RPA is ready for its close-up

Three little letters are proving they can deliver accounting firms big benefits in the quest for greater process automation.

Enter RPA.

Robotic process automation, more commonly known as RPA, involves programming software to automate manual processes across applications. The goal: to reduce the burden of repetitive, simple tasks on employees. The potential impact: big.

The technology has gained greater attention in recent years as more firms look to drive efficiencies and improve workflows through automation. Despite this, the profession’s adoption of RPA remains in the earlier stages, according to sources.

“As far as adoption goes, I think it’s still a little bit slow. Certainly at the larger firms it seems like people are moving in that direction, and certainly integrating RPA with other technologies that you’re purchasing seems to be something that is in play as well,” said Andrew Spikes, shareholder at Topeka, Kansas-based Mize CPAs.

Added Gary Baumgartner, CIO of Armanino in San Ramon, California, “I think everyone kind of agrees they are in the early stages of exploring what RPA brings to the table. I am excluding the Big Four for right now simply because of the amount of investment they have available there further on. But in general I think everyone is in the earlier stages.”

That could soon change, however.

A 2019 report by consulting firm Protiviti found that companies, including those in the accounting and finance arena, are planning to up the ante on RPA. According to the findings, 67 percent of financial services organizations said that they are in the planning/implementation phase, and 29 percent are in the maturing/advanced stage. More than half (53 percent) of the respondents said that they expect to be in the maturing/advanced stage in two years.

It’s not coming for your job

Let’s take a closer look at what RPA is, and what it isn’t.

RPA is deployed to automate mundane, repetitive work and improve workflow. It doesn’t seek to mimic human intelligence like artificial intelligence, or gather new insights like analytics. And, much like other emerging technologies, RPA is not designed to replace employees, but rather to augment their roles and enable them to focus on higher-value work.

Take, for instance, the example outlined by Baumgartner: “The way we see RPA is providing that connection to multiple solutions. Through leveraging RPA, we can assign work to a digital worker through a task management solution, retrieve client information from, for example, our tax filing system, file the prepared tax return, and then let the team know when the IRS has accepted the return. A significant amount of automation can be applied in a very high-volume transaction, like tax filing,” he said.

As outlined in the report from Protiviti, the greatest benefits of RPA include increased productivity, a stronger competitive market position, higher customer satisfaction, and better quality.

One firm that is seeing the benefits of RPA firsthand is Mize CPAs. “We really looked at RPA as a way to remove some of the manual, lower-level processes that we do and allow our staff to be able to work on higher-level things that bring more value to our clients,” Spikes said.

One of the first projects, said Spikes, was to use RPA to retrieve invoices from the website of a vendor used by many of its clients, then process those invoices through the firm’s bill pay software. “A task that we used to do manually: Twenty-five-plus people a month going out there and pulling invoices, coding recast, and [doing journal entries]. By moving to [RPA], we basically got rid of that entire process but we also then elevated our process,” said Spikes. “In the past, our clients saw a journal entry recap of the expenses for the month. Now, they actually can see invoice-level detail through our AP system for each and every invoice for the month. So, it reduces some of the questions surrounding costs associated with that vendor and, for us, it saved probably just over 2,500 hours of work a year.”

It has been about a year since the use of RPA has picked up steam at Mize CPAs, which has developed the technology internally. And, today, there are a handful of RPA-related projects currently in play within the firm, Spikes said. Among the more recent is a project to use RPA to automate the creation of client tax packets to save staff time and expedite the delivery of client tax packets to the tax department. In addition, Mize CPAs is currently working with its internal IT department to develop RPA solutions to assist with the firm’s monthly financial statement preparation work and the creation of month-end ratio sets for clients.

“So, we actually have about five [RPA projects] in play right now, all focused on client work. And then we have several internal-focused ones. One [internal project] is surrounding health checks on our servers. We have staff that do that today, but now we can actually have the robots do the monitoring itself, as well as also monitoring and creating different file structures for document storage,” Spikes said.

Spikes noted that the firm is currently saving about 3,200 hours annually on mundane tasks. Once it completes the remaining RPA projects now underway, it expects to save about 7,500 hours annually. That’s time and energy that staff can redirect to higher-value work.

Challenges and lessons learned

Despite the benefits, the journey to RPA can be fraught with challenges. According to Protiviti, the top obstacles to adopting RPA include:

  • Inability to prioritize potential RPA initiatives (40 percent);
  • Concerns about cybersecurity/data privacy (40 percent);
  • High implementation costs (37 percent); and,
  • Difficulty in scaling applications (37 percent).

To overcome such hurdles, Protiviti highlighted lessons learned by those that are most advanced in their use of RPA. These include:

  • Start with processes that are simple to automate through RPA;
  • Make processes as efficient as possible before automating them;
  • Plan and budget for ongoing maintenance;
  • Consider the scalability of the application and whether it can support advanced AI technologies;
  • Assess and address up front any cybersecurity and data privacy risks associated with the planned RPA application;
  • Make sure you are purchasing the right RPA software; and,
  • Analyze and track the cost-effectiveness of each RPA application.

To assist executives looking to implement such innovation into their business, as well as leverage the learnings internally, Armanino recently unveiled its AI Lab, which aims to take a practical approach to integrating AI in the day-to-day working environment using three core components: predictive analytics, RPA and virtual assistants. It provides members with access to ready-to-deploy, proof-of-concept AI models that can be modified and customized to their specific organizational needs.

“The services and the value-add that we want to provide to our end clients we also try to use internally everywhere that it makes sense. So, today our AI Lab is currently focused on democratizing AI and automation and making it accessible to our clients. It’s all about the clients, even though, again, we want to leverage the learnings we are delivering to the clients and also repurposing them inside the organization and vice versa,” said Baumgartner. “Especially important is to help [clients] scale an enterprise RPA program. To be able to go from a proof-of-concept, to scaling, to a more mature operating model is not simple.”

The AI Lab, for instance, helps clients navigate such issues as selecting a vendor, selecting the right use cases, setting up the technology platform, and building internal development capabilities. “In our case, team Armanino has helped companies overcome these hurdles and provide support as they set out on their RPA journey,” Baumgartner said.

When discussing the benefits of RPA, Baumgartner said it extends well beyond cost savings, though those can certainly be significant. “It definitely increases efficiency. There’s also a pretty significantly improved quality of deliverables, and the overall reduction in human-related errors,” he said, noting that, “How we build integration between multiple systems can be massively complemented by things like RPA.”

Daren Campbell, Americas tax innovation leader at Big Four firm EY in Washington, D.C., said that it’s important to look at automation holistically and to leverage the right tools for each specific project. EY leverages RPA in a variety of instances such as onboarding new hires and automating repetitive data entry activities on the back end, as well as within client services.

“The No. 1 area that companies seem to start with for automation or use bots is usually in the extraction of invoices. … Primarily related to sales and use, where companies need to be able to pull down invoices on audit or for any of their reviews for monthly compliance. … A lot of companies have looked to RPA to do that. In the sales and use space we see a lot that have used it for managing the exemption certificate process,” said Campbell. “We’ve seen it a lot with being able to pull down source data — when the client has a lot of different ERP systems and it needs to pull down trial balances from all of those ERP systems that it will later be using for tax provision or tax compliance purposes.”

Added Campbell, “We’ve used it both internally and externally in a lot of the reconciliations [in] comparing current-year to prior-year reports and schedules. We’ll use RPA to pull down the reports that are needed. This is oftentimes integrated with data integration as well, but bots will pull down the information, will process it through one of the data integration tools, and then the bot will pick up the results of that, prepare that final workbook, and email that to the reviewers. … With any of this, the human in the loop is an important element.”


When one thinks of RPA, perhaps one of the more recent developments that comes to mind is the ban on RPA in the submission of applications for Paycheck Protection Program loans.

Given the speed and efficiency of RPA software compared with human data entry, some banks were using RPA to enter PPP loans applications into the Small Business Administration’s E-Tran electronic loan system. However, in a statement issued in April, the SBA and Treasury announced a ban on RPA in the loan application process. “RPAs burden the processing system and diminish its capabilities. Without RPAs, the loan processing system will be more reliable, accessible and equitable for all small businesses,” the statement read.

In commenting on the news, Campbell said, “It sounded like the issue was that the underlying system that they had couldn’t handle the volume and the speed at which the bots were pushing data. I think it was their E-Tran loan system, which isn’t a surprise. … A lot of times, whether we’re working with a system that is internal or a third-party system where we are using RPA, one of the things that we try to do is when it is interacting with those systems that it is operating more at a human pace. So, we actually build in delays and other things to prevent it, [which] sounds like what happened here.”

Said Mize CPAs’ Spikes, “I think the SBA’s concern was that some of the larger banks using RPA to automatically transmit applications into the SBA system was going to overload it. … Not overloading the system and clogging it up, and allowing smaller banks to be able to submit as well without having the system crash.”

The future of RPA

Most accounting professionals agree that automation is key to future growth. But given the rise of application programming interfaces, the choice between RPA and APIs may be confusing. However, as stated in a blog post by RPA vendor UiPath, “Each has advantages and disadvantages. In many situations, the right answer is a hybrid strategy that uses RPA as the larger framework and API integrations for specific functions.”

Campbell of EY noted the rise of APIs, but maintained that RPA will still have its place as a critical automation tool. “For the foreseeable future, there will be systems where APIs just aren’t possible. … So, utilizing RPA to be able to interact through the front end is going to continue to be important,” he said.

Also strengthening RPA’s foothold, he added, is the fact that RPA capabilities are increasingly being brought into other solutions. He cited as examples SAP, which has invested in RPA to help automate repetitive processes, and Microsoft’s recent acquisition of Softomotive, a provider of RPA and creator of WinAutomation.

In announcing the news in May, Microsoft stated, “Together with [Microsoft] Power Automate, WinAutomation will provide customers additional options for RPA desktop authoring so anyone can build a bot and automate Windows-based tasks.”

“I think increasingly we are going to see RPA being embedded as a capability into some other tools and products,” Campbell said.

August 14, 2021