Most RPA Implementations Revolve Around Document Processing. Here Are the Five Most Common Ways Companies Are Doing It.
By Joe Labbé on November 23, 2020
With all the hype around Robotic Process Automation, it’s helpful to remember that the vast majority of all the data running through RPA automations originates or terminates with a document. If we’re spitballing, I would call it about 80 percent.
Granted this number is based on experience, and I work for an intelligent document automation company, but in my 16 years prior to working at KnowledgeLake, I observed similar trends. Because so many RPA automations involve documents, we can see the emergence of common document processing patterns.
Here are the five most common ways companies are using RPA to enhance document processing:
1. Transactional Data Entry, or Straight-Through Processing
Transactional Data Entry is the mother of all unattended bot automations.
Unattended bots do not run on a user’s personal workstation. Rather, they run on one or more physical or virtual machines tasked with performing more batch-oriented and longer running transactions. Unattended bots are tailormade to process documents that flow into the enterprise on their own timetable, and where timely and hyper-accurate data entry into downstream line of business systems is critical. We are talking about an organization’s most important operational documents such as invoices, insurance claims, electronic medical records, shipping documents and more. This is the data lifeblood that courses through the veins of every organization. No matter the document, the processing pattern is often the same:
- Documents are acquired and ingested
- OCR and machine learning auto-categorize documents and extract transactional data
- Extracted data is validated against internal data sources
- Validated data is automatically entered into downstream line of business systems via the application’s user interface (just like the user does it)
- Real-time prompts are generated and presented to skilled knowledge workers to handle exceptions
- Transactions are recorded and logged for audit and compliance purposes
RPA makes the elusive, Brigadoon-like promise of straight-through processing a reality. While transactional data entry is usually handled by unattended bots, there are many attended bot uses as well. The bang for your buck will vary depending upon document volume, duration of the transaction and verification requirements.
2. Document Acquisition
Document acquisition is not always a simple task. While most document processing platforms can acquire documents from scanners, cameras, watch folders and email accounts, this is far from an exhaustive list of the channels through which documents enter the enterprise. Many organizations are also forced to acquire documents from locations such as websites, FTP sites, EDI translators and more.
One example: ECI Defense Group provides government procurement solicitation information to its subscribers. Prior to implementing RPA bots, ECI staff members would trawl multiple government sourcing sites, multiple times per day, looking for new and modified solicitations. This is time-consuming and tedious work.
After implementing RPA, staff members were freed up to handle more customer service-oriented functions while letting the bots comb the sites, determine which solicitations are new versus updates, and update their internal solicitation tracking system. The results are faster, more frequent and more accurate processing for a fraction of the manual cost. (For more regarding ECI’s use of RPA, click here.)
Another client receives most of its orders from one large customer that posts orders on its various divisional web sites and also sends orders via EDI. Before implementing RPA, this client had dedicated personnel to monitor both the websites and its EDI VAN (value-added network) account for incoming orders. After implementing RPA, those document sources are now monitored and processed by bots, which in turn send the orders to fulfillment and accounting. The results were spectacular. The client reports order processing costs have declined 25 percent and order volume from the customer is up 12 percent based on increased customer satisfaction.
RPA can be an incredibly powerful multi-channel document acquisition tool.
3. Document JumpTo, or Short Cuts to Needed Content
The top reason organizations give for not centralizing their document repositories or duplicating documents across line of business systems is the need to locate documents from within different contexts. Meaning, accounting may need to retrieve an invoice from a supplier record within an ERP system, while an HR professional may need to view a change of life event form from within an EHRS.
Satisfying these requirements would normally involve creating a custom connection to the repository from each application screen, in each line of business system. Not only is this a ton of work, but in most cases, it is impossible to do. However, via an RPA attended bot, a “jumpto” button can easily be placed onto any screen with no changes to the application required.
When pressed, this button kicks off an automation that extracts screen data to establish context and then uses that data to “jumpto” the related document. The attended bot functions as a universal adapter that can determine application and data context for a given user and present the user with the right document(s). The same is true for contextual requests flowing in the opposite direction, or something we call “reverse jumpto”. For example, if a user is viewing an image, such as an invoice, the user can click the jumpto button, which extracts the metadata required to establish context and navigates through the ERP system to the related bill.
The possibilities of creating a powerful, contextually hyperlinked experience among data, documents and system records are endless.
Despite the power of machine learning and its ability to perform document auto-classification and data extraction, many documents are still manually indexed or supplemented with metadata from line of business systems.
“Quickcopy” is the process by which a user opens a record in a line of business system, and via the same button described above, kicks off an automation that extracts select pieces of data from one or more application screens and quickly copies them to the metadata fields of the open document. As with jumpto, the user can reverse quickcopy by pushing data from the metadata fields accompanying an open document to fields within the selected record in the line of business system.
Note the more you adopt the jumpto pattern, the more jumpto allows you to interact with documents in a central repository from any application screen on the desktop.
5. Document Assembly and Forms Extraction
Document assembly and forms extraction have been around forever. By leveraging the power of RPA, any application screen(s) can easily be used as the source to assemble a document, or as the destination for data extracted from a form.
In the case of document assembly, data is extracted from application screens and assembled into a document or email. Think of it as a universal mail merge capability that can be added to any application that assembles or injects data into any document. A great example is the creation of a contract in DocuSign that gets its inputs from one or more selected records in an application.
Forms extraction is really just the other side of the same coin. In this case, data is extracted from a form and inserted into the fields of an open application record. Document assembly and forms extraction are equally used by both attended and unattended bots.
The Way to RPA
RPA and document processing go hand and glove to boost productivity. Borrowing from these five patterns can inject greater efficiency into the productivity-draining tasks currently slowing your organization down.
To learn more about how to intelligently process your organization’s most important documents, contact KnowledgeLake for a discussion or demo.
Stay in Touch
Receive the latest blogs from KnowledgeLake in your inbox!