After nearly two years of companies diving headfirst into digital automation projects, grumblings are beginning to emerge.
Gartner noted that 69% of Boards sped up their digital business initiatives after COVID-19, with 86% feeling technology will transform their strategic business priorities. It has also seen an increase in creating chief digital officer (CDO) type positions and vendor rhetoric.
Yet, a Boston Consulting Group report noted that 70% of digital transformations fail. In a separate survey conducted by digital intelligence company ABBYY, 58% of C-level executives felt their companies were not digitally ready despite having digitalization efforts. Twenty-two percent abandoned their projects, and 32% said that the technology did not work as intended.
The disconnect is creating fresh challenges. The ABBYY survey noted that failure to deliver what senior leaders want is becoming a significant roadblock for digital transformation, especially digital automation, success. The bad experience is also creating another headache for CDOs: budget approval.
Hammering out wisdom
Anthony Macciola, chief innovation officer at ABBYY, felt part of the problem was how many companies started their digitalization projects. Many used digitalization to automate via RPA. The problem was the haphazard and hurried approach. It was like “the wild, wild west,” he adds, pointing to the early periods of digitalization after COVID-19.
“It’s like some person figured out in the business unit one day what they could do with RPA, and a wildfire starts,” he explains.
IT had a better approach. They had a workable formula after years of working with procurement and CFOs to get budget approvals and refine their best practices.
“And initially, a lot of IT departments tried to stem off RPA because they realized it was going to break their best practices approach. But they also realized very quickly that they either get on the train or off it,” he describes.
That’s because business users want to automate quickly. With COVID-19 making business more challenging, they wanted agility and resilience. And they became adamant automation and RPA were the way forward.
“You know, when you're a hammer, everything looks like a nail. So, people were just automating things because they could, regardless of whether or not the impact [to the business] would affect the bottom or the top lines,” says Macciola.
It made the problem worse.
“So, a lot of the best practices that IT became comfortable with over the last 30 to 40 years just got thrown out the window. And the business units just started doing stuff. But they then began to realize that just because you can automate doesn't mean you should. And in some cases, you may be creating challenges in other places [of the organization],” adds Macciola.
Other issues started to crop up. For example, many business processes have inherent dependencies that may not be clear from the onset. It takes time to discover these dependencies and find out whether the platform is well-suited.
“Whether they are back offices processes or customer-facing ones, there’s a lot of dependencies on content, media, unstructured data, etc. Not all automation platforms are really well suited to deal with these,” he explains.
Another major issue is whether the RPA technology or vendor will remain relevant. Macciola points out that many companies began centers of excellence (CoEs) to study RPA. “And a year down the road, they realize that 70% of their robots aren’t going to be in existence this time next year. The payback or the impact on the business isn’t what they thought.”
Such issues saw 35% of the ABBYY survey respondents saying it is becoming more difficult to approve the budget. It was named one of the top three barriers to intelligent automation projects. The others were having difficulty replacing legacy systems (36%) and finding the necessary skills (34%).
Truth is still out there
So what can companies do?
The first step is for CDOs and their business users to find a common understanding of whether the automation or digitalization projects are what the company needs. Macciola also advises companies to temper the opportunistic approach to automation and become realistic. However, he agrees that this is not easy as many millennials and Gen Z join workforces and question why they are overloaded with manual tasks.
Macciola urged digitalization teams to determine which processes need to be impacted to create the most significant impact. Unfortunately, the answer is often not that obvious. Just shooting for those low-hanging fruits to do the easiest projects may not be the right step forward. If the difference to the top or bottom lines is minimal, no CFOs with limited budgets will increase the funding easily.
Companies also need to understand their processes first. “Identity process variance, and I can’t tell you how many times we had to tell this to an organization.” It is why ABBYY is investing a lot more time and money in the process and task discovery “so that we can build process maps quickly and identify the bottlenecks.”
Then, companies need to find out how they are going to prevent adverse outcomes. This matters because Macciola believes that companies were not spending their digitalization budgets efficiently or effectively.
“Tools are emerging that will help them become more focused. They'll be able to monitor and report on their success towards achieving their intended outcomes,” he explains.
Macciola believes many companies realize that they need to take back control of their digitalization efforts. Many are beginning to use Six Sigma practices to CDO budgets and initiatives.
Vendors are also helping out. They are helping their customers to understand the dynamics, bottlenecks, opportunities, and impact on their business. This allows CDOs to make corrective measures midway through a project if they feel it will not deliver on its promises.
Lastly, Macciola urges companies to be less biased when picking the right processes. For example, many focus on customer-facing processes as they readily see the top-line benefit. Yet, backend processes get overlooked as they benefit the bottom line and do not bring in huge amounts of revenues.
“Many organizations now realize that to create better customer-facing experiences, you need to increase back-office operational efficiency,” says Macciola. Again, it points back to understanding the dependencies and where the investment should be made for maximum effect.
Develop the right tools and use cases
Two of the biggest challenges, when businesses drive RPA projects, is that the problem statement is defined by the business unit, and toolsets are limited. This creates a challenge when companies try to scale the RPA and leads to failures and budget worries.
CDOs need to become pragmatic. “You know, you talk to [business users], and they think RPA will solve world hunger and global wars. But you need to peel back the onion [layers], turn down the volume knob on the rhetoric, and understand what you will use it for. If you understand the task behavior and the impact, you’ll probably do very well,” says Macciola.
He also advised CDOs to build the right toolset first. It means they have to go beyond what business users want and look for tools and vendors that deliver on their promises.
It is where Macchiola believes a best-of-breed approach works best. “So I pick vendors based on their core expertise and not their marketing narratives. And I will pick vendors who get that their success is predicated on interoperability with other tools in the ecosystem.”
Most importantly, CDOs need to have a broader set of use cases in mind. They need to question if the RPA works in finance, how it can be replicated in other functions. “But when you ask someone for a use case, they start looking at the clock on the wall. They will say we'll get back to you on that,” says Macciola.
This attitude is slowly shifting. Macciola observes that more companies are looking at using technology around a horizontal capability that the organization needs. And this creates more use cases right at the beginning.
“When you're horizontal-focused, you're trying to anticipate the next use case without actually knowing what it is. Then you're continually striving to make sure that you're on the cutting edge, even though you don't maybe have a material example for why you need it,” Macciola concludes.
Winston Thomas is the editor-in-chief of CDOTrends and HR&DigitalTrends. He is always curious about all things digital, including new digital business models, the widening impact of AI/ML, unproven singularity theories, proven data science success stories, lurking cybersecurity dangers, and reimagining the digital experience. You can reach him at [email protected].
Image credit: iStockphoto/Andrii Atanov