This article by Wayne Ariola was originally published on cio.com
And so began Gartner analyst Tom Murphy’s keynote at Tricentis Accelerate—a conference dedicated to exploring “Automation @ the Speed of Change.” Now that many organizations are adopting Robotic Process Automation (RPA) and showcasing quick wins, we’ve reached an inflection point. The focus has shifted to scaling RPA, and this exposes the danger of moving full speed ahead with automation you simply can’t control.
So how do you avoid hitting that brick wall and bringing your newly-automated processes to a screeching halt? Murphy offered a number of recommendations that apply to organizations just starting off with RPA, as well as early adopters who have plunged into the so-called trough of despair.
Look beyond the “low-hanging fruits” for long-term ROI
Repetitive tasks that create a “people-process” bottleneck are obviously prime candidates for RPA. But although RPA use cases are typically demonstrated with functions like HR and insurance, almost every professional is held back by mind-numbing tasks that make them hate their job. For example, even a top-level analyst like Murphy must spend significant time closing out Siebel records after each and every client interaction, every day.
Murphy offered the following 4-step approach to maximizing RPA’s business impact:
- Identify: Look for high-volume repetitive tasks that involve a lot of keying, have high error rates, and slow down critical processes.
- Validate: Consider the input and the trigger events you might want to go after. Where do you need validation? Are the related processes and systems relatively stable? Of course, you’ll need a way to identify and manage change (more on this below), but beginning with your most dynamic processes could create an uphill struggle from the start.
- Quantify: How do you quantify the value of automating this task? How do you account for task complexity as well as frequency? What are the current error rates and what’s the real value of reducing them? Typos in transcribing call notes is one thing; transposed digits in financial records is another.
- Prioritize: Don’t start with your most complicated process, even if it’s stable. Aim for projects that are small, simple, and provide compelling proof points. Projects that lie beyond the boundaries of other automation approaches are great candidates for showcasing the value of RPA. As you prioritize, consider what’s going to resonate most with your business (improving efficiency? Increasing capacity?) and how you will demonstrate business value.
Test automation is a great bridge to RPA
According to Murphy, many organizations have successfully used test automation as a bridge to RPA. There’s a significant area of overlap between RPA and software test automation. Organizations who have already achieved mature, sustainable test automation processes are much better poised for success with RPA. In many cases, they are able to apply many of the same automation experts and even the same automation assets to both test automation and RPA.
On the flip side, brittle automation—the same core problem that has doomed so many software test automation initiatives—has already emerged as the #1 enemy to RPA success and ROI. RPA users are realizing what testers learned years ago: if your automation can’t adapt to day-to-day changes in interfaces, data sources and formats, underlying business processes, etc., then maintaining the automation is going to rapidly eat into your ROI. Moreover, with RPA, the repercussions of broken automation are much more severe. A test that’s not running is one thing; a real business process that’s not getting completed is another.
Test automation is arguably more challenging than RPA, and resiliency is undeniable more critical in RPA. Find the people who mastered it in test automation, and you will be on a much smoother path to RPA success.
But recognize that test automation is not the same as RPA
Enabling fast creation—and updating—of resilient UI automation and API automation is essential for both RPA and test automation. Nevertheless, Murphy emphasizes, there are some important differences. Test automation tools lack the enterprise-grade orchestration, high-availability, and production focus required for RPA success. On the other hand, RPA solutions don’t provide many capabilities that are essential for test automation success: test case design, service virtualization, test data management, etc.
Put simply, RPA tools aren’t fit for enterprise software testing, just like software testing tools aren’t fit for enterprise RPA. Nevertheless, core automation assets and experts certainly can (and should) be leveraged across test automation and RPA if you want to take the fast track to RPA success.
Commit to both short- and long-term goals from the start
Wrapping up, Murphy recommended a very specific action plan:
- Next week, clarify your organization’s RPA position and partnerships.
- Over the next 3 months, examine your processes, identify a core set of quick wins, then build an enterprise automation roadmap focused on business outcomes.
- In the next year, plan how you will scale RPA and extend RPA to reduce TCO and increase business agility.