Improve Document Processing Workflows With Intelligent Data Capture: Part 3

This is my third article in a series on the topic of intelligent data capture. I’d like to again underscore the point I’ve made previously that effectively managing document processing activities is critical to business success. It can help raise employee productivity; reduce the cost and cycle time associated with processing documents such as invoices; strengthen security and enhance customer service. Intelligent data capture is an approach that can help you streamline document processing activities and realize these and other goals.

In my previous articles I focused on the potential business advantages of intelligent data capture and I spotlighted three of five key steps to successfully implementing a solution. (The three steps discussed previously include centralize intake, automate capture and classify documents.) I’ll now address steps four and five and then highlight case history examples that illustrate how businesses are leveraging the intelligent data capture process to achieve substantial benefits.

4. Extract Data

Step four revolves around cost-effectively extracting data. This means reducing labor costs associated with data entry. The key point is that in its current document processing an organization may use manual data entry which, being labor intensive, causes the overall document management process to be more costly and error-prone than it needs to be. Additionally, manual data entry keeps the document processing system from being scalable because there is a limit on how many characters an individual can key in a given period of time. Finally, if more processors are required, the learning curve for training them is going to differ for each person.

To resolve this situation, an enterprise can take a few initiatives. One is to determine critical data points and the rules for finding them. Then, leverage software to automatically extract those data points based on clear rules. Finally, assign manual steps only to exception processing.

5. Export Data

The objective of step five is to distribute—or export—the extracted data to areas within the organization where it will be used. As with the previous steps, there is a least one key challenge involved. In this case, creating custom links between capture systems to downstream platforms such as an ERP system can be expensive and usually require professional service support. There are a few ways to help alleviate this pitfall:

• Document the requirements for the downstream systems. These systems may be able to accept information in more formats than expected, enabling an “out of the box” export from the capture system to work.
• Use a phased approach, connecting to one system at a time. This potentially reduces the amount of professional services needed at once.
• Leverage a managed services provider with the right experience and technology. The latter might include a connector that works “as is” or could work with a few changes.

Case History Examples

At this point in our article series, we’ve examined five steps that can help support a successful implementation. Now let’s see how the intelligent data capture process can work in real world scenarios via three brief case history examples, two which I’ll spotlight here and the third in my next and final article in the series. The first case history, as illustrated in Figure 1, concerns an insurance company that needed to capture data from the approximately 10,000 claims it received monthly.

Figure 1

The company receives the forms in several formats (PDFs, scanned typewritten forms, handwritten forms and faxes). Because the insurer receives the forms at its main location, centralizing the inbound document stream was a relatively simple step. Additionally, as part of its information intake strategy the company decided to use an offsite processing model in order to maximize space and contain cost.

The other elements of the solution included automatically capturing data from typewritten and electronically generated forms. Then, business rules were applied to the extracted data, which facilitated moving the majority of the forms efficiently through the approval process workflow. This had the effect of lowering the company’s claims settlement costs and reducing processing time to less than 24 hours. Manual activity was only used for processing handwritten forms or forms that required exception processing. This system not only helped raise efficiency and lowers costs, it enabled the company to better meet industry regulatory and compliance requirements

In our second example, highlighted in Figure 2, a manufacturer needed to digitize unstructured data in the form of vehicle and dealer files, comprising approximately three million pages.

Figure 2

The company teamed with us to help address several key challenges and implement a solution. The challenges included:

• The documents were stored in multiple locations.
• The company needed a secure central repository, but didn’t want to install or manage it on the company’s premises.
• While some files could be digitized offsite, some sensitive documents had to be scanned onsite.

Our solution was to create a temporary scanning operation onsite for the sensitive documents and manage intake for the other documents at an offsite facility. Our team used OCR technology to capture important vehicle file data in full text. This data includes automobile VIN number, vehicle type, region and year. We also installed a secure cloud storage system for the company, into which we imported the documents. The repository included separate network access for each location in order to provide added security. The project, scheduled for six months, was completed earlier than planned.

In my fourth and final article in this series I’ll examine a third case history example of intelligent data capture in action and close with a discussion of the pros and cons of managing data capture internally versus outsourcing the process.