Humans are key to the Intelligent Automation Puzzle!

Isn’t Robotic Process Automation (RPA) New?

In the recent years, Robotic Process Automation has taken the world by storm, be it through the stock market valuations of Automation Software vendors, that have crossed Billions of Dollars or that the Automation companies have triggered both, awe and suspense as the magic solution to many RPA proponents. RPA however has been around for quite a while. It is notable that most of the RPA software vendors started somewhere around the year 2000 and had initial versions of their products out as early as 2003! So clearly it is almost two decades that RPA has been around – RPA therefore is nothing new and has been around for quite a while!

When I first explored RPA, my initial thought process was that, it looked like Testing software had gone mainstream! For folks who had tested software via automated scripts or had been through software implementations such as SAP implementations, where they had conducted Data Migration using Batch Data Communication (BDCs) or Legacy System Migration Workbench (LSMWs), RPA will not have any novelty value. What was however, impressive was the ability of Automation software to transcend different heterogeneous environments and work across multiple systems. As an Enterprise Architect, I felt that this was almost as good as providing an integration layer across multiple systems, on the top layer i.e. via the user interface – thus minimizing expensive and time consuming integration between disparate systems.

While recent advances in Automation software does allow DLL integration to make the process faster and seamless, the philosophy is largely based on input via the user interface. Another area that was significantly impressive was the variety of business environments where bots were being deployed – be it for capturing data out of shopping carts, to carrying out transactions spanning varied systems ranging from email to ERP, to trades being carried out in a trading system, RPA is literally being used in all possible contexts.

Automation creates Jobs != Myth

While cost saving (read downsizing) is touted as one of the key benefits of RPA, RPA may land up creating newer jobs. While on one end, manual data entry based jobs may be most prone to automation, newer jobs such as RPA developers, solution architects and even roles tiered to RPA centers’ of excellence have emerged. RPA is still not intelligent to take cognitive decisions which require human intervention.

Historically Automation has always created more jobs, some intended and some unintended. Email did not render postal workers obsolete – Amazon and ecommerce which came along with the internet, led to utilization of their services. Similarly internet based apps contributed to delivery jobs. The rise of the Automated Teller Machine (ATM) did not render tellers obsolete, in fact it contributed to increase in branches. In some cases inadvertently, it led to the requirement of physical security for servicing the ATMs and guarding them! Internet Banking has led to the generation of cyber security jobs.

Jobs created due to new age Business Models, Internet and Automation

While RPA can transform rule based, repetitive processes, it still requires people to act when exceptions occur. The intelligence that the bots deployed display, is wholly pre-configured by humans and bot operations need to be monitored by people. 

 People are at the Heart of Intelligent Automation

While bots deployed in RPA are work-horses in automating processes, they are unable to navigate cognitive decision making where human intelligence reigns supreme. While it can be argued that machine learning can possibly perform cognitive decision making, economics will ultimately determine what level of autonomy is achieved by machines. While Autonomics (A.I. powered Automation) is not theoretical any more, with advances both in Automation and in A.I., the path forward is neither smooth nor simple for Automation.

Automating complex processes through organizations have their own challenges. Interestingly Organizational Change Management is one impediment that affects most RPA initiatives when they are required to scale through the enterprise. Lack of buy-in from stakeholders of an enterprise, unclear governance and lack of ownership from IT may be some of the impediments that impact the success of an automation initiative.

Having the right Automation strategy where Process selection for Automation is optimized is also another key area. Trying to fully automate all the processes may not be the right strategy – the key impediment being ability to scale. While Agile methods are most suited for automation initiatives, a deep dive is required to ensure that processes that are selected for automation provide Business value. Too complex and heavily customized processes may not be the right fit for automation.

Scalability both related to technical and process are impediments, that have to be thought about, before embarking and designing an automation initiative.

Robotic Process Automation is a fantastic tool with exceptional capabilities. However at the end of the day it is what it is – a fantastic tool. Like any other transformation it requires expertise and people engagement to be successful!

The Author is a Top MBA, Enterprise Architect, SAFe Agilist and Practitioner, certified Scrum Master, Prince 2 Practitioner, Blockchain expert and an AA certified Advanced RPA professional.

This disclaimer informs readers that the views, thoughts, and opinions expressed in the text belong solely to the author, and not necessarily to the author's employer, organization, committee or other group or individual.