Source: COGNITIVE WORLD on FORBES

Robotic Process Automation (RPA). It’s a hot topic among the C-Suite. Where to implement. How to implement. How many headcounts can be saved through robotic implementations? Business leaders want robots – powered by AI – to drive automation into every mundane task people now do … but there’s a lot of misinformation and misunderstanding about what bots can and can’t do for an organization.

Last year, Deloitte’s survey of 400 global firms found that 63 percent of surveyed organizations did not meet delivery deadlines for RPA projects. For the ones that succeeded, longer-than-anticipated bot implementations delayed return on investment (ROI). And an EY study found 30 to 50 percent of initial RPA projects fail.

Yet, we’re seeing expectations for RPA spending increase nearly 60 percent over last year, with no signs of slowing down. The narrative being pushed is RPA technology can automate and transform every aspect of every business. This enthusiasm for RPA has even spilled into the investment market, leading firms to make investments based on what I believe are unrealistic valuations. The industry needs a reality check to better understand where and how RPA can best add value and plan accordingly.

Where RPA goes wrong – unrealistic expectations

Let’s start with clarifying what RPA is. RPA is the industry’s blanket term for two main types of software robots: “attended” bots that run on the desktop and assist human workers with simple, repetitive tasks; and “unattended” bots, which are designed to stand alone and automatically execute tasks in the background.

The challenge in using robotics effectively comes from the misuse of the “P,” or “Process,” in “Robotic Process Automation.” In fact, most bots are designed to automate tasks, a far cry from redesigning and automating the end-to-end business processes that are at the heart of real transformation. Used appropriately, RPA can be a very useful tool in a strategic transformation initiative. But if approached incorrectly, they can perpetuate legacy system problems.

RPA goes wrong when organizations that have been promised significant savings by using lots of bots to try to automate as much as possible, including increasingly complex operational processes. Most bots are designed to automate tasks – they won’t address the transformational need to optimize or redesign processes for the digital world. And they can’t deliver meaningful digital transformation on their own.

Business leaders that have high hopes for huge returns from massive RPA process implementations are frequently disappointed. Problems arise when organizations mistake tasks for processes, and either vastly underestimates the complexity of the processes they’re trying to automate or the time it takes to fully integrate and automate unattended bots. The result can be delayed or abandoned projects.

Poorly planned RPA implementations can create an additional problem. Unattended RPA can become another silo – another group of disconnected legacy apps in the IT stack that fragment workflow and processes. Plus, when bots are designed with custom-built logic, it makes them difficult to modify and scale. Lacking the benefit of centralized rules management and orchestration, they inevitably become elaborate, siloed, background batch processors.

RPA can’t fix bad processes – it just speeds them up. When businesses try to use unattended RPA to remedy poor processes, not only is the process not improved, but the resulting errors and bottlenecks are typically shifted down the line, creating new problems that diminish real transformation and ROI. Organizations need to shift their thinking to reevaluate existing processes and modify or redesign them to give customers and employees better experiences. 

Where RPA has succeeded – repetitive task work

By integrating attended bots into the desktop workspace to automate specific tasks, organizations create an environment where robotic and human work complement each other – automatically executing repetitive, low-value tasks, and completing work far more quickly and accurately. These desktop RPA implementations also free up staff to focus on more complex, nuanced, and high-value work that is difficult to automate. The ROI is even more significant when desktop-based attended bots are leveraged as part of overall end-to-end case automation.

The right way to think about and implement RPA

Both attended and unattended RPA has a place in digital transformation, but instead of saying, “We need RPA,” what leaders should think about is how an organization’s interconnected processes are automated from end-to-end. Real digital transformation is bigger than bots. First, it requires a design thinking approach to identify desired outcomes, and then a holistic look at sales, service, AI/decisioning, and robotics as an integrated, end-to-end workflow. Once organizations understand those, they can begin to execute on a vision of how their systems for process automation, case management, email automation, app development, APIs, and data can be orchestrated together to seamlessly deliver the outcomes that will improve customer engagement.

What’s the right way to implement RPA? Instead of trying to build a program around staff reduction and thinking about how to achieve 100 percent automation to replace a handful of FTEs, focus on some level of basic task automation for every staff member. Then focus on refining overall processes and streamlining operations. In many organizations, staff is responsible for a series of very complex workflows or decisions, making them difficult to replace with full automation. Organizations will see a much faster and higher return on investment using an attended desktop bot to automate the low-value, repetitive task work that almost every employee performs.

For example, the process of copying and pasting information from one electronic form to another can be automated with assisted desktop bots, shaving seconds or even minutes off-task time. Now multiply that by hundreds, thousands, or tens of thousands of workers. The ROI is immediate.

Embrace RPA as part of end-to-end automation

We’ve seen what happens when organizations jump on the latest technology only to find they’ve wasted time and money on an incomplete project or created yet another disconnected data island or channel. Do some honest analysis of your organization’s process needs, identify the outcomes that matter most, then develop a robotics strategy that deploys bots as part of an end-to-end operational approach. In most cases, the fastest way to earn ROI is to automate repetitive task work with attended desktop bots and use unattended RPA for the last mile of integration, automating integrations to apps, data, and legacy systems.

Alan Trefler is a visionary leader, technology change-agent, innovative philanthropist, and trusted advisor to global business executives. Alan founded Pegasystems in 1983 and has built the company into a $840 million provider of customer engagement solutions with more than 4,200 employees in 37 global offices. His book Build for Change describes a new generation of customers that have unprecedented power to make or break brands and the changes businesses must embrace to succeed in today's digital world. Alan is a member of the IT Steering Committee of the World Economic Forum and a chess Master who was co-champion of the 1975 World Open Chess Championship. A staunch education advocate, Alan and his wife Pam established the Trefler Foundation in 1996 to improve educational outcomes.