CONTRIBUTION BY Matthew Cook –Senior Supply Chain Technology Leader
Optical Character Recognition software has been available commercially since the 1970s and the technology keeps getting better but you might ask so what?
So what if you can capture data more accurately? Where studies have been done, they have found that more than half of businesses (in the USA) still use paper invoices, and globally less than 10% of the estimated 500 billion invoices are paperless.
Agreed. Accuracy on its own isn’t a compelling enough argument to move to automation. But if that accuracy is combined with other tools in your enterprise it can bring huge value. It can even transform B2B business. How? Use your imagination. Pairing OCR with other applications is the key to bringing amazing benefits. I’ll explain how, and give a few examples within a supply chain context.
A brief history of recent OCR advances
To see what I’m talking about, first digest this quick recap of OCR advances over the years:
First we had template-matching. This enables software to ‘look for’ and find the right value from a document in the right place on the page.
Then came self-learning. Self-learning OCR systems ensure that word and character images are compared to a standard. And, over time, the software learns which images are likely to match or have a ‘high probability’ to match.
And now… we can even re-purpose the digital content created from paper scanning!
So how does OCR evolution help make business better?
When combined, these advances (and others) have opened three large avenues of opportunity for nearly every business – when it comes to auditing documents, managing exceptions, and interrogating data. Here’s an explanation of how these advances work, and an example of each.
Opportunity area 1: audit
With these advances, it is now possible to compare almost anything on paper documents with almost anything else. Here are just a few examples: digital contracts, shipping documents, tax laws, official government documents, calendars, tariffs, tax tables, utility rates, pricing tables, or any other established numeric parameters! And once the comparisons have taken place, these documents can be processed and mismatches can be stopped, investigated and corrected.
An audit example:
How can you ensure your company is billed the proper freight cost? Simple – paper freight invoices can be scanned, digitised, and compared with tariff schedules, by carrier.
Opportunity area 2: exception management
What happens if things go wrong? Exceptions happen. But now, exceptions can be identified, and quickly (by email) returned to their sender. And because everything that has been digitally captured can be audited against numeric parameters it’s possible to see precise results and check: is the value within the contract limits? Does it match the contracted rate, etc.? It’s now a relatively simple matter to write reports that summarise huge amounts of information such as: the number of invoices processed and paid according to their contract parameters; and the number that have been rejected (along with the contents of those rejected invoices).
An exception management example:
In the freight invoice example above, you can stop non-compliant invoices and 1) send them by email to the shipping or accounting department, or 2) return them to the shipper, or 3) send them for payment automatically (but with the correct freight rate applied).
Opportunity area 3: analytics
Automation ensures that the tens of thousands of paper documents (like Bills of Landing / Invoices / PODs / GRNs, Orders etc.) can be dealt with appropriately. We know that with clever technology, thousands of these may have adjusted quantities or notations that can be isolated, summarised and categorised. But then what? Actually, analytics takes it a step further: patterns can be established depending on the item, customer, delivery point, or mode of transportation (just to name a few).
An analytics example:
Picture a database of scanned and digitised delivery documents. With this database it is possible to create (using simple queries) all sorts of interesting reports. Perhaps you want to figure out what costs are due to damage or to late deliveries. You could even analyse this by delivery point and shipper.
Getting to a stage where these practical examples are possible may sound simple, but there are companies out there with billions in sales that are still trying to manage paper delivery documents with a large number of dedicated employees and customer service representatives.
In fact, paper will likely be prevalent for many years to come.
In conclusion: automation is a no-brainer
To me, it makes sense to make the most of technology and ensure that people can do what people do best – and let systems take care of the data behind the scenes. In this way humans can then act on the intelligent analytics and information they provide in order to make a real difference to customers (perhaps, even, to make your customers smile!). After all, if automation exists, shouldn’t it be used to make life easier?