The global market for robotic process automation (RPA) technology continues to increase. In fact, analyst firm Gartner recently named RPA the “fastest-growing segment in the enterprise software market". That’s why in 2019, Pega surveyed more than 500 global business leaders from varying industries for their insights and experiences with RPA. (You can read my blog about our most important findings here.) By far, the biggest challenges faced by these respondents included: Broken bots (87%), harder than expected bot deployment (50%), and maintaining automations (41%).
The problem: Lengthy deployments and manual maintenance
Historically, to deploy new bots or maintain existing RPA bots, businesses needed highly trained RPA mechanics at the outset of any robotic automation design working under the hood to identify controls. Without this unique identification, robotic automation is simply not possible. The process is largely manual and requires automation specialists to spend a copious amount of time identifying application controls, but is a necessary step to deliver to powerful, robust automations. Some form of control identification is required in any RPA solution, so making this step easier and quicker can, in turn, reduce the amount of time needed to deploy and maintain bots.
The solution: Leveraging AI-enabled, adaptive RPA to automate control identification and integration
Instead of using humans to manually search and identify every control, the newest versions of RPA can actually automate this process. In Pega RPA, we use a capability we call X-ray Vision, which identifies and matches controls in enterprise applications. Pega is able to do this because of its patented Deep Robotics technology that has been running under the hood of Pega RPA for 15 years, powering faster and more effective automations. Unlike screen scraping techniques that focus on scraping data from the surface of applications, Deep Robotics operates at the foundational level of applications – the binary code – wrapping around that code, making it possible to automate bots in lockstep with the natural motion of an application. Together, X-ray Vision and Deep Robotics work in tandem to remove the layers of abstraction that slow bot implementations, allowing businesses to integrate bots more quickly into their enterprise systems and implement RPA at scale.
The future: Self-healing RPA
But what happens when an underlying application in a robotic automation changes? In most cases, the automation breaks, taking an entire business process offline until fixed. To say the fixing of a broken robotic process is time-consuming is an understatement. In situations where the enterprise is not using Deep Robotics, this requires the process owner to go back to square one, back to the developer, back through regression testing, and so on. A broken bot basically means a production outage, which inhibits the organization from meeting their business objectives with RPA. Given the dynamic nature of today’s enterprise applications, there needs to be a better way.
Going forward, more and more businesses will take advantage of artificial intelligence capabilities to identify and fix broken bots. The AI will automatically identify an underlying application change, even at the most microscopic level, then determine the maintenance that needs to take place to allow the automation to self-heal. With this capability, end-users go from viscerally experiencing a broken process – with a problem originating at a level far beyond what they can see – to a solution that automatically fixes the problem out-of-sight before an end-user notices anything is wrong. This self-healing capability eliminates the need for manual control identification, dramatically increasing the time-to-value for new robotic automations, getting to that “it just works” experience in record time.
The result is fast, reliable, and resilient automations that work invisibly behind the scenes of your most critical business applications. For example, agents rely on RPA in the contact center to save time on data entry and help customers reach outcomes faster. A broken automation can lead to an unsatisfactory customer experience, so fixing it quickly is essential. Or, consider the role of RPA in pulling a credit score during a loan application process. When RPA “just works,” the loan process is resolved much faster. The role of RPA in employee and customer experience is tremendous, and having RPA bots work without interruption is critical for RPA to meet its business objectives.
Self-healing RPA is the next-generation RPA experience. Beyond reducing hours spent on maintenance, this X-ray Vision ability to “see” into automated applications will give RPA developers new powers to build even more robust automations and help the enterprise, their employees, and their end-users have a frictionless experience that just works, no matter how many times supporting applications change or how complex they are.