Insights

RPA And Y2K

A cautionary tale about writing bot software scripts

Robotic Process Automation (RPA) is perceived by many large organizations as a way to streamline data collection from multiple sources and execute basic reporting tasks.

Instead of an individual (or team) hunting in different and often incompatible databases, software robots can more efficiently collect the data and then execute the task at hand.

But, because business units, rather than enterprise-wide IT departments, are often responsible for writing bot scripts to solve their particular problems, companies risk introducing another helter-skelter stack on top of the current legacy stack. There is a clear analog to today’s scenario—the way in which business organizations tried to patch over legacy systems in the early 2000’s.

If RPA remediation takes budget and management focus away from badly needed application modernization investments, it will delay the arrival of that agile, digital infrastructure.

Coming off the apocalyptic Y2K scare, companies had spent huge portions of their technology budgets remediating mainframe and other legacy applications for the century time-clock change, rather than building new systems. To circumvent legacy limitations, business users embraced the increasing power inherent in Microsoft Excel and Access to create complex, business-critical applications on their desktops.

As these custom-made computing tools proliferated, so also did the problems due to the lack of a strong controls framework, quality assurance, release-management processes, and other formalized IT protocols. Companies then subsequently spent large sums of money tracking down all their wayward tools and slowly eliminating them from critical functions. Fifteen years later, this work continues.

Today’s explosion of ad hoc bots threatens to repeat this pattern. Without thinking about the enterprise-wide consequences, a script designer might simply push data into an Excel file as a proxy database. That potentially creates another custom-tool remediation exercise—a large number of scripts, running on a larger number of bots, without the necessary standards and monitored source code that is critical in any modern, enterprise-technology platform.

Most organizations understand that RPA is not a long-term solution, but a short- to mid-term patch to keep unwieldy legacy systems functional until new digital systems are fully installed. But if RPA remediation takes budget and management focus away from badly needed application modernization investments, it will delay the arrival of that agile, digital infrastructure.