Let’s paint a scenario: Your company has invested in the tools and training to undergo a true digital transformation. Your processes are more streamlined, documents are easier and less costly to store and retrieve, you’re saving time and money, and after a brief adjustment period, your employees are happier as a result.
There’s just one problem. Not every company is on the same page as you, and you’re being flooded with paper mail as a result. Naturally, that information needs to be transferred to your ECM solution, but manual entry is the kind of time-consuming, repetitive task that you invested in a solution to avoid.
Short for optical character recognition, OCR can help automate the process of transcribing these documents into computer-legible text. Below is an excerpt from our expert OCR guide detailing 4 key factors that impact your organization’s OCR strategy. Interested in learning more? Download Your Guide to Mastering OCR.
A Wide-Range of Strategies for Varying Business Needs
As a discipline, OCR implementation can be diverse. Its practice varies widely on a case-by-case basis and depends greatly on an organization’s size, process maturity, technical sophistication, resources, and objectives for the solution’s use. A common thread, though, is that the success of an implementation initiative depends to a great degree on an organization’s understanding of its needs.
1. Capture Accuracy
Many factors play into the accuracy of an OCR solution, such as font type and density; the use of irregular fonts and logos; and most notably, the source of the image. A PDF or quality image of the original file will almost always yield the best results, but even in these cases, solutions are typically 99.7% accurate. Still, there are many cases where 100% accuracy is required for the task, such as:
• Sharing information with other line-of-business applications
• Indexing documents that need to be retrieved quickly
• Searching emails for data points to help assign tasks
When information needs to be 100% accurate, strategies that limit dependence on OCR or use formulaic checks to ensure accuracy are typically the most effective.
Example 1: Suppose your company plans to index insurance forms and has a highly accurate source of related data. In that case, an OCR strategy may involve capturing one key index field to identify the document and pulling the other index fields from that data source.
Example 2: On the other hand, if your company does not have an accurate reservoir of data to pull from, insurance information such as member ID and group number can be checked for accuracy by noting the format of these numbers and searching the document for other numbers with that format to see if they match.
2. Age of Document
Older documents that are stained, torn, or faded may all impact capture accuracy.
Some capture solutions clean up scanned documents before beginning the capture process by adjusting for cockeyed scans; removing spots, lines, and other print imperfections; and compensating for stains, fading, and signs of age. If you work with high volumes of documents that contain these types of imperfections, it may be worth looking into systems that perform this cleaning process.
3. Fast and Frequent Retrieval
Accuracy becomes a high priority when documents need to be retrieved quickly or frequently.
Example 1: If your business needs to reference invoices to respond to the needs and questions of your vendors, partners, or customers, having them readily available is your top priority. Accuracy is critical to locating this information quickly, as you would likely be using whatever index field is available on-hand.
Example 2: Inversely, documents that are being captured because of mandated retention rates and potential audits are usually rarely referenced. Often companies have a day to provide these documents. In these less time-sensitive cases, the typical 99.7% accuracy will suffice since one incorrectly captured index field simply requires you to search with another field, after which you can quickly correct it for the future.
4. Capture Volume
The number of pages you’ll need to capture regularly plays a vital role in determining your OCR strategy. While OCR speeds up the process of leveraging your information quite a bit, capturing large volumes of paper documents can still take time, especially if accuracy is a priority. In situations involving high capture volumes, several implementation strategies can be used.
Example: An educational institution plans to capture student documents. In this case, all documents related to a specific student can be grouped and labeled with a bar code to ensure index fields common to each document are accurate.
There are other possible strategies, such as implementing validity checks and limiting reliance on OCR, but ultimately which strategy makes sense depends on your organization’s needs and available resources.
In Summary
By balancing the factors that affect document capture with your organization’s needs, an effective solution can be designed to help you fuel your organization with the data it needs to thrive.
Square 9 is an end-to-end provider of digital transformation solutions with innovative approaches to document capture, enterprise content management, web forms, and workflow automation. To find out more about OCR, its place in digital transformation, and how best to leverage it, download our complete OCR Guide.