A wrong way sign on the road way before construction - negative impact concept

In research carried out about a year ago, just as the COVID pandemic was really spreading, the Boston-based global consultancy Bain & Company looked at the forces that were not only forcing workers to stay home — COVID-19 — but also the technology that enabled them to stay at home. Based on survey responses from nearly 800 executives worldwide, it found that many companies had no choice but to turn to automation to keep the business running.

The crisis forced companies to move their operations remote within a matter of days, underscoring a greater need than ever for automation technology to help maintain business continuity.

Entitled Intelligent Automation: Getting Employees to Embrace the Bots, companies reported cost savings of roughly 20% since 2018 with the implementation of automation. However nearly 45% of respondents reported that their automation projects had not delivered the expected savings.

Since then, the global business community has been working remotely and organizations have turned to automation to keep them running. It is unlikely that any automation investments of the past year will be stripped away now, especially as so many workers will continue to work remotely once the pandemic has passed. So will automation really bring savings or will it create even more problems for enterprises?

Automation Challenges

Even with the many benefits of automation, organizations still encounter challenges that prevent them from effective implementation, Rich Waldron founder and CEO of San Francisco-based low-code workflow automation company Tray.io, told us. According to recent research they carried out, some of the key challenges organizations encounter when implementing automation programs are securing internal resources, securing budget, and an inability to identify high-value automation opportunities. Every company’s automation journey is unique and can bring its own setbacks along the path to success.

However, he points out that there is a silver lining in that most of the research respondents expect their challenges to be resolved within the next three to five years. Organizations are understanding the tremendous value of automation and expect to see significant growth and investment in these initiatives moving forward.

Related Article: Automation Is Still Growing in the Workplace Despite Concerns 

IT and Business Friction

There are other gaps too. Sagi Eliyahu, CEO and co-founder of San Francisco-based Tonkean, which provides process orchestration software, thinks one of the biggest issues with automation is the gap between the priorities of IT teams and their business teams. On the one hand, he said, you have business trying to deploy automation on their own with minimal to no IT support, which leads to simple tasks being automated without thinking about gaps or impact to the overall process. On the other hand, you also have IT running automation projects without full buy-in and participation from business, which leads to a disconnect between what is automated and what delivers business value.

More and more, the priorities and needs of business teams, who want to move fast and innovate to solve hard problems, prove diametrically opposed to those of IT teams, who seek standardization, maintainability, and security, to limit risk.

Even when their incentives are aligned, business teams lack the technical ability to build, implement, or maintain needed automation on their own — efforts undertaken on their own often directly comprise IT’s ability to preserve the integrity of the tech stack and limit risk. This renders the business by necessity subservient to and reliant on IT.

But IT almost always has other things they would rather be focused on than helping the business use another app for automation. "The result is not only animosity between these two sides of the enterprise, but inefficiency on account of broken processes; an entrenched inability to move fast or adapt; and, if business users shop for automation and other tech solutions independent of IT, increased risk. Somehow, all sides lose," he said.

The Problems With Repetitive Behavior

According to Chris Nicholson CEO of San Francisco-based Pathmind, which applies AI to industrial sites, automation is another word for programming, since computer programs automate repetitive behavior. If you want to automate business processes, you need them to be repetitive — they must be standardized. It is no good telling a computer program to just analyze a situation and go with its intuition. Many business processes include a little wiggle room so that people can adapt them to the unexpected, but computer programs do not handle that well.

It also takes time to automate business processes. Someone must do it, and the people with the skills to program or automate are not always available. In addition, the domain experts who can teach the programmers what must be done may not have the time, or the inclination to help automate their own jobs. This is a problem of incentives as well as scarcity. The time required is not just up-front, it is ongoing. “Software requires care and feeding because software operates in a changing environment. Sometimes the business needs that are served by the business processes change. Who will rewrite and maintain the automation? Who understands the process and the changing needs?” he asked.

Dealing with unexpected contingencies, and adapting to changing business needs, are things that humans can do well. But those are not problems that computer programs can solve or respond to. “We should be realistic about the nature of business. Humans are under-rated. Automating tens of thousands of years of evolution (the human brain), is not easy. You can't just do it with a few lines of code,” he added.

Many people try to automate business processes with low-code/no-code solutions like RPA. Like computer programming, RPA is a skill unto itself. Instead of hiring software engineers, companies must try to hire RPA consultants. To expand the number of people who have the skills to automate a process, you must teach people who can think algorithmically.

It's Early Days Yet With Automation

Automation's progress in the enterprise seems all but inevitable. But organizations have some work ahead of them to realize the full benefits that automation can bring.