Introduction to RPA: Process Automation Software, best practices, and the right approach

A simple, comprehensive introduction to the topic of robotic process automation (RPA). Why should you automate processes? How does process automation work? What examples and best practices are available? 

Why should organizations automate processes?

Repetitive, tedious processes are time-consuming and costly. They tie up the resources of well-trained and well-paid professionals. Such manually implemented processes slow companies down. That’s why companies should automate their processes, because the external pressure they face is increasing: more competition and a rapid pace of innovation, the increasing shortage of skilled workers, rising customer expectations, more compliance requirements, and crises like Covid-19. When companies automate processes, these processes become more stable and more resistant to external influences.

Process automation offers very high potential:

  • While the degree of automation in industrial production has increased by a full 75 percent since 1980, in the office it has increased by only three (!) percent over the same time period (source: study by the Dutch Vrje University).
  • The McKinsey Global Institute estimates that 30 percent of all office activities can be automated.

RPA software: How can processes be automated?

Organizations can automate processes with modern technology such as robotic process automation (RPA). When companies automate processes, they speed up and improve processes – by high double-digit percentages.

The orange and yellow robotic arms are a common image for modern industrial production. However, our lives are becoming increasingly digital as well. And with it, a new era of automation is also dawning: automating digital processes with the help of software robots.

Software robots are basically computer programs that automate rule-based, recurring processes by executing them around the clock. In doing so, they mimic the interaction of humans with computers. This means the software bots perform routine tasks such as searching for information or files and documents or matching data in different documents and systems. They work via the graphical user interface of the programs – i.e. the screen.

The generic term for the technology behind software robots is “software robotics”.

Process automation – what are the concrete benefits?

What are the benefits of process automation? With robotic process automation, organizations can accelerate and improve their digital processes, while making them more stable at the same time. This is because software robots work around the clock, without downtime and without errors, even when performing monotonous activities. Moreover, everything that software or RPA bots do is transparent and traceable. This traceability also improves compliance, i.e. the ability to manage and prove adherence to internal and external specifications.

Global consulting firm Deloitte describes the added value of process automation this way: “The benefits of RPA adoption are significant.”

In its 2018 Global RPA Survey, Deloitte quantifies the benefits quite specifically:

– 59% lower costs

– 86% higher productivity

– 90% better quality/accuracy

– 92% compliance improvement.

Moreover, Deloitte adds that in most cases, the return on investment for robotic process automation is achieved in less than twelve months. This great potential made the technology the fastest-growing segment in the enterprise software market in 2018 (source: Gartner, June 2019).

Process automation – here’s how it works

Compared to other technologies, process automation works exclusively via the graphical user interface (GUI), i.e. via the screen. Companies can therefore automate processes without having to laboriously integrate the technology into the IT landscape. In fact, no intervention in the IT landscape is required at all.

RPA examples and best practices: Who is RPA most relevant for?

Due to the relative ease of use, all types and sizes of companies in every industry and department can automate processes and drive their digital transformation – because without digitalization, there is no automation.

Since the software is relatively simple, employees can automate processes themselves after a brief introduction, completely independently. This quickly creates critical mass and is a big step towards an automated organization.

Process automation in business departments

According to the forecasts of independent market analysts, more and more companies will automate processes because software robots have a wide range of applications: In addition to the IT department, which still automates processes most frequently (as of 2018/2019), their use is also increasing in business departments. Controlling and Finance & Accounting, as well as Production, Customer Service, Human Resources, and Procurement are departments that are increasingly automating processes.

In Controlling, companies can automate numerous processes, for example in accounting, tax returns, receivables management, and financial statements.

Organizations can also automate many processes in purchasing, such as procure-to-pay processes, contract and supplier management, and master data management.

Process automation in SMEs

Process automation is particularly interesting for SMEs because the barriers to entry for the technology are very low:

  • No complex IT implementation necessary
  • Fast implementation time of a few weeks
  • Fast return on investment (usually within a few months)
  • Manageable costs.

The German government is also pushing the technology associated with process automation, even going so far as to launch a separate research project so that small and medium-sized enterprises (SMEs) can also benefit from Robotic Process Automation. After all, process automation is an important driver for digital transformation. That’s why the German Federal Ministry of Economics and Technology is funding a project called “RPAsset”, which will develop an implementation strategy for process automation, including a guide for SMEs, by 2021. German process automation specialist Servicetrace is also actively involved in RPAsset.

Industry best practices for RPA – a few examples

Retail: Link Retail Story

Pharma: Link Merck Story

Telecommunications: Link T-Mobile Story

Finance: Link MLP Story

How to automate processes completely or partially? What do “attended” and “unattended” mean in this context?

There are basically two ways to automate processes:

  • Unattended automation: full automation. A software robot executes a process or a process substep completely independently. There is no intermediate step or interaction with a human.
  • Attended automation: supervised (or partial) automation. A software bot acts together with a human who, for example, makes a decision during the process, on the basis of which the software bot then continues to work.

When combined with artificial intelligence such as machine learning, companies can increasingly automate processes where bots learn and make decisions themselves (if this is desired).

Automating processes versus classic business process automation

Classic business process automation (BPA) is not a new technology. With BPA, companies focus on optimizing existing business processes. A process that does not work or does not run optimally is reengineered. In other words, BPA is an approach to increasing process efficiency and value creation across the enterprise.

In contrast, robotic process automation (RPA) is a technology in which software robots operate existing systems, mimicking the actions of human employees. No change to software solutions – such as ERP – is required. However, BPA and robotic process automation are not mutually exclusive. They can also be used in combination.

Criticism of process automation

More and more companies want to automate their processes. Alongside the hype, however, there has also been criticism, especially in the beginning. Proponents see the automation of processes as an important driver for digital transformation. For companies to be able to automate processes, they must first be digitalized.

There have also been a few critical voices: to the contrary, automating processes inhibits digitalization, they claim. And RPA in particular is only a bridging technology. This means that process automation only serves to keep technically outdated systems (legacy systems) alive. These complex, business-critical, legacy systems frequently exist in corporations such as banks and insurance companies. They usually run very stably, but are outdated from a technical point of view. Replacing them is very cost-intensive, however. In addition, there is a lack of skilled personnel for these legacy systems. And this is where process automation comes in. Software robots are sometimes used to operate these systems and thus keep them alive longer – at least that is the criticism.

But the (unsustainable) use of a technology is not the fault of the technology itself, but of the user. In addition, a survey confirms that 78 percent of banks that operate their legacy systems with software robots do so with the aim of better planning their often very complex replacement.

Process automation with artificial intelligence – extensive new possibilities

By using a combination of robot-controlled process automation and artificial intelligence, companies can automate processes even more extensively. The potential usage spectrum and the possibilities are increasing massively. By using machine learning or even natural language processing (NLP), for example, they can automate even longer and more complex processes. NLP is the automatic recognition and processing of natural language. Read an example of process automation in customer service using an intelligent chatbot and RPA here.

In this respect, the process automation story has only begun its development. But there are already names for it: from Intelligent Process Automation to Cognitive Process Automation or Hyperautomation.

Which processes are suitable for automation?

Areas of application for process automation can be found across all industries and departments. Examples of specific starting points are activities where employees have to enter data manually, copy data back and forth between systems and/or documents, check data, consolidate and prepare data from various sources, e.g. for reports, fill out forms or registrations, etc.

Select and evaluate processes

Characteristics of processes that are suitable for automation:

– Transactions that are prone to repetition

– Standard processes

– Logical decision trees (if x, then y)

– Processes that are highly susceptible to careless errors

Automate processes: Initial assessment of automation potential

– Are the data inputs structured?

– Is the volume or value of transactions high?

– Can the process be mapped with clear, logical rules?

– Is the process prone to repetition?

– Is the process prone to errors and rework?

– Does the process/systems change infrequently?

Model and document processes

Identifying suitable processes is already the first step on the journey to automation. It is advantageous if processes are already documented. Otherwise, they will be modeled or documented in the next step. With the help of Servicetrace’s automation platform XceleratorOne, companies can implement process documentation directly in BPMN, the common standard for business process management.

Design, implement, and monitor process automation

Based on this process model, users can then create the automation workflow. After testing the created workflows, users then configure the operation and can make the automation live. During operation, the software bots are of course monitored and optimized if necessary. In this respect, it is a classic project process.

Process automation – success factors

Automating processes is relatively simple compared to other technologies. This promotes a high level of acceptance among employees, who can automate processes on their own. Incidentally, once a process has been automated, employee satisfaction often increases afterwards, because employees are relieved of monotonous, error-prone tasks. Automating processes does not necessarily mean job cuts. According to HfS Research and KPMG, only 7 percent of companies that automate processes aim to cut jobs as a result.

If many people throughout the company automate processes, however, these developments should be controlled, which means transparency and a control option for process automation are needed at the management level. This applies to individual projects as well as to all automation initiatives in the company.

This is why analysts such as Gartner warn that many organizations underestimate the complexity of automating processes. This lies primarily in successfully automating processes throughout the entire company and thus achieving a high return on investment.

Important success factors for process automation are therefore:

  1. Transparency and evaluation of projects and their progress
  2. Governance of individual and all projects in the company
  3. Collaboration: successful, effective cooperation between different experts
  4. High scalability of the implemented technology for process automation

Process automation – a look into the future


In the beginning, the focus of automating processes was often on relatively simple activities and processes. This corresponded more to a pure automation of tasks (task automation). However, this also meant that there were often only individual silo projects and the successes were limited. With today’s technology, however, companies can automate longer, more complex processes – especially when RPA is combined with other technologies. Gartner has coined the term hyperautomation in this context.

Enterprise Process Automation:

For Servicetrace, a German software robotics company, the focus has long been on automating even complex and long processes. Servicetrace’s goal from the beginning has been to develop an automation platform that enables organizations to successfully manage all aspects of process automation. On one hand, this includes the complete lifecycle of RPA, including all stakeholders. On the other hand, the platform is designed to introduce process automation throughout a company. This also means bundling and controlling all projects and software bots centrally on one platform. Servicetrace thus offers a truly comprehensive, neutral platform for process automation and coined the term Enterprise Process Automation.

Automate processes with an Enterprise Process Automation platform – the benefits:

  • Orchestrate all RPA initiatives across the enterprise
  • Also control RPA bots from other vendors – all on a single platform
  • Automate complex processes
  • Scale dynamically
  • Continuously increase your organization’s level of automation

Intelligent Process Automation / Cognitive Process Automation:

Intelligent Process Automation (IPA) and Cognitive Process Automation describe the combination of robotic process automation with intelligent components. This includes technologies for intelligent vision and recognition of texts, images, and characters (optical character recognition, OCR), as well as machine learning and natural language processing, NLP.

The Capgemini Research Institute estimated that large organizations achieved around $500 billion in cost savings through IPA in 2020.

Register for our newsletter: Robotic News in your mailbox

Stay up-to-date with us, register here for the Servicetrace Robotic News