across all sectors are looking for ways to increase the efficiency of their administration functions to help reduce costs. In finance many companies have already realised the benefits of automating invoice processing within their accounts payable department, but for those who are currently considering this technology we hope our short ‘best practices’ guide will be useful. At Datafinity we have many years of experience in implementing invoice processing systems so we can recommend some best practices which enable our customers to get the maximum automation and efficiency from their accounts payable systems.
How do purchase invoices arrive?
In most organisations invoices arrive through the post (paper) or via email as PDF attachments. Because nearly all invoice processing systems use OCR (optical character recognition) it is very important to have good quality invoices so that manual intervention needed to handle unread or misread invoices is kept to a minimum. If you are receiving invoices through the post then make sure they are scanned with a good quality scanning device that produces 300 dpi (dots per inch), black & white or greyscale images. This is the optimum quality for reliable OCR results. Better still, try and encourage suppliers to send you invoices by email as PDF attachments. However, if the supplier’s PDF file is not generated directly from their accounts system but from a scanner, ask them to adhere to the image quality recommendations above. It will save you a significant amount of time!
How do I separate multi-page invoices?
This is a question many of our customers face, especially when scanning paper invoices arriving by post. If you receive paper invoices, and you want to scan them in one batch, then you have three options. Use a separator sheet (blank page or barcoded) between each invoice, use a barcode sticker on the first page of each invoice or setup the scanning software to separate on specific keywords. The last option is by far the least labour intensive but does usually need some fine tuning to make sure it is reliable. If invoices are arriving as PDFs then separation is much easier because each file can be regarded as a new invoice. For single PDF files containing multiple invoices the software keyword separation method can be used.
What data do I need to capture from each invoice?
This will depend on what the data will be used for. If the invoices are just going be stored in a document management system for later retrieval then only document search data is needed, such as invoice number, purchase order number, invoice date, etc. If the invoice data is also being imported directly into an accounting system then all pertinent financial data is needed, including data retrieved from external systems, such as cost code, VAT code, supplier reference. If the supplier provides products or services that span multiple cost codes, that have to be allocated separately in the accounts system, then its likely that line item data will also be needed from certain invoices. Its important for future proofing that the invoice processing system you choose has the capability of automatically extracting line item data if needed.
Do I need 3-way matching?
Three way matching is where the data captured from an invoice is matched against a purchase order (PO) and a goods received note (GRN). In many cases a match check between the invoice and the PO is all that is required. However, if an invoice contains only some of the items ordered, because of a staged delivery or lack of stock availability, then a further match against the GRN may also be required. This is usually referred to as 3-way matching. Checking the invoice against the PO and the GRN will also require data from the GRN to be captured and stored in a database or document management system.
If you would like more information or advice on the subjects covered in this article, please don’t hesitate to get in touch.