In a preview of his feature in the winter 2020 Pennsylvania CPA Journal, Adam Costa, CISA, CCSP, a robotic process automation (RPA) solution architect for Schneider Downs & Co. Inc. in Pittsburgh, offers a crash course on RPA, including the ways it
is different from standard spreadsheet software, how structuring data helps CPA firms and other businesses, and the areas of accounting that are already seeing significant gains from this emerging technology. Also, don't miss these related courses.
By: Bill Hayes, Pennsylvania CPA Journal Managing Editor
In the winter 2020 edition of the Pennsylvania CPA Journal, we are lucky enough to have two tremendous features on technological developments affecting the accounting industry. Next week, we'll be releasing a podcast on audit data analytics. But why celebrate what is coming in the future when you can enjoy what you have in the present, which is a discussion with Adam Costa, CISA, CCSP, robotic process automation solution architect for Schneider Downs & Co. Incorporated in Pittsburgh. His winter 2020 Pennsylvania
CPA Journal feature is on, of course, robotic process automation, and he offers us a preview of it today.
There's probably going to be some uninitiated in the audience. Can you define the term "robotic process automation," which we'll also be referring to as RPA?
[Costa] Sure. First off, I'll say I have a love-hate relationship with this term. People have a strong association between the word "robot" and mechanical machines. I often go out to introduce this topic to clients and find that people want to
see the actual physical robot. There's no real robots here in RPA, right? We're talking about software. It's software that you can use to replicate human interactions with computers.
And also, since RPA is software, it can use a lot of
traditional APIs and other programming interfaces to get the application data, databases, and other sorts of files.
In what ways is robotic process automation different from standard spreadsheet software and other macros that are out there?
[Costa] I'd say the big difference here is in scope. Macros are generally limited to one application and allow you to automate within the functionality of that specific application. RPA has a much bigger scope of interactivity.
With RPA, you're automating all the different pieces of the process, not just the parts within a specific application. You're automating end to end.
One of the main benefits of RPA is the structuring of data. Can you give us a concrete example of how this helps CPA firms and businesses?
[Costa] A common use case here that a majority of businesses are going to find useful is in their accounts receivable processing, and in that process you've got a lot of invoices. Invoices are a semi-structured form of data, right?
So think about all those invoices out there. Everyone's got a different format. It's all over the place, right?
You might have 50 different vendors, 50 different formats, but they're all carrying very similar sets of data, invoice number,
vendor number, quantities, prices, so forth. If you're not using some sort of electronic invoicing system already, you're going to get these in paper format, email format, a bunch of different formats, and you're going to need an employee to read
and key that information into your accounting software or whatever system you're using to keep track of everything. This is going to be slow and tedious, and there's going to be all sorts of chances of mistakes and so forth. Well, with RPA, you can
train them to read those invoices and output a set of structured data that then you can use to pull into your accounting software and other forms of business processes.
Let's talk about some of the advantages RPA will bring for businesses. How will it bolster efficiency?
[Costa] This is a hard question to answer. There really isn't one specific answer to this question. The interesting thing about working with RPA is that when you meet these companies, you're seeing many different ways that people are
doing business. So the efficiency that one company is going to get out of RPA isn't going to be the same as the efficiency that another company gets out of it. My experience with working with clients, it's hard to predict where that efficiency can
really be gained.
I walk into somewhere knowing an industry, knowing where other people have been successful with RPA, and I find that this other company's having a completely different pain point, and they want to use RPA in a completely
different way that I didn't even think about. So that's kind of the beauty of it. It is specific to what you need and can help you create those efficiencies in your processes depending on your company and your needs.
Interesting. So, not really a one size fits all sort of thing.
[Costa] Right. Exactly. That's the beauty of it. It's really meant to fill in those gaps for everyone in different ways.
It might be a similar type answer here, but how does RPA increase data traceability?
[Costa] Traceability, in general, is something that you're going to get when you look at a software solution, right? Software is generally traceable because it does the same thing over and over again, and that's true for RPA. A well-designed
RPA bot is going to have built-in auto-logging. That's going to allow you to trace your data as it moves through the process, right? You give the bot some data, you can see exactly where it's going, what it's doing with it up to the output where the
bot stops working off it. Once you've got your bot quality controlled and tested, you're going to know exactly what's happening to that data, where it's going, and what's being done with it.
Would you say there's a role for RPA to play in an area such as risk management?
[Costa] For sure. Just like traceability, having computers perform the task is going to give you some advantages here, and the security benefit for RPA is that it's taking the data out of people's hands. People are usually considered
the weakest link in a security chain because you can't guarantee that someone's not going to do something unexpected or nefarious with that data. So if you've got some sensitive personal information, you're working with HIPAA data or private financial
information, you can keep that information out of people's hands, and then they never even see the sensitive data, so your security risks go down.
As it applies to RPA in the accounting field, what changes would you say are already well underway and taking place, and what areas are being most affected?
[Costa] At the moment, as far as accounting, I think the focus most companies have is on the accounts payable/accounts receivable areas. Really, people want to first tackle things where they're going to see a cost savings and a value,
and there's a lot of value in the AP/AR function, right? Right now, you're spending a lot of time with people manually processing documents and putting them in the systems. It's a low-hanging fruit. Bots can do that. You can potentially reduce the
size of your AP/AR staff or even maybe reduce growing your AP/AR staff by just having the bot do much more of the heavy lifting of processing those forms and documents.
Are there areas that might not be seeing the impacts right now, but could be susceptible to a large degree of impact in the future?
[Costa] I think there's really two areas here. The first is in auditing and the second is in tax. For audit, I think we're going to see a lot more real-time and recurring audits being run with RPA, largely because companies want
accurate financial data and they want it now, and bots will make that much easier for people to achieve. In the tax area, I think it's because of clerical work. RPAs are going to streamline that work substantially. There will be a lot less manual
data manipulation and all that form filling-out where you're just copying data onto government forms from your actual financial documents. That's all going to get automated away.