What is optical character recognition (OCR) and why is it important for RPA?
Optical character recognition (OCR) is a key feature of any good robotic process automation (RPA) solution. In short, OCR is a technology used to extract text from images and documents via mechanical or electronic means. It converts typed, handwritten or printed text into machine-encoded text – this data can then be used in electronic business processes without someone manually capturing it.
OCR has been around in various forms for more than 100 years, but unlike the earlier versions of the technology that need to be trained one font at a time with images of each character, today’s artificial intelligence (AI) powered OCR solutions can recognize and capture data from machine printed documents with high levels of accuracy. Their ability to accurately decipher handwritten text is also rapidly improving.
What is the value of OCR in RPA?
OCR in RPA enables organizations to automate a greater volume of their operational business processes, especially those that still depend heavily on scanned paperwork such as customer-completed forms. NICE’s automation tools, for example, include surface connectivity capabilities, which give our RPA solution the ability to pull information in from images, PDFs, and remote applications. Not only is this pulled information read, but actions can be performed as well. OCR for RPA combined with NICE’s advanced Shape Analysis technology makes this magic happen. The result is accurate, fast and high-quality textual output, enabling robots to automate more tasks, while reducing operating and training costs.
The different types of OCR for RPA
There are two broad categories of OCR business cases within the RPA realm. The first of these is about converting unstructured data from scanned documents into structured, digitized data, which can in turn feed into digital business processes. The solution reads and extracts information from a scanned document like an invoice. The data can then be transferred to any enterprise application such as CRM, ERP or legacy system.
The second involves more complex automation capabilities. One example would be using surface connectivity to automate applications off remote machines. Advanced RPA OCR could be applied to read the image and extract the necessary text from the screen image or simulation of the application. This capability enables organizations to automate more processes and expand their automation projects.
RPA OCR use cases
Most mid-sized and large enterprises want to automate and digitize their business processes as well as take advantage of big data. One of the major barriers to this goal is the expense and complexity of storing and analyzing scanned documents. Even in today’s digital world, most companies still process large volumes of documents such as invoices in hard formats with no digital copies available.
That’s where RPA OCR comes in. An OCR engine working alongside and within the RPA platform can automate the time-intensive tasks associated with manually processing these invoices into readable data. A sophisticated OCR solution can be fully integrated into the workflow of complex business process automations, seamlessly transitioning between fully robotic and human-supported input as needed.
Since advanced OCR can pinpoint ‘suspicious data’, any data that should be verified will be sent to a human via a call out screen through an attended process automation and the human will check and, if necessary, correct the information. The verified or updated data will then be submitted back into the system in order to complete the automation.
A practical example of an RPA OCR use case might be extracting information from a scanned customer application form and inserting it into a CRM system. Text analytics can then be utilized to categorize the data, after which, the automation will kick-in and update the CRM application. In the event of an irregularity, the robot will alert a human to intervene.
Another use case could be Know Your Customer checks in financial services. The RPA OCR solution could verify a customer’s identity from their identity card or driver’s license. Once again, if there’s a discrepancy, such as a difference between the customer’s name on the ID document and on the account application form, it can be referred to a human. AI enables the robots to learn from the human behavior or input enabling greater proficiency to manage similar exceptions in the future.
Here is a video example showing how cognitive tools (including advanced OCR and machine learning) can be applied to a banking scenario of opening up a new account.
Towards smarter RPA solutions
As RPA evolves towards cognitive automation – which can handle more complex tasks than the repetitive processes at which RPA excels – advanced OCR is a key feature for any enterprise-grade platform. Cognitive technologies largely mimic human capabilities – in the case of OCR, reading. In addition to OCR, NICE Robotic Automation’s open platform is designed to integrate with cognitive and AI features such as chatbots and machine learning technologies to support the automation of more complex business processes.
Check out this video to see how NEVA (NICE Employee Virtual Attendant) assists employees by handling unstructured data at scale.
