The challenge for CPA firms: Applying RPA in the real world
It’s 2020, and by now it’s likely you’ve heard of robotic process automation (“RPA” for short). For those of you who haven’t heard of it, put as simply as possible — RPA does not involve little robots rolling around an office performing tasks. It does, however, utilize cutting-edge software packages, or “robots” to automate business processes and perform nearly any structured, repetitive, routine, low-judgement task that a human can perform on a computer. Quite literally, in most cases, you can watch as a bot moves a mouse across the screen, types, searches and logs into business applications, without any human interaction.
RPA is all the hype right now. It’s quickly making waves in the banking, health care, and insurance industries, among others. According to Gartner Inc., the RPA industry grew nearly 63 percent in 2018, making it the fastest-growing segment of the global enterprise software market. According to a report by McKinsey, automation technologies like RPA are expected to have a potential economic impact of $6.7 trillion by 2025. This impact is especially impressive when you realize that the term “RPA” dates back only to the year 2000. (RPA does, however, rely on a number of pre-existing technologies that date back to the 1990s, such as screen scraping, workflow automation and artificial intelligence, so some people do consider it to be less of a “new” technology and more of an “extension of technologies that came before it.”)
Basically, RPA is here, it is here to stay, and it is quickly changing many industries. This brings us to the challenge for CPA firms. How can we utilize RPA to create efficiencies both in internal processes and/or financial statement audits? In theory, RPA is great — it saves time and can do manual tasks for you! The challenge for many CPA firms, including ours, is how to actually apply this in the real world. Here’s a real-world application from our firm:
Imagine auditing a client whose business consists of pre-settlement funding — the business provides plaintiffs with cash advances to pay for necessary expenses as legal action is pursued. Essentially, the client’s entire business is a roster of legal cases in which they have funded plaintiffs who are in various stages of litigation. Cases, especially in New York, can go on for years. As an auditor who is auditing this company, how do you ensure that a case is still outstanding and therefore a true receivable at year-end? How do you know that a case is still active?
The old way: In New York, and many other jurisdictions, there are public websites where you can search the status of any case and find out all relevant information about it. This is great! Luckily for the auditors, they can search these cases and find out the status from a third-party source. The client gives the audit team MS Excel files with the case details. Auditors must then search the court websites to match the correct court case to a case number or partial case number from the data provided by the client for hundreds and in some cases thousands of cases, based on the sample size. The results from the search are either entered into a spreadsheet manually or copied and pasted. This is extremely repetitive, labor-intensive and very much an inefficient use of any auditor’s time (at any level).
The new way: Utilize RPA to the point where each task turns into a programmed process in which a bot performs all the repetitious steps:
- First, the robot cleans the data. RPA can be used as an ETL (extract, transform and load) tool to fix improperly entered case numbers and flag incorrect or incomplete for review. The data will then be automatically separated into specific Excel spreadsheets, based on the court (local court, appellate court, etc.).
- Second, the robot logs on to the court websites and conducts a search for the current case status. The robot will open a browser, go to the website URL, copy and paste the first case index number from the spreadsheet of case index numbers into the website’s search feature. The robot then scrapes all the relevant search results from the website, including the status, the most recent or next appearance date, among others, and writes them to a new spreadsheet that contains all the existing data and several new columns that contain the data points from the website — with near 95 percent accuracy. All cases verified by the robot are written to an Excel spreadsheet, while cases with insufficient data are written to another spreadsheet to be reviewed by the auditor. This frees up time for the audit team to actually analyze the results and exercise their judgement rather than spend hours on data collection.
Tasks that once took a human auditor days can be performed in hours. Tasks that took hours can now be performed in minutes. Process accuracy is increased dramatically. Humans no longer perform repetitive tasks. In an RPA process, humans have a supervisory role and are enabled to work on more valuable work.
Our recommendation for CPA firms interested in RPA is to start small. You won’t be able to automate every single process on every client engagement overnight. Identify larger, more data-intensive client engagements and start with those. Look for processes that are manual, repetitive, and require little judgement and tackle those first. Eventually, you will see the positive impact to implementing RPA and will be able to identify even more engagements it can be applicable — we are now in the age of the automated audit.