Click to learn more about author Tom Thaler

Robotic process automation (RPA) is the fastest-growing segment of the global enterprise software market. That said, it’s not surprising that 41% of all operational process improvement programs focus on process automation and that RPA skills rank in the top three most important skills in achieving operational process excellence. Furthermore, McKinsey estimates that about 50% of today’s workforce activities could be automated and 10% of those are automatable at a rate above 90%. All of this underscores the importance and the great potential of RPA for the digital transformation of enterprises.

Despite this, we are at a crossroads. Technologies follow a typical hype cycle and, for RPA, we are actually on the way from the “peak of inflated expectations” to the “trough of disillusionment.” The simple reason for that is that RPA is considered in isolation. Pureplay RPA vendors position the technology as “the” solution for business process automation and tend to oversell the technology as a silver bullet solution for all integration and automation problems. Although this might be true from a purely implementation standpoint, the real challenges – especially in large enterprises – are completely ignored.

This leads to a number of issues that are primarily caused by a missing management approach. For example, 27% of RPA-applying organizations are confronted with a lack of understanding and capturing the opportunity and 48% grapple with talent and employee resistance. Because of this, 70% of change programs fail to achieve their goals and only 4% of RPA-enabled companies reach a 50-robot scale in their automation projects.

How the RPA Industry Will Shift in 2020 and Beyond

Consequently, 2020 is a crucial year that will lead to a major shift in the industry. Following Gartner’s hype cycle model, this will prove to be the year where the industry will follow the “slope of enlightenment” and aim to reach the “plateau of productivity.” This shift will not be technology-driven; instead, this will come about as more enterprises start to understand how to capture impact, manage complexity and gain commitment from top management.

1. Capturing Impact: Companies that successfully implement RPA start with a top-down assessment of the automation candidates. They have a clear understanding of their ambitions and how to predict and capture the impact. The focus is no longer on simple back-office processes and is instead increasingly shifting towards core competencies and customer relevance – in a recent OPEX survey, customer satisfaction passed traditional measures of success such as increased throughput and efficiency. Process mining tools will make it easier to spot opportunities for automation based on relevance criteria like this, but also deliver the insights for a post-assessment of the achievements.
2. Managing Complexity: A holistic automation approach makes end-to-end use cases available for automation. Handling that requires a specialized team at the core – otherwise known as a center of excellence (CoE). These competence teams act as enablers to the business and manage the complexity of the robotic process landscape. Following a structured approach, they produce new insights that are leveraged for continuous process improvement initiatives – something that is addressed by process management respectively enterprise business process analysis (EBPA) tools today.
3. Gain Commitment: Successful RPA projects have a strong buy-in of the top management, who drive the project across the organization. Employees must be aware of the project’s importance and trust in the direction of the transformation. The key is to ensure employees that automation is used to augment rather than replace the workforce – and that automation leads to better jobs, not fewer. This requires management to engage all parts of the business through consistent and clear communication on what is being automated and how people remain valued and part of the company’s plans.

Addressing the Key Phases of the RPA Lifecycle

As RPA continues to develop, it will continue to mature into a fully integrated intelligent automation platform that will enable businesses to massively scale the automation of business processes. As this growth occurs, it will be essential for enterprises to ensure that their RPA platforms address all six phases of the RPA lifecycle:

1. Discovery: Process management and process mining are used to identify and prioritize the most business-relevant automation opportunities. Both deliver insights like bottlenecks or weak points in the end-to-end processes, either on a conceptual data basis or based on the observed process behavior. This effectively allows organizations to localize problems with the biggest business impact and create the automation pipeline.
2. Analysis: After localizing the problem, the second phase is about scoping it. Discovery bots are applied to selected employee computers through intelligent task mining, which generates an unbiased view on the as-is processes and applies machine learning to find patterns in employees’ interactions with the application systems they use to execute their business processes, which is the basis for the robot implementation. From there, the solution identifies and recommends processes that can be improved based on any issues or bottlenecks it observes. Since up to 70% of RPA project resources are spent on pre-automation, it’s crucial to find a platform that can automate most parts of this phase to speed up RPA progress with fewer resources.
3. Design: It is then important to comprehensively document the robotic process landscape in detail, which includes systems used, ownerships and activities. Using this information, the platform should be able to reconstruct the observed process behavior as a visual representation to design the desired automation solution. Another important aspect the platform should have is an initial view on improvement potentials, which is key to preventing a poor process being automated.
4. Development: During this phase, the robot should be enriched with detailed step-by-step operating procedures. These can be automatically generated by task mining. The easiest implementations can be developed through low-/no-code, which can be used to both configure and refine the robots.
5. Deployment and Operation: The robotized process landscape is rolled out and the robots are deployed to the production environment. Deployment means two things here – the technical deployment of the robots and the rollout to all relevant stakeholders. Management must communicate key details to employees so they can understand what is happening – including what is done by robots and how the robots are doing it – since this is crucial for a broad acceptance within the organization. In this phase, the robots are continuously monitored and maintained for operability and compliance. This is especially critical in regulated environments such as chemistry or healthcare.
6. Benefits Capture: Finally, process mining and dashboarding capabilities should help to collect all business-relevant data, visualize corresponding key performance indicators and communicate the RPA success story to top management to ensure buy-in, ownership, and budget for growth.

GOL Linhas Aéreas (GOL), a Brazilian airline operating primarily in Latin American countries, is one organization that has truly reaped the benefits of RPA. Since GOL’s founding, one of its biggest challenges was its limited ability to continuously review and adapt schedule plans in response to operational requirements.

Prior to implementing RPA automation, processing 1,300 flight changes required 20 working days, as updates were made in two asynchronous systems – the route network management system and the reservation and ticketing system. Now, with RPA, GOL can process the same workload in only one day and is able to plan flights three months in advance – increasing ticket reliability and availability for customers.

Looking at another case, one of the largest energy suppliers in the world worked on an RPA project to automate its financial transactions. The company leveraged process mining to identify key factors that contributed the most to automation breaks. Once identified, these key factors were eliminated step by step to resolve weaknesses. This year, the company expects that 45% of all financial transactions will be performed automatically.

While the benefits will vary from industry to industry – and even organization to organization – it’s important to carefully consider all business processes and integration challenges when selecting automation candidates. Once a clear project has been identified, it’s important to secure support from top management, set up a CoE to manage the project’s complexity and ensure that the right platform is in place to design the right solution.