Repetitive, monotonous and dull tasks eat up a lot of time in everyday life and are usually rather annoying and boring. But that doesn't have to be the case, because precisely such tasks can be automated. Robotic process automation (RPA) helps to avoid having to press the copy-paste buttons prayerfully every day or laboriously enter information into a system by hand.
Robotic Process Automation (RPA) is not a T800
Robotic Process Automation is not about physical robots replacing humans in a company. RPA refers to the automation of business processes by digital software robots, so-called RPA bots. RPA is particularly comfortable with repetitive, rule-based work tasks that rely on digital data. These tasks include queries, calculations, creating and updating data sets, filling out forms, creating reports, cutting and pasting, and other high-volume, transactional tasks that require moving data within and between applications.
RPA aims to improve efficiency, increase productivity and save money by supporting - or replacing altogether - the routine and error-prone digital processing tasks still performed by humans in many organizations. Done right, RPA not only saves companies time and money, but also frees up employees to focus on higher-value activities.
A software robot can be used around the clock, 365 days a year, takes no vacation and is never sick. While a human being works 220 days a year and a 40-hour week for 1,760 hours, a bot can theoretically work 8,760 hours - and at a higher speed. According to management consultants Deloitte, 60 to 80 percent of resources can be saved in individual processes. For standardized back-office tasks, the average is usually 25 to 40 percent.
How does a software robot work?
A software robot operates at the interface and user interface level by mimicking the keystrokes and mouse clicks of human workers and completing the task in a manner similar to how employees log in to programs and apps, enter things, perform calculations, and log out. To do this, integration scripts are being developed to retrieve information from systems and transfer it to other systems. The scripts are designed to replicate the actions of a person interacting with these systems or documents, for which there are typically no effective APIs. An RPA tool works by mapping a process that the software robot can follow through computer paths and various data stores, allowing RPA to operate in place of a human.
For example, a software robot recognizes when someone repeatedly enters the same things and can fill out the forms itself. Robot-controlled process automation is thus used for surface automation. The goal: as little human interaction as possible.
Unlike a human worker, however, the bot does not need a physical screen to complete the task, but performs the process steps of the task in a virtual environment.
Where does Robotic Process Automation come from?
Process automation with software robots has a long pedigree dating back to the introduction of macros in the 1950s. The term itself and the core technologies underlying today's products are of more recent origin. In the early 1980s, banks developed data scraping applications to capture data from various financial data services such as Reuters. In subsequent years, more sophisticated tools were developed to cut and paste data from mainframe terminal applications into more modern web applications. The true origin of RPA is therefore difficult to determine, especially since automation as a principle has been the driving force of IT for decades.
The spiritual forefather of the term Robotic Process Automation is undoubtedly Phil Fersht, founder and principal analyst of HFS Research, who first used the term "Robotic Automation" in the blog post "Greetings from Robotistan, outsourcing's cheapest new destination" in 2012. A little later that year, Pat Geary, RPA chief evangelist and marketing director at Blue Prism, added the word "process" to describe a new category of automation that would complement business process outsourcing (BPO) and business process management. Ultimately, RPA is based on key technologies: screen scraping, workflow automation and artificial intelligence. Screen scraping is the method of collecting and translating screen display data from one application so that another application can display it. This is typically done to capture data from a legacy application and display it through a more modern user interface. It is sometimes confused with content scraping, which uses manual or automated means to rip content from a website. Very often, screen scraping refers to a web client that parses the HTML pages of the target website to extract formatted data. Workflow automation is a technology that uses rule-based logic to automate manual work such as data entry. By using self-operating processes that perform manual tasks, it can help the company save time and money, reduce errors, and increase productivity.
Artificial intelligence has the ability to allow computer systems to perform tasks that normally require human intervention and intelligence. While each of these techniques is an advance in its own right, the development and deployment of robotic automation and its ability to combine, refine and redefine certain elements of all these technologies actually makes RPA a very powerful technological platform.
How do companies benefit from an RPA bot?
The use of RPA brings companies numerous advantages:
First, there is no need to waste human brain on requirements that can be executed much better, faster and cheaper by software robots, and thus are often completed with higher quality. Employees do not have to deal with simple and monotonous things, but can pursue the challenges that really drive the business forward. For example, there is more time to take care of customers or to drive innovation. This allows employees to make higher-value contributions to areas such as customer satisfaction, innovation and scalability. Set up correctly, bots perform the process the same way every time and don't get tired, reducing errors and inconsistency.
Second, Robotic Process Automation improves process time and accuracy by eliminating human errors and the need to correct them. In this way, software robots pay heavily towards a better customer experience, higher NPS (Net Promoter Score) and lower customer churn.
Third, it offers companies a way to automate parts of critical business processes without having to replace the costly legacy systems that support them, because RPA works at the user interface level. Rather, RPA is also quick and easy to implement as well as flexible to customize and integrates well with legacy systems without complex and costly (interface) programming.
Fourth, RPA increases operational flexibility, improves the ability to audit processes, and provides insights that can be used to identify, analyze, and proactively improve existing problems in processes. In addition, software robots lower human interaction with sensitive data, reducing the potential for fraud - so it's a win-win situation.
RPA use case: automate personnel management
Taking a look at the HR industry, there are many processes that can be automated, especially in HR. From simple payrolls and reports, to talent acquisition and recruitment, to employee onboarding or an overview of the number of workers needed and further data management. HR departments can benefit from digital support and focus on other challenges, such as conducting hiring interviews or resolving conflicts.
Specifically, RPA bots can take over resume screening, for example, and compare the information with the requirements from the job ad. The best candidates then receive an invitation to an interview, while the others receive a rejection notice. Furthermore, offers, for example for freelancers, can be filled out according to fixed templates. If a new employee has to be trained, it is necessary to integrate new information into the system, such as mail addresses, access rights or passwords. Process automation helps here and automatically creates new user accounts. Travel and expense reports can also be automated, saving a lot of paperwork. The employee database thus remains much clearer and error-free. To ensure the productivity of the company, the presence of employees can be tracked.
In summary: If a company wants to grow quickly, process automation can save a lot of time - provided the right strategy is in place.
Typical tasks of an RPA bot
Almost all recurring activities can be automated by one or more software robots:
Launching and using various applications
- Open emails and attachments
- Log in
- Move files and folders
Integration with enterprise tools
- Establishing a connection to system APIs
- Reading and writing in databases
- Scraping data from the web, including social media
- Follow logical rules like "if / then" rules
- Perform calculations
- Extract information from documents
- Entering information into forms
- Extract and reformat data into reports or dashboards
- Merge from multiple sources
- Copy and paste data
Attended or Unattended - that is the question here
Attended (partially automated) and unattended (fully automated) RPA bots are two modes that can operate with or without human intervention. RPA bots can operate in both attended and unattended modes. Usually focused on front-office activities, attended bots are created in a situation where it is not possible to automate the entire end-to-end process. In such cases, the RPA bot is triggered by system-level events and works in conjunction with human workers. Unattended bots work independently without human intervention. They are designed to take over the time-consuming manual processes and execute them in the background without requiring any input or intervention from a human worker. An unattended software bot can execute repetitive, rule-based requests - often back-office activities - that follow a predefined pattern or series of steps, the same way every time.
As a digital worker, an unattended bot is typically triggered remotely to run behind the scenes. Unattended bots can be triggered automatically by an event or launched at a specific time to run in a batch mode model around the clock, for example, batch operations. Unattended bots can extract unattended customer data from a spreadsheet and automatically enter it into the required application. Predominantly, unattended bots are used in the back office where large amounts of data are collected, sorted, analyzed and disbursed.
When it's impossible to fully automate a process from start to finish, humans and software bots can combine their strengths to get the job done more efficiently. In other words, attended bots are created for processes that require human input. Typically, the attended bot is triggered manually and operates locally on the employee's computer, executing the routine, rules-based components of the process while the employee focuses on work that requires expertise, human judgment, empathy, creativity and/or strategic thinking. Often, attended bots are used in more complex, longer-running or front-office processes where they act as virtual assistants. For example, an attended bot might work with a call center agent to transfer customer data from one system to another while the agent continues to speak with the customer. This increases efficiency and creates a better customer experience.
Whether an attended or unattended bot is the right choice depends on the time perspective and certain contextual characteristics. As a rule of thumb: attended bots for short-term efficiency and unattended bots for longer-term strategies. A more nuanced decision regarding an optimal automation solution can only be made by carefully examining the contextual specifics.
Which business processes are worthwhile for RPA?
RPA is not suitable for every business process. Automating a mission-critical, long-running and complex business process is typically a task for IT using API-based automation. In fact, in its July 2019 Magic Quadrant on RPA, Gartner noted that the word task in the RPA acronym would be more accurate than process.
Processes best suited for RPA have high transaction throughput of structured digitized data with relatively fixed processing paths and/or user interfaces that do not change frequently and are rule-based. RPA tools work best when they have direct access.
Here are some characteristics that make a process suitable for RPA:
- It is high volume and repetitive.
- It relies on structured digital data.
- It has clear business rules and little or no exception rates.
- It is prone to errors when human labor is used.
- It is time-critical or highly seasonal.
Business processes and their associated systems should be stable before using RPA. RPA is well suited for simple processes that run at high volumes.
In general, almost all digital business processes can be automated. Bots can take over repetitive processes. RPA can start with simple requirements, such as filling out a template, but can end up taking on significant, more complex challenges in accounting, bookkeeping, human resource management or logistics. For example, an RPA bot can automatically record employee time, monitor inventory levels and merchandise shipments, or generate billing statements. It can also be helpful for customers if, for example, signatures are checked automatically or approvals are created independently. Back-office processes in business systems thus become much more efficient.
Robotic Process Automation can be used in a variety of different areas.
Here are just a few of the many application examples:
RPA in Finance
Financial services companies have to write mountains of profit and loss reports. Most of the time, employees type everything in manually. An error-prone, time-consuming and monotonous process that an RPA bot can do much faster and more efficiently after just one mouse click.
RPA in Customer Care
Another starting point for process automation is customer centers. Employees often have to put customers on hold in order to gather information from various systems. A robot can easily take over such processes while the employee continues to talk to the customer.
RPA in Insurance
Insurance is another industry with many repetitive processes that are well suited for Robotic Process Automation. Although insurance lags behind banks in RPA adoption, some insurance use cases are already being successfully automated. These include various claims processing operations, policy administration, underwriting, processes, regulatory compliance and more.
RPA in Healthcare
For healthcare organizations, the accuracy and compliance of all internal processes is essential, as the health and well-being of patients depend on them. For this reason, large hospitals in particular are using Process Automation to streamline information management, insurance claims processing, payment cycles, prescription management and other processes. This results in fewer errors and a better patient experience.
RPA in Retail
Retail companies are investing heavily in Process Automation to improve the customer and employee experience. Popular applications of RPA in retail include fraud detection, inventory and order management, customer feedback processing, and customer relationship management.
Cross-industry RPA use cases
Processes predestined for RPA can be found in every industry.
Here are just a few of them:
- Human Resources: HR processes involve a lot of information management and standardization across many systems and applications, making them well suited for automation.
- Finance and Accounting: This area offers many opportunities as the processes are mainly rule-based and require a high level of accuracy and speed. Some common RPA use cases in this area are order management, billing, accounting and reconciliation.
- Procurement: Due to the structure of the documents used in the processes, procurement is an excellent choice. RPA applications here include invoice processing, purchase order management and contract management.
Are there security risks when using RPA?
RPA also brings security risks, but not the ones that are widely expected. So far, there have been no major data breaches in RPA implementations. That's largely because the technology functions as an upper layer and is not integrated with the lower layers where the data resides. Nevertheless, there are security risks. These include fraud, inappropriate access to sensitive information, the potential to compromise system availability and continuity, and abuse of administrative privileges.
The use of RPA requires change communication
Challenges posed by change, fears of job losses and media coverage can result in potential resistance and even sabotage automation plans. From the employee's perspective, automation is not always a positive issue. Many fear for their jobs. Generally, however, this is not the goal that companies pursue: The focus is more on error reduction and an increase in quality and productivity. The crucial success factor is therefore communication as part of change management.
Companies should therefore
- Inform employees early and transparently about the purpose and goal,
- Reduce prejudices and fears and highlight the need and benefits,
- actively accompany individual phases of the change process and mobilize employees for automation.
In this case, communication is not a one-time task, but will accompany the company throughout the entire project. It is therefore advisable to develop standardized procedures and formats and to set up specific communication channels and systems so that the project can be communicated successfully and efficiently.
How can employees benefit from Robotic Process Automation?
Incidentally, process automation using RPA does not mean that people lose jobs. On the contrary: They are relieved by the elimination of tedious routine tasks and can use their know-how specifically for more important challenges. In addition, more and more service providers are offering advanced training and recognized certificates. This gives companies the opportunity to train their freed-up employees as Automation Experts and use them internally for the analysis and optimization of various processes. Incidentally, this saves companies the cost of expensive consulting firms and external RPA developers.
Here is a small list of possible job titles and education or training in the field of process automation:
- Automation Starter
- Automation Agent
- Automation Leader
- RPA Developer
- Cloud Automation Engineer
- Citizen Expert
Moreover, the future of process automation is not only reserved for companies. Individuals who want to change their career or simply want to specialize will also have the unique opportunity to gain a foothold in a still young but rapidly growing field and make a career as an RPA expert.
The future of the RPA industry
The market for Robotic Process Automation is growing continuously. Numerous companies are already using the technology. In a study of 141 companies surveyed by the consulting firm PwC, more than half said they were already working with robotic process automation. More than a third of the companies that do not yet use bots intend to follow suit soon, according to the study. The most common areas of use are controlling, reporting, quality assurance and data validation. In some cases, companies even use RPA for customer contact (chatbots) or price negotiations.
According to a study by the consulting institute BARC, the market leaders in the area of robot-assisted process automation are the companies UiPath and Automation Anywhere. Pegasystems, Blue Prism and Kofax also offer RPA solutions. According to a market survey by consulting firm Gartner, the U.S. has so far dominated the software market with more than 50 percent market share. The number of RPA providers is growing continuously.
The market for process automation is still very young and, with a market share of just under 850 million dollars, relatively small. However, it is one of the fastest growing software areas in digitization. According to a study by Gartner, it grew by around 63 percent in 2018 and the market volume is even expected to triple by 2024.
Impact and technological outlook
There is no shortage of opinions about what the future of RPA will look like. According to the current state and the latest trends, different types of development can be predicted. What does the future of RPA look like?
Process Automation and its impact on workplaces will accelerate the adoption of RPA technology in enterprises as more companies reap the rewards of successful implementation of RPA and AI technologies. In addition, there will be a convergence of digital and human workforces. Software robots will augment the work of humans by taking on the most tedious, repetitive problems. This will lead to the creation of a digital workforce that works closely with the human workforce.
In addition, there is growing interest in new technologies that combine RPA and Artificial Intelligence. Through the use of Artificial Intelligence (AI) and Machine Learning (ML), Robotic Process Automation is becoming Cognitive Process Automation. These tools can self-correct and learn from mistakes to continuously improve processes. Intelligent Automation offers much greater possibilities by automating judgment-based processes with unstructured and non-digital data.
Overall, technologies such as Artificial Intelligence, Business Process Management, Optical Recognition (OCR), Machine Learning (ML) and others will be increasingly used in combination with RPA in the future. Although the popularity of RPA technology is steadily increasing, analysts agree that companies that rely solely on RPA are falling short because, although software bot is an excellent way to start automating, RPA offers limited opportunities for scaling.
The future of technology lies in the convergence of RPA with Machine Learning, orchestration and advanced analytics, providing scalable solutions for enterprise-wide digital transformation.