Digital Technology
May 16, 2017

Beyond RPA: Intelligent Automation and Artificial Intelligence

May 16, 2017 - 

Campbell Brown: “What is happening with automation and globalization, that's not going away.”

Much is being written in popular media today about automation, but this phenomenon is definitely not new. Arguably, it goes back at least to the invention of the wheel around 3500 B.C. The progression of technological innovations and automation over the millennia mirrors the march of human progress. Ignoring it is only for Luddites.

Automation has had a profound impact on rising standards of living around the world. For example, technology and automation have re-shaped the agricultural industry—which in the late 18th century employed over 90% of the U.S. population, compared to less than 2% today—dramatically increasing its productivity and enabling the growth of a significantly larger population.

Inside today's modern enterprise, Robotic Process Automation is viewed as an essential productivity tool—one that is necessary for thriving in global, hyper-competitive marketplaces. Enterprises are faced with decisions about where, what, and how to invest to achieve success with their automation initiatives. This post provides guidelines on key considerations for how to best employ RPA, and how to leverage the resulting new digital workforce during your enterprise's automation journey.

Key RPA considerations: Where, What, and How

Since RPA technology has been around for several years, some may believe high ROI is guaranteed. Unfortunately, this is not the case since there are many considerations for an enterprise to succeed in automating a business process at scale. In the order of where, what, then how, these considerations include:

  1. Prioritize best use cases/workloads (i.e., where): Begin by working to understand which use cases are most appropriate for automation. Look first for labor-intensive workloads with deterministic and rule-based processes, and, ideally, standardized or structured data. Start small and work to build broader stakeholder endorsement.
  2. Determine realistic ROI expectations (i.e., what): Some enterprise ROI expectations for RPA are unreasonable, leading to disappointment and disinvestment. Be sure to consider longer development cycles as well as total costs, including costs associated with RPA software licenses, configuration, implementation, and IT infrastructure.
  3. Develop implementation plan (i.e., how):
    • Establish well-defined governance: This is a key ingredient for successful RPA projects. Consider defining roles and responsibilities, creating frameworks for internal change management, and involving business, IT, legal, risk, and other entities early in the automation project.
    • Ensure strong collaboration between business and IT: RPA project success is greatly enhanced when both business and IT collaborate from the outset. This seems simple, but is often much harder in practice. 
    • Re-engineer processes to maximize impact: Processes should be re-engineered prior to applying automation technology in order to maximize business benefits. Look at RPA as just a part of your broader process improvement and digital transformation effort. 
    • Select capable tools and solutions: Evaluate RPA tools with pre-built automation libraries and re-usable components to make connection to back-end systems easier. Consider data extraction capabilities and cost and licensing options. As enterprises mature, they can progress beyond RPA to look to integrate other advanced technologies to realize additional benefits of intelligent automation.

Consider service providers (i.e., who) with proven expertise and domain knowledge of the specific processes your enterprise will automate. The only consideration that does not deserve consideration is when—if not already underway, your enterprise needs to begin its automation journey now.

George Keith Funston: “Automation does not make optimism obsolete.”

Looking forward: AI as next-generation automation tool

Recent advances in technologies, such as data ingestion, computational linguistics (or natural language processing/generation), machine learning, and computer vision, have enabled the creation of new AI-powered solutions. AI solutions will turbo-charge the productivity gains seen with prior technologies.

About the author

Dan Glessner

Dan Glessner

Vice President, Digital

Follow Dan Glessner on LinkedIn