When it comes to the day-to-day and year-on-year running of any company, there are always inevitably going to be mountains of financial information to process. From cashflow and receivables, to loan payments, one-off costs, employee expenses, utility bills and virtually everything else related to staying in business, there are endless invoices, accounts to track, people to pay and payments to chase. And this in itself can generate a substantial amount of extra financial costs, in terms of the time spent by your finance department and other staff members in dealing with this never-ending train of numbers.
Fortunately it is now becoming possible to automate large swathes of this numerical burden, by using software that can recognise all relevant figures and actually feed them into your accounts and relevant databases. But if you’re planning to entrust such vital duties to machine minds, you’d better take steps to ensure their accuracy, or you could end up losing more than you stand to gain in the long run.
The promise of optical character recognition
The key technology which enables the automation of financial processing tasks is known as optical character recognition, or OCR. In basic terms, OCR technology, when incorporated into the document scanning process, can recognise all of the numbers and text in your invoices, balance sheets etc. as not just ink marks on a piece of paper, but as the actual numerical figures, letters and words that they represent. They can then be outputted to a chosen and electronic format, and inputted directly into your accounting software and financial reports.
Avoid the ‘garbage in, garbage out’ trap
If you’re learning this for the first time you’re probably quite excited to give it a try, but a word of caution must be voiced. There are of course a variety of free and inexpensive software solutions out there on the internet which purport to offer OCR document processing, but these are often inaccurate in the results they provide. Just how far wide of the mark they are can vary widely, but even small mistakes in the scanning and character recognition process can have disastrous consequences.
For example, imagine a situation in which all invoices for a year are scanned using OCR technology to calculate the total outgoings to contractors and suppliers for the purposes of a financial audit. In 10% of scanned documents there are single digit errors, such as the misidentification of a digit, or the addition of extra digits. Over the course of the hundreds of documents scanned, how much discrepancy will there be between the reported totals and the reality?
In another hypothetical example, an extra 0 is added to a scanned invoice in error, resulting in a payment being issued for £1,000 instead of £100. Even if the error is then noticed and rectified by human intervention, this takes additional time, nullifying the main point of automating the task in the first place. Poor, inaccurate solutions will make your document processing and document management practices worse not better, due to the oft-used adage of ‘garbage in, garbage out’ – if the quality of the input (the OCR conversion) is poor, the results will also be poor. And if you can’t trust the results, the solution is no solution at all.
Choose a trusted solution
Fortunately there are also highly accurate OCR document processing solutions on the market which, for an affordable fee, will allow you to dramatically automate your financial processes. Nuance’s Omnipage is a full-featured OCR software package used by Fortune 100 companies, and it uses innovative ‘3DC’ technology to intelligently identify text, whether print or handwritten, digital or analogue. What’s more, it offers support for most commonly used business productivity applications and formats, including PDF, Excel, Word and HTML.
Why not take a closer look at what OCR document automation can do for your company today?