How to Overcome Enterprise RPA Challenges
When scrolling through any tech-focused social media feed, you’ve likely seen hundreds of RPA videos recently that promise your enterprise an innovative forecast of the future. Unfortunately, for all the hype around it, Robotic Process Automation (RPA) still has yet to fully deliver on its promise. According to a recent report from Gartner*, a significant number of businesses that have invested in RPA are questioning its initial ability to drive expected results. “Through 2021, 40 percent of enterprises will have RPA buyer’s remorse due to misaligned, siloed usage and inability to scale,” Gartner analysts say.
While roadblocks can be a regular occurrence when launching any new type of innovation, there’s hope for enterprises to overcome these initial RPA challenges in order to reap the full benefits of process automation.
RPA is struggling to scale
Though corporate inertia can sometimes be the culprit when innovations fail to perform, research suggests that the initial lag with RPA does not lie in a lack of leadership buy-in or low adoption of process automation technologies. Deloitte’s annual Global Robotics Survey** for 2018 found that 67 percent of businesses have started to implement an RPA strategy and that eight of out 10 business leaders support the use of robotics in the workplace—two key metrics that both increased over the previous year’s results.
So why are so many businesses getting it wrong? Too many have so far failed to put in place the planning and processes required for RPA to yield reliable results. Respondents to the Deloitte survey said they are struggling to scale due to a combination of factors that include process fragmentation, the lack of a clear RPA vision and a lack of IT readiness. In the meantime, much of RPA’s potential continues to go unfulfilled.
RPA has the potential to pay out in dividends
What’s hanging in the balance? Vast cost and productivity improvements, to start. Across industries, RPA has been shown to reduce the number of human hours spent on repetitive tasks, bringing down staffing costs and freeing employees to dedicate more time to functions that yield greater business value. By reducing manual tasks, such as entering the same information again and again across systems, and the potential for error that comes with it, RPA can also improve accuracy.
Accenture’s review of the benefits of process automation*** in the financial services and insurance sectors revealed an 80 percent reduction in costs and 80 to 90 percent reduction in time spent on repetitive tasks. What sorts of processes can be automated in these industries? Everything from new account information entry across systems and tracking for fraud detection purposes in banking, to compliance reporting and comparisons of customer-reported claims history against actual history for insurers.
In life sciences industries, RPA could also reduce costs, manual work and improve results by enabling the automation of data entry for clinical trial results, adverse drug effect reporting and other key data, as well as improve sales force effectiveness and other important objectives. But that’s only if businesses are able to overcome their existing RPA challenges.
The RPA data challenge
To understand why the results of RPA initiatives are underwhelming in the face of such great potential, it’s important to first understand how robotic process automation works. Since robots require high-quality data inputs to deliver, the most basic prerequisite for process automation involves getting a handle on organizational data. But steep expectations for automation are directly at odds with the current state of most organizational content: the vast majority of corporate data is unstructured, meaning it exists in formats that cannot be readily digested by machines. Unstructured data that exists across locations or business lines can also be rife with duplications, redundancies, errors and inconsistencies, all of which can further hinder RPA results.
While an organization’s goals for RPA will determine the type of data they need, the first step in any automation initiative has nothing to do with robots – it involves ensuring that corporate data is relevant and usable so that robots may utilize it efficiently. This means standardizing unstructured data, analyzing information assets to determine what data you have, eliminating redundant and erroneous data, and other steps to improve its overall quality and accessibility. Only then can you unleash the robots – and start reaping the benefits of RPA.
Though companies are investing heavily in robotic process automation, many have yet to see its full potential. To avoid ranking among the 40 percent of businesses that are experiencing “buyer’s remorse” about their RPA investment, enterprises must first take steps to deal with their unstructured data, and transform it into clean, process-ready formats.