After helping several clients select a robotic process automation (RPA) application. It now looks perhaps the easiest business case ever to be made.
Case: The company had a set of over 1 million records, both digital files and scanned images, and it was seeking a way to quickly scan these files for information about when invoice payment was due. The RPA system took 3 seconds to scan a test bank of 400 records. A similar exercise using a group of people took four to five hours and the RPA system was much more accurate. In addition, the RPA system even uncovered additional information useful to the company, such as certain clients were charging late fees before an invoice came due.
It is stories like these that have made RPA the darling of the automation industry. Forrester Research has projected that the robotic process automation market, while only $250 million in 2016, will grow to $2.9 billion in 2021. Unfortunately, dazzled by successful case studies and impressed by numbers such as Forrester’s, buying companies are sometimes inclined to jump straight to RPA as a solution when in fact they need a more holistic approach to automation, such as business process management, or BPM.
It is easy to understand the confusion. The two methodologies are highly complementary, but have very distinct purposes and scope. BPM and RPA are really two sides of the same coin. Both are valuable tool sets that management can leverage to drive operational excellence within their organizations.
The technologies are not mutually exclusive, Indeed the best results are often achieved when both technologies are deployed to drive true transformative changes.
The Difference Between BPM and RPA
While both technologies deal with automation and both are being increasingly used in digital transformation projects, one cannot serve as a substitute for the other.
BPM is holistic and big-picture oriented. This technology employs automation in appropriate places, but it also permits integration with other, external technologies and offers a means to optimize and increase the efficiency of humans involved in the process.
By contrast, RPA is application agnostic and most useful for repetitive tasks. RPA technology is based on low-level events or interactions such as mouse and keyboard use or web page scraping and it is relatively easy to do everything that an application with a user interface, running on a computer, can do.
There is another key difference between them, which is the period required to generate the desired returns on investment. Today’s executives are under relentless pressure to deliver transformation and instant results. In this regard, implementing RPA is akin to process excellence on steroids.
How Is RPA Used?
There has been strong take-up of RPA in the finance sector, as these businesses often have a great deal of data that lives across legacy systems — a prime use case for RPA. There is a lot of data, there is a lot of process at a scale that when automated, you can see the business impact in terms of savings, efficiency or lowered risks and business outcomes can be easily identified.
RPA’s true worth, though, will be developed when artificial intelligence is added to these applications. Adding AI to RPA will free it from an exclusive focus on rote tasks. AI will account for an increasing portion of the digital workforce, and in the end, RPA will be a small fraction of the overall AI cubicle market spend.
For now, though, RPA is mostly being used for such routine chores as scanning records.
How Is BPM Used?
By contrast, BPM is used to automate end-to-end processes. It does this through a variety of ways:
- It identifies and clarifies all the administrative duties around the process.
- It outlines the work that people accomplish, i.e., their workflows.
- Then it identifies the areas in which those workflows can be optimized.
There is a standardization and documentation aspect to it, but BPM also is key to engaging people to work together. Ultimately, BPM is about mapping a process, tracking its performance and looking for opportunities to improve that performance.
How to Choose Which One You Need
With these nuances between the two technologies explained, it becomes easier to decide which is most relevant for an automation project. Use RPA to automate everything that can be automated without integration and is not subject to frequent change. As for BPM, delegate all human integration — specifically, action requiring judgement — to this technology. Also, use BPM to integrate with third-party systems.
More likely a complex project will call for both technologies. A typical use case of both BPM and RPA would be BPM automating an entire business process — such as the process at an insurance company to pay a claim — while RPA would be used for one or two steps within that process, such as moving records from one database to another. In the past, BPM would have routed that work to a human user. Now it can be routed to a robot which has been trained to do the same task