RPA Use Cases Based On A Developer’s Experience
We’ve addressed Robotic Process Automation (RPA) in several previous posts, but generally we look at RPA from a business user or C-suite perspective, which only addresses some of the important considerations around this type of automation. In this post, we’ll look at RPA from the vantage of a programmer with considerable business experience, as well as deep experience building more complex bots and using RPA tools that don’t necessarily cater to the business user.
Consider the common use cases
One of the most important considerations, from any perspective, is the use case or application of the automation in question. From our experience over the last year, two of the major things RPA is great for are automated testing and integration with legacy systems where no API is available.
In the first situation, using RPA in an automation testing setting provides benefits like delegating repetitive and time consuming tasks to bots that run systematically through an interface or process, identifying bugs and breaks throughout the system. This gets employees back in more fulfilling, value-add roles, decreasing the resource costs and increasing both accuracy and compliance.
Likewise, an RPA testing approach can capture repeated system/interface interactions and locate redundancies or inefficiencies in a given process by forcing the RPA developer to understand the complexities and business rules of the system’s workflow. In this context, the RPA developer is like an auditor, showing you where in a process or workflow the most improvements could be made.
In terms of automating processes, RPA really shines where more robust options like SFTPs or APIs aren’t available. It’s also well-suited to low/medium lift automations, where it’s possible and desirable to get a solution implemented quickly. This is because, generally, most RPA Tools provide native activities or actions as well as bot scheduling and orchestration capabilities that allow for a quick setup of project environments, facilitating automation by providing a full ecosystem right off the bat. This quick turnaround time to get any automation implemented allows for more companies to adopt RPA with little or no investment risk and makes it a major reason that RPA has become such a popular solution in recent years.
Ease of use is another reason for the popularity. In the past year, RPA has been increasingly marketed as a no-/low-code solution that can be built out and implemented by business users without the help of programmers — or at least with very little help. Tools like Power Automate, Automation Anywhere, and UiPath are examples of RPA that business users can leverage without much prior experience in programming.
While there are obvious benefits to these easier-to-use platforms, like letting those with the keenest business knowledge solve immediate problems, they are also not as flexible or comprehensive as tools that leverage more programming skill. They’re simply more limited in the scope of problems they can solve, and they can add further complexities into solving a simple problem that would otherwise require a few lines of traditional code. From the perspective of someone with development knowledge, that can be a major drawback. But the trade in speed and ease-of-use might make sense depending on the complexity of problems.
If what you need to automate demands more complexity, and you have access to developer expertise, then reaching for tools like Robocorp or our forthcoming engine, Orca, would be the way to go. Both are rooted in Python, with access to Python packages, and so the extensibility of solutions is much greater than with no-code options in which you might be limited to the set of activities established by a specific RPA tool. In addition, because these tools use more traditional software development methods, it’s much easier to employ software dev best practices, helping with testing, version control, and risk management.
Developer friendly tools like Robocorp also employ an open source framework, which means there is a fairly large and growing community of developers regularly contributing their expertise to projects. Those projects are available to anyone using the tool, so developers can leverage community-made packages as business solutions, reducing the amount of work required for developers to get solutions in place.
All this information suggests there’s no one simple answer to which RPA tool is “right”. It simply depends on your company and the processes you need to automate within it, as well as the resources available to invest in the workflow you’re planning to automate. If you don’t have access to developers and you need simple automations rather quickly, no-code RPA platforms will be your best options. If you do have access to developers, but time is still a concern or you plan on making multiple bots, then the low-code options that run on programming language engines such as Python are a win. Of course, if time is less crucial and you have a team of programmers, developing in a programming language such as Python is a great option. It just takes more time and expertise, but you should end up with a more risk-averse solution and a more complete overall product.
Regardless of your use case, taking the time to explore what options you have will always be time well spent. We’ve worked with and experienced a number of RPA software, and are working on creating our own, and one thing we consistently land on is that the decision on which tools to use is almost always determined on a case-by-case basis, taking all the factors above into consideration. The one thing that is certain is that RPA developers of all levels are needed in all stages of the industry.
