Speaker 1: Hi, everyone. How's it going? So, I'm going to talk about how RPA can be applied in finance. So, first of all, it's seen across the industry as a whole in banks and credit agencies, the use of data analytics and financial services. In banks and credit agencies, you're going to see it in commercial banks. Credit card companies are utilizing it. And the use of data analytics is seen in a bunch of finance and accounting firms for manipulation, extraction, visualization. And then many insurance companies, investment firms and accounting firms are all looking to apply RPA to their processes. So, first, I'm going to talk about what RPA is. I know we talked about it a lot, so some of this may be repetitive. I'll try to skip over it if we talked about it before. But RPA takes the place of a human user on rule-driven repetitive digital actions. And sometimes it's referred to as software robotics. And technology is used across many industries, like I just said. And it takes the monotonous process done by humans and automates them. These software robots do it, and they'll perform the set task or rule in a fraction of the time without the implication of human error. And like we discussed, the leading players in the industry are UiPath, Automation Anywhere, and Blue Prism. And then some of the functions of RPA, just to touch on a few. By UiPath, elimination of human error, perform back-office functions, manipulate, extract data, fill out forms, and work with programs and databases. Like mentioned, 100% accuracy. It allows you to manipulate, extrapolate, and organize data in various programs like Excel. And it is a cost-effective method of doing so. So these robots are capable of mimicking many, if not all, human user actions. They log into applications, move files and folders, copy and paste data, fill in forms, extract structured and semi-structured data from documents such as PDFs, emails, or forms, and scrape browsers as well. So I'll run through a quick little example from an HR service provider to show the benefits. So there was an HR service provider in Europe that would process 2,500 sick leave certificates per month. And these certificates would take an average of four minutes per item. And it took the company three weeks to implement an RPA solution, achieving 90% automation. And the system then extracts the data from SAP, the company's business application software, inserts the information into the customer's system, and then prints it. The error rate was 0%, the manual effort was reduced to 5%, and the processing time reduced to 80%. And the return on investment was actually seen in six months. So the implementation of RPA, like we talked about before, a cost-benefit analysis is going to need to be done to kind of see how the upfront cost and time save can bring value to the firm. And another thing that's not really monetary, but it can focus on retention as well as growth drivers for a company and allow employees to focus on revenue-producing actions. Now, RPA isn't for everyone. A business would have to see what's important and see if any other processes can implement RPA. Typically, high volume of data will result in faster return on investment. And this figure right here kind of lays out what the robots can perform in a premiums example. And then I'll touch on some examples in the finance industry as I go throughout the examples. So how is it being applied in finance? Many areas of the finance industry are applying or looking for ways to apply RPA to their business processes. Credit card companies are using RPA to give fast responses to consumers seeking lines of credit or loans. Banks are using RPA to automate filing processes and fill out forms, open accounts, and comply to the extensive rules and regulations that they must follow. And RPA is being used in many firms to manipulate, depict data, especially large data sets. And these methodologies pertaining to data can be seen in auditing, consulting, and financial analysis in many firms. Insurance companies are seeing benefit in filing claims. Investment firms are looking to RPA to fill out paperwork required for compliance, as well as document scanning for analysis purposes or extrapolating client data to better assist customers in their advisory practice. And accounting firms are looking to integrate RPA into their audit processes, as we discussed before. I know there's some challenges, but it's definitely a goal of a lot of these big four accounting firms. So first, I'm going to talk about banks. Sorry. Banking is known to be extremely repetitive and conservative, perfect for using RPA. So we went through a couple of the examples before, and it can act similar to Excel macros in a banking software, and they focus on labor-intensive, low-value tasks. And I know it's contrary to what people may think, but it can actually increase employee morale by taking away some of the repetitive, monotonous tasks that they're expected to perform and automating that and letting them deal with more higher-level work. And then how they would implement RPA is identify a subprocess on the process map and then find the subprocess that yields the most benefits and then develop a use case requirements, rules, keystrokes that these robots will perform. And to touch on a quick example, a consumer loan process time can be reduced from 30 to 10 minutes, eliminating copying information from one banking system to another and also mitigating the risk of human error. It can boost the speed of customer verification when processing auto loans by validating customer data in various government web bases or the DMV, things like that. And then to touch on credit agencies and credit card companies, transaction-based companies that are highly regulated represent a perfect opportunity for RPA usage. Companies like American Express are beginning to utilize RPA to open doors for other forms of automation, and they're focusing on employee engagement and financial integrity to start for the uses of RPA. And it's helping companies like American Express actually allow their employees in entry-level roles to perform more tasks that they could learn from and even learn about RPA itself, which makes them more valuable assets to the firm. And to kind of further explain and touch on machine learning and AI, two other forms of automation, credit agencies will invest in RPA and business intelligence tools to build copious amounts of data that, when combined, can begin to make assumptions using machine learning on client activity. And machine learning is something that can predict outcome by utilizing vast sets of data to statistically predict, at a certain confidence level, whether a payment or transaction should be flagged or fraudulent. And RPA is what can put that data together, and the data analytics is what tools like machine learning can use to make those assumptions. And now to talk about data analytics. Like I said, manipulation, extraction, and visualization are things that RPA could be used for. And the reasoning behind many finance and accounting companies to shift to robotic process automation is to reduce errors, decrease cost, improve efficiency, and like I mentioned before, open doors to machine learning and AI. So many finance companies will use big data to make business decisions, and the companies that are on the edge of changing are looking to machine learning and artificial intelligence to start to make those decisions for them. Companies that wish to work with machine learning and AI generally have already integrated robotic process automation and data analytics, two of the main things that a company would need to carry out an implementation of further automation. And RPA is the natural supplier of data, which is needed by artificial intelligence to function. And data is the fuel to AI. And this diagram shows RPA collecting and sorting data and then sending that data to a data robot, where the data robot will come up with a model, produce a prediction, and then feed it back to the RPA system, which will act based upon that decision. And these data robots can update and change with new data to assure accuracy in the data being used to make those decisions. And this process is a major goal and a big opportunity for firms to gain a comparative advantage over the smaller firms that maybe lack the resources to do so. And then in the insurance industry, they use it for many back office functions. So employees can focus on client retention and revenue driving tasks like claims processing and new business preparation are some things they use it for. And this diagram right here, if you can't read it, it's a little small. It shows RPA facilitates communication between legacy and newer systems like we talked about before. And then automation using RPA enables enterprises to process data from various systems with incomparable accuracy and successful implementation can free up 20 to 30 percent capacity. Like I said, to allow for employees to use their time, minimizing operational risks and enhancing the customer experience. So another UiPath example from I'm sorry, another example from a UiPath customer was PCU, one of the largest insurance groups in Europe. They wanted to focus on employees bringing value to the firm, so they implemented robotic process automation. I'm not going to go through all of them, but just to highlight a few, they automated calling for assistance after an accident has taken place, refunding the cost of vehicle repairs based on filed invoices, replacement vehicle rentals through third party liability insurance, corresponding for calculating compensation and for customer insurance history. Just looking at one of these in less than two months after implementation of one of those processes, like I said, PCU saw a 15 percent increase in the number of decisions issued per person. So people got the chance to actually make business decisions opposed to filing paperwork and doing monotonous tests. And they also achieved 100 percent accuracy of entered data with automation and 50 percent shorter call times. So investment firms, they're kind of taking a big shift towards machine learning and AI also in some algorithms they use for trading and portfolio management. But as for RPA, hedge funds, REITs and investment management firms are all using RPA to produce a higher quality of work and data. Processes such as setting up trust, moving funds, operating trade desk reporting, workflow and customer service functions as well, such as meeting requests, client advising, asset allocation and portfolio adjustments. And I came up with kind of a little example to show how much time RPA can save in a day. So Mike, the person in my example, works at an asset management firm as an investment analyst. He updates the broker estimate reports manually, which typically takes him 15 minutes per report. The process without RPA involves obtaining information from e-mails, Excel and web cloud portals. So before RPA, the process would look like this. Brokers would estimate change. Mike gets e-mails of the estimates. Mike would then open the e-mails, save the attachments to a network drive. He then reads through the attachment, copies the information within the attachments to the master totals located in an Excel spreadsheet. This process is repeated for each attachment. And then when this is finished, Mike will open his browser, go to a stock Web site to copy the estimates onto the same spreadsheet. And then after updating, Mike saves the spreadsheet, then opens Power BI, a data visualization program that many finance companies use. And he uses that to assemble a report. Mike changes the reports and charts to reflect the new estimates. And Mike spends about 80 percent of his day copying and pasting information between sources. With RPA, Mike receives the e-mails and then a robot opens Outlook and searches for the e-mails, then saves the attachments to the network drive. And then the robot reads each cell in the rows of the attachments within a second and enters the information into the master spreadsheets in six seconds. The robot will open a browser, go to a stock Web site, copy the weekly estimates from the site to the master spreadsheet. And then the robot will ask Mike if the information is correct, allowing him to check the accuracy and confirm that the data is in fact accurate and meaningful. And then once Mike confirms to the robot that the data is correct, the robot then will open Power BI, refresh the report based on the new saved master spreadsheet. And then Mike is asked again to check the report before it's sent to whoever needs it in the firm. And then Mike can go ahead and e-mail the report to the traders. This step took three clicks of a button compared to the hundred clicks required before the use of RPA. This just shows what took Mike 15 minutes before the implementation of RPA. Now it can take him one minute and allows him to handle hundreds of these reports per day opposed to 20 to 30. And the last thing I wanted to touch on was the use of RPA in financial services, specifically accounting. We talked about it a lot. All the big four accounting firms are currently using RPA or trying to implement it into various processes. And they're seeing more use in testing and assurance practices. And I'm going to go through a quick example that I came across in my internship with Ernst & Young to see where I thought RPA could have been used. And in a recent survey from Information Services Group, it reveals that RPA was the second most in-demand automation scale in 2019. The first was artificial intelligent technology. So just to kind of run through a personal example, obtaining bank confirmations and confirming those is something that I thought could definitely be automated. I dealt with a company with multiple entities and each entity had its own spreadsheet. I would first match the entity's bank statements to the master spreadsheet once obtaining the PDF copies from the client. And then I'd sum the monthly data and highlight the matching figure on the big bank statement if it matched. Then I'd copy the bank statement to the master PDF to confirm the support and put that in a folder. And then I'd go back to the monthly data in Excel, copy and paste the monthly data to a master spreadsheet with all the prior months. And then pivot all the prior months and confirm them with the bank statement and make sure they're all in one place. This test took a very long amount of time and required communication with the client to obtain sufficient support if something couldn't be read or wasn't accurate. And some entities and different currencies required calculations. I know we touched on that before. And also manipulation of data such as reconfiguring dates for foreign entities so formulas can have consistent data in the master spreadsheet. And this timely process, like I said, could have been reduced with robotic process automation to organize the support, put the data together in a fraction of a time. So this is kind of how the industry is changing and applying RPA to various processes. And just kind of a comment. I think that within the next five years, RPA will be widespread across the finance industry and all companies will have adapted to it. And they'll make the shift towards adapting machine learning and artificial intelligence to actually make decisions for them.