Reginald J. Twigg, PhD, Director of Product Marketing for Enterprise, ABBYY
Is Digital Transformation All It’s Cracked Up to Be
In the everyday realities of document processing, whether from electronic or scanning sources, ‘digital transformation’ seems like an over-hyped buzzword with little connection to the actual work we do. There is truth in this assessment, but there is also a need to understand how digital transformation is going to change the realities of document processing sooner than we think. Are we ready for them? In order to answer that question, we need to get a handle on the practical side of digital transformation, what impacts it is having on document processing, and what are the first steps we can take to be ready for it.
Digital Transformation Simplified
Digital transformation marks a seismic shift in how technology is used to solve ongoing problems of customer service, operational efficiency, compliance, and Service Level Agreements (SLAs) we have with our business and its customers. As a seismic shift (one could even call it a paradigm shift) in the relationship between technologies and business, digital transformation simply is about using smarter technologies in innovative ways to change how our businesses operate in meeting our SLAs.
In doing so, this transformation is less about inventing new technologies than it is taking advantage of innovations in AI (artificial intelligence), ML (machine learning), NLP (natural language processing), process intelligence, micro-services architectures, and new consumption models (for example, subscription, cloud, mobile) to understand how our processes actually work and how content flows through those processes. With this understanding, we get practical guidance on where change really is needed and how automation can be targeted more effectively to get results.
We have heard all the hype about RPA (robotic process automation) revolutionizing business with new levels of automation, and with it a latent fear that bots will take our jobs. The reality is that bots can only perform specific, repetitive tasks within a defined process, which makes our jobs easier – that is, if they are used effectively. According to many sources, 65-70% of RPA implementations are in serious trouble or failing entirely. Why? The most commonly-cited reason among our customers and industry sources is that bots are being used to automate the wrong processes.
Second behind this is the fact that many problems bots are being deployed to solve are actually around the problem we solve every day – intelligently extracting data from documents and content. One of the most common scenarios we see at ABBYY is using bots to gather emails with attachments and send them off for OCR and extraction. Another is using bots to verify data fields extracted from documents. All of these functions are performed more than adequately today by the intelligent document processing (IDP) solutions we have today – FlexiCapture, for example. So why are companies looking to RPA to solve this known problem?
Some of the reasons for this are complex, even shocking. One in particular is that the need for document processing has moved out of the mailroom, BPO, and scan center into the business process, and, suddenly, it is the business users and no longer just the back office, who are demanding these solutions. RPA vendors have caught the attention of this new buying audience, but IDP already provides the more complete, proven solution for the challenge of getting content out of documents and messages into business processes, which tend to run on clean data.
Getting Started: The Process-First Approach
Fundamentally, the challenges of automation generally, and RPA specifically, are process problems, so we need to take a process-first approach to solving them. Many attempts to use smarter technologies to automate business processes fail because they are automating the wrong processes. At the core of this problem is not actually knowing how your processes are working – how much time does it take to perform specific jobs, where are bottlenecks occurring, which document review exceptions are taking the most time and by whom, where are tasks being repeated by multiple people.
All of these real problems can be assessed with data, visualization, and, most importantly, seeing process timelines. While we have schema, dashboards, and tools to show how these processes should work, a fact-based look at where time is spent in processes provides the more complete picture and starting point for change. We call this timeline-grounded approach to assessing and targeting automation to solve the most urgent process problems process IQ. Process IQ is having an informed, factually-based understanding of where problems are occurring in your processes, where documents and content are creating bottlenecks, then knowing where the best opportunities are to solve them with IDP capabilities.
IDP, according to Everest Group, is ‘any software product or solution that captures data from documents (e.g., email, text, PDF, and scanned documents), categorizes and extracts relevant data for further processing using AI technologies such as computer vision, OCR, NLP, and machine/deep learning. These solutions can be integrated with internal applications, systems, and other automation platforms.’ IDP, which includes the OCR and capture capabilities for which ABBYY has been known for many years, takes the next step to become more process aware – knowing where processes consume document data, how those processes work, and how time is spent on them.
Getting a Handle on Processes and Documents
Digital transformation as a concept can be overused, but it presents a real opportunity for IDP to make a difference in day-to-day processes and activity. So many of the challenges we face as imaging and capture professionals are now getting the attention of new audiences in our organization, forcing us to get smarter in how we apply document processing capabilities into our everyday processes. While scanning and capture have been around for decades, what’s new is the growing awareness of how these capabilities need to be consumed in our processes.
Process awareness, or process IQ, is what makes IDP different from the historical back-office functions to which capture historically has been relegated. As a set of document processing capabilities intelligently targeted to processes that need them, armed with a fact-based knowledge of how time is spent in these processes, along with finding easier ways to consume these capabilities through microservices, AI and deep learning, IDP can make digital transformation work more efficiently and effectively for a change.