This article appears in the Fall-Winter 2019 digital issue of DOCUMENT Strategy. Subscribe.

Image by: VLPA, ©2019 Getty Images

Robotic process automation (RPA) is Dilbert’s dream come true. As a cubicle dweller, imagine having your own personal digital assistant that takes care of the repetitive and boring bits of your job. As a manager, imagine being able to use robots (that don’t complain) for standardized routines while using humans for high-value work. In essence, that’s the promise of RPA.

RPA may bring to mind the idea of robots diligently performing office routines, but it's actually software that automates repetitive tasks that employees perform on their computers, such as keying in data, logging into applications, and clicking buttons or screens. However, RPA just imitates. It's not intelligent.

If we had a Turing test for white-collar work, it would consist of a computer completing tasks, such as purchase order preparation, report generation, and data reconciliation, as proficiently as a tenured employee would. Alas, RPA will not pass such a test, because it only remembers and repeats exactly what it’s been taught before. RPA does not handle tasks it has not encountered before nor does it get better by learning on the job. In other words, RPA is different from machine learning or artificial intelligence (AI).

The current size of the RPA market, including software, consulting, and implementation services, is nearly two billion dollars.

To use a phrase popular in Silicon Valley, this seemingly simplistic behavior of RPA is "a feature, not a bug." It does not subtract from its usefulness. In fact, RPA’s simplicity makes it easy (even for non-technical business users) to get started on automation projects and is one of the main reasons for the scorching growth of the RPA industry in recent years.

Organizations—both small and large—across industries are experimenting with RPA, and many have completed several pilot projects. rpa2ai Research estimates that the current size of the RPA market, including software, consulting, and implementation services, is nearly two billion dollars, and this growth is expected to average about 40% for the next three years.

The power of RPA, when paired with AI and machine learning, goes up significantly, and the next phase of industry growth will be driven by the so-called cognitive or intelligent automation. RPA and AI technologies are very complimentary, and greater adoption of RPA will also spur greater demand for related AI solutions in multiple ways.
  • Identify Work Activities Not Amenable to Simple Rules-Based Automation: In RPA implementations, there are areas where you need human intervention or some basic machine intelligence. RPA vendors, such as Automation Anywhere, Blue Prism, UiPath, and WorkFusion, are all creating such intelligent add-ons. Add-ons from their partners (who tend to be domain specialists) are also available in RPA vendors' bot stores (i.e., app stores for add-ons).
  • The Opportunity to Redesign Processes: RPA projects are mostly about as-is automation—layering RPA on top of existing systems and old workflows. Many such projects only deliver incremental benefits. Today, enterprises have the opportunity to redesign their decades-old business processes (which are limited by legacy technologies) and can take advantage of machine learning capabilities, such as image recognition, natural language processing, and text analytics.
  • Machine Learning as a Service Has Matured: Several APIs for natural language processing, image recognition, optical character recognition, document processing, speech recognition, video analysis, emotion detection, and sentiment analysis are available from the likes of Amazon, Google, Microsoft, IBM, and others. Using these APIs, the RPA platform becomes an integration point to introduce advanced AI functionality into your existing business processes.
  • The Business Case for RPA Has to Address Reskilling and Reassigning Employees: As enterprises gain experience with scaling RPA projects, they are in a position to better understand the softer issues related to change management, such as the challenges associated with employee reskilling or reassignment. Larger AI projects also face similar issues, so you can leverage your learnings from RPA projects there.
In the 90s, organizations combined their business functions, such as finance and accounting and customer support, under the umbrella of shared services to take advantage of centralized scale. This led to the shared services operating model.

The next decade was about globalization of shared service centers, giving rising to the business process outsourcing (BPO) and knowledge process outsourcing (KPO) industries. After such centralization and globalization, the next transformation lever for business services is automation that is enabled by RPA and AI working in tandem.

Kashyap Kompella is the CEO and Chief Analyst of RPA2AI, a global industry analyst and advisory firm focusing on automation and enterprise artificial intelligence. Kashyap is also the co-author of “Practical Artificial Intelligence – An Enterprise Playbook,” available on Amazon. Follow him on Twitter @kashyapkompella.
  • Generative AI (Gen AI) has captured the imagination of industries worldwide, but the true potential lies in its practical applications
  • Digital Asset Management (DAM) is a system designed for organizing, storing and retrieving media files and managing digital rights and permissions. DAM systems have become a core component of creative
  • Is Generative AI tipping the scales in favor of building Enterprise Content Management (ECM) software, or will it ever get to that point?
  • Information technology has undergone a major transformation in recent years, sparked by the rise of “big data.”
  • Every day, large organizations face multiple challenges with the hundreds or thousands of pieces of mail received through the USPS and other carriers, documents that include general business mail