On-Demand | How to Achieve a 'Hands Free' Financial Close
- Mar 19, 2019
Technology impacts every aspect of our businesses. Understanding change and how to capitalize upon it is essential information for every executive-level employee. During this webinar EisnerAmper’s Digital Solutions specialists discuss how Robotic Process Automation (RPA) is being leveraged to improve the monthly financial closing process.
Greg Fritsky: Thank you. Good afternoon, for those on the East Coast and morning to the others. Thanks for joining us. This is a busy time of the year. It's the last day of winter. So it's springs upon us. So, happy days ahead. Thanks for making the time during your busy day, during your lunch on the East Coast here where I'm joined with my colleagues Steve Palomino and Ben DiNapoli. My name is Greg Fritsky. I'm the director, a practice director. I run the Practice Digital Solutions and Process Automation. Focus on number of different technologies, but helping clients understand these technologies and helping bring solutions to make your lives easier. My background is accounting and auditing. I worked at Deloitte for a number of years before I transformed myself into a consultant, focus on technology, work for a software company as well. Most recently, a company that actually had its own proprietary robotics process automation software, where I actually met Steve Palomino. And I'll let Steve introduce himself.
Steve Palomino: Thank you, everyone for taking your valuable time and spending it with us. We do appreciate it. Steve Palomino. I'm also with EisnerAmper, based out of Dallas, Texas. And I am a CPA and a CATP. I've been a practicing accountant for many years and moved into consulting and process transformation where I support Greg in the digital transformation space here at Eisner, it's very exciting time. We're actually applying not only RPA but other technologies around accounting and finance in helping organizations make that transition from some of their legacy environments into their more modern environments. And with that, I will hand it over to Mr DiNapoli.
Ben D.: Thanks, Steve. Yes, my name is Ben DiNapoli. And thank you all for your time this afternoon. I've been with EisnerAmper for over two years now. And I've handled a lot of the technology implementations and kind of assessments on that. And so with that, I'll turn it back to Greg to begin the presentation.
Greg Fritsky: Thank you, both of you. And thanks again for joining us. Number of outcomes, we want you to achieve today. I want you to be able to take away something, if anything, it's that robotics process automation is really a driver for change. Really, at the end of the day, we keep talking about new technologies, folks are always talking about digital transformation but RPA is really kind of the starting point. It's a good place to take a look at your own processes, your manual repetitive activities, and look for ways that you can leverage a technology like RPA or other software's to achieve automation. So today we're going to talk about identifying month end closed processes that can be fully automated end to end.
We're going to discuss how to leverage RPA to transform and streamline those finance processes. And we're also going to demonstrate how bots can actually complete manual repetitive month end closed processes for you. So today's agenda, we want to focus on the finance processes and operational challenges. The reality is everybody's trying to do more with less. We hear this all over the country as we speak too many different folks, we see it fall down.
The reality is that we aren't going to be able to add the people as fast as the work is coming in. These are good times, but folks are looking to improve and increase scale. They're looking for ways to leverage new technologies and analytics capabilities as well. We also see that ... Finance processes and we're talking specifically in this space today, but finance processes are complex. So oftentimes it's hard to understand how do I take a process where I know there's problems and deviations. And how do I standardize that and how do I address that?
The other thing I hear oftentimes is how do we get people to think more and do less? The reality is that people honestly one of these doing, the thinking analytical work, we want to repurpose them to do so. We've spent many many many years in finance accounting doing the same processes. I oftentimes talk to people and they mentioned how the processes that they have in place, the shared service center that maybe they built, it's 20 30 years old on technology that it's that old as well. Nothing has really changed. We're able to meet our deadlines, but how we get there is often a challenge. So how do we transform? What's this whole digital transformation? We want to talk about how to eliminate these manual repetitive tasks, how to reduce costs, increase scale, and also how do you leverage this tool to help you enhance internal controls.
And finally, we'll talk a little bit about how do I start? This sounds great, but how do I really get this jump started this kind of program? What kind of folks do I need to engage? How do I leverage the people I have today? And what's this whole concept of process transformation? And how can I do it at maybe a small scale to learn and then start to push it through the enterprise? So with that said, let's start with operational challenges. We mentioned finance processes are complex. How do we get a level of transformation and control?
The reality is, as I mentioned, folks are sitting there, they're challenged with getting things done, especially on a monthly basis. When you think about things like the reconciliations you perform, the manual journal entries, inner company transactions, think about all the different things you do on a daily, monthly and quarterly basis. Oftentimes, is it really complex or is the challenge really that we just do it because that's the way we always did it.
Steve Palomino: And the other thing to Greg, that's important to talk about is in the complexity. And you'll see later on in the demonstration, this complexity is often driven by personal intervention. What individuals are doing using Excel and other desktop tools and not necessarily leveraging either the capabilities of the tools they have at hand, or now leveraging this new technology that's actually going to help put things together. So really looking at that transformation and applying controls that right now again, we rely on humans to enforce and not necessarily the most efficient way to approach it or effective.
Greg Fritsky: That's right. And in terms of reducing costs and increase scale, hear this oftentimes, your own organizations are probably looking for how to do more with less, how to get costs down. The only real way to do that is to automate and standardized, but how exactly do I achieve that? Many of you may have tried to go to the shared service model. Maybe you've outsourced it to a lower cost center, the lift and shift strategy is kind of played itself out. At some point, people are starting now to explore how do I bring some of this control and processes back? I can't add people, so automation, leveraging technology is really my only solution.
And the other component is making the work challenging. We have a whole generation of folks coming up, including my own son who's studying accounting, but he's also studying management information systems. He probably spent more of his time programming than he does in the intermediate accounting classes, just because he's going to be building the solutions of the future. And this is where I see the transformation of our own teams and staffs here at Eisner as people are looking to leverage these technologies and learn how to automate things like how do I do an audit more efficiently? How can I leverage to do tax preparation, we're in the middle of busy season here. How do we just build that kind of grandiose scale?
Steve Palomino: And also the value add, as an audit firm and accounting firm, we hear often from both external and internal audit clients, which is they want more value, they want to see more information brought out of this. So they are not looking for just the standard, you check the box sort of activities. They're really looking for insight into the business. And given the amount of work papers and load that each organization has, it becomes more and more of a challenge. So again, using automation and standardization is going to be the key to moving that needle forward.
Greg Fritsky: That's right. And as these jobs change, and these roles change, we need to reassign the tasks, redeploy and basically retrain people. So there's a number of different things happening here. This slide right here just kind of illustrates a sample of many different financial and accounting processes. Things around record a report and closing your books. There's more transactional activities around procure to pay in order to cash. All of these have in common, they're very much rules base, they're established, you have policies, procedures in place. And people execute these tasks end to end, meaning any of these types of like, for instance, you could take tax provisioning. Taking that process start to end from pulling in data, putting into Excel, doing something, transforming that data, sending emails, taking that information once it's been analyzed and then making some sort of posting or some sort of entry. That's sort of an end to end chain of events.
Steve Palomino: Well, the other thing to Greg when you think about the topic of this which is the hands we close and trying to make your process more efficient. If you look at tax provisions for instance, the work that goes on, the information is often available but there are a large number of individuals who have to work it and then your auditors, your tax support groups, ultimately, to be able to support whatever financial information do you need to or financial activities, you need to post to the balance sheet. Many of those tasks could be automated or made more efficient through robotics or other sort of technologies. So again, as you start to think about this, not in terms of towers, financial close or fixed assets or so on, but also looking at operational activities. All the core activities that occur within your business that ultimately result in that financial statement, that P&L or that balance sheet position that you have to present to either regulators or to your investors.
Greg Fritsky: And the other thing to think about too with each of these activities, like I said before, if you can sit there and define any of these things end to end, they'll lend themselves to process automation. The fact is that when people come to us, and we have conversations, it's usually around, where do I start? What are my best use cases? And oftentimes the answer is typically, where's the pain? Where are people spending inordinate amount of time? Where are folks ... how to stay late last night, because they're just gathering data from multiple systems performing numerous reconciliations, just to get comfortable that the data is correct. When you're in the business of data management, and you're not spending the time analyzing what the data is telling you, that's probably a good place to start. And oftentimes, they'll be not necessarily high transactional items, but just where people spending the time.
Steve Palomino: And also related to risk. That's another area that you can focus on from whether it be a regulatory reporting or order to cash is a great one. And how do you necessarily know that you're collecting all the cash you should? How do you necessarily know that your commission payments are accurate and timely and then you're not putting your organization at jeopardy around those types of activities. Right now we rely on individuals, checklists, reports and so on, not in internal and external auditors to help us with that. But it tends to be after the fact as a business owner and as a person who produce reports in the past, I'd like to have known before the test at least an idea of what happened or rather than all the answers after the exam.
Greg Fritsky: Steve and I had spent quite a bit of time over the years traveling around speaking with CFOs, controllers, accounting managers about their processes. We've actually tabulated a lot of the results of those activities that they perform. And one thing that was telling is that again, and again, it's the same type of activity seem to come to the surface and things around reconciliations, inner company transaction and journal entries tend to take consume a great deal of time.
Steve Palomino: Which is fascinating in itself in that 2019 those are the same issues that we face as accountants and business managers and business process owners that we faced 15, 20 years ago.
Greg Fritsky: That's right. So with that said, we're going to have our first polling question.
Moderator: Polling Question.
Greg Fritsky: We have a good representation today for those of you who joined, I actually counted I think 28 of the 50 states and two provinces in Canada. So thanks again for joining us today.
Steve Palomino: And please take the time to respond to our questions. And also if you have any questions so far, we do know this is a webinar and understand that but we'd love to respond any questions that you might have, as we present this information would be glad to respond to those if we can.
Moderator: Okay. I am now going to close the poll and share the results.
Greg Fritsky: Okay, interesting. Yep. So reconciliation goes right to the top and that we see again and again. And this kind of gets back to the point about managing data. Really, why do we do reconciliations. I've done them, Steve's done them, we spent a lot of time. Just trying to get comfortable that the information's aggregated correctly, that the information has been validated. And then even after you do your reconciliations, you always have that feeling that something doesn't feel right, I got to send it to two other people. And we can never say seem to get to a source of the truth.
Steve Palomino: The thing too about reconciliations is again, I started as a staff accountant in 1997, which doesn't seem like that long ago. But it is, the reality is we are still doing most and again whether you're small, medium, large, public or private company, reconciliations still are managed in Excel. Again, it's not a bad tool, but is not the most efficient tool, and they're still managed by individuals, very seldom do organizations really leverage tools, even when they have tools such as BlackLine, or RX, or one of these other applications that supports those activities, you tend to not leverage that tool as efficiently as you could, and you end up still relying on individuals to do those types of activities. So this is interesting that we're seeing this, the same sort of finding or response.
Greg Fritsky: So with that, that's very interesting. So, let's look at some numbers, some interesting telling figures that we've seen over time. This poll actually indicates that transaction processing, just doing is 49% of the activities that we perform.
Steve Palomino: Which is where accountants are spending their time. So again, when you look at from the APQC, you're not unlike your peers, and this is also industry agnostic. So it's not ... we also run that to you Greg, where organizations say, is it just an industry X or Y, again, it's a process, is part of running and managing your businesses.
Greg Fritsky: And interestingly, the controls, decision support, management support more around the analytics, or to a lesser degree, and I'll tell you again and again, we'll hear people say, they really want to spend less time doing and more around the analytics, as data analytics is keep talking about that, how we can use that, the reality is, it will become more and more part of the accountant's role as we start becoming ... building these models that the business will now own because finance is the stewards of the data. So the reality is a lot of that information is stored here, and who better to project and validate and share it than the folks in the accounting office.
Steve Palomino: No, I think we have some questions that I'd like to respond to real quick, which is one that came in, how do we overcome the pushback on RPA due to lack of knowledge of RPAs? And one of the things that and this is a common question, which is why I wanted to respond to it. One of the challenges, but not one, but there are many, which is where to start? How do we deal with this pushback? What Greg and I like to do is start with, what are pain points? What is your environment? What is your roadmap? What are your IT strategies or business strategies, and then look at what technology might be available to support you.
Oftentimes, what we see is if you're starting in Excel, right now, if you're in this transaction processing world, you're probably not going to jump from people in Excel to RPA, you're probably going to need some way to make that journey, whether it be using BlackLine, or some other tool, adaptive for planning, whatever that may be, in marching towards that RPA environment. So it's that ... I think that's where organizations run into that pushback, because they say leadership, and rightly so says, "Hey, gosh, right now, we can't even get past Excel, how are we going to jump to RPA?" So that's how we approach it is take that journey.
Greg Fritsky: In terms of another chart here, you see a lot of the time is being spent on manual postings, manual effort. Oftentimes, systems interfaces and data quality as well, in our companies, half that time, so productivity is challenged. And I think this is kind of where we're going to kind of go to is what does this all mean from a monthly close perspective and the time that it spends. I don't know, each individual in a organization, you'll have a chance to respond in a minute, but we see the top performers typically around, five days to be able to prepare their close and consolidate their statements, medium performance all around, six to seven days. And then the last tier is really around 10 days or more, where you're spending two weeks or greater to get the work in hand.
Steve Palomino: Right. And what else is misleading about this chart on the top performers, we actually ... Greg and I worked with an organization that boasted one day close and actually did have a one day close. But what we ultimately found out is, it took them about 30,000 person hours to accomplish that one day close. So they were able to do it in one day, but it actually took quite a bit of effort to get there, and through automation, they were able to reap some serious benefits around that.
Greg Fritsky: And then, chaos of the close, maybe this is your organization. Many people have said, it's a very negative experience. 60% report increased stress levels, 36% reported impacts personal relationships. That's sad, but 18% report missing regular life events, and I've been guilty of that in the past. But the reality is, it's a challenge. And this is an area where it can be helped if it's focused.
Steve Palomino: Yeah. I worked for a very large company, their logo is blue, and they provide telecom support or communication support. And I close the books there 60 times. And I have to say that from the first time I was introduced to it to when I left in a long performance close activities, that it never got easier. As a matter of fact, I think over time, it got more complex and more challenging. And I most certainly missed more and more family events. So I went into consulting which seemed like an easier way to go with the time.
Greg Fritsky: Yeah, unfortunately, consulting busy season doesn't end in April.
Steve Palomino: Correct.
Greg Fritsky: And lastly, I want to just point out the three out of four folks are not confident in their close process. And this kind of gets back to the whole transparency and getting comfort and controls around the close. Oftentimes, people say, I would like to get some sort of technology or automation in place just so I can really understand and track some of the benchmarks and statistics around, where are we spending our time? How can we improve? Where are the trends? And ultimately, most folks when they're running things on Excel, and people are just running out of time, that stuff doesn't get captured.
Moderator: Polling Question.
Greg Fritsky: One of the things I just want to mention that, Steve's actually based in Dallas, we also have offices based in Miami and San Francisco as well. I'm going to be spending some time on the West Coast, for our clients have joined from California. We'd be happy to stop in and pay you a visit if you have any questions. We'll be making the West Coast tour next week.
Moderator: Okay, now I will be closing the poll and sharing the result.
Greg Fritsky: Okay. So, it looks like we got kind of most of the folks in the moderate to high effort zone. Okay, 53%. And we still got a good quarter there greater than 10 days. And yeah, I would say nearly 75-79%.
Steve Palomino: Well, I love this, hobby statistics is my thing. So basically, if you look at this, we've got a bell curve, and you've got one standard deviation on each side from the bell curve. So the reality is that most organizations are sitting right there in that middle of the bell curve, which means that most organizations are people on this call, who responded to this query, this question, have an opportunity to really make a difference in the world and push yourself over, forward into that one standard deviation beyond the mean, it can be done.
Greg Fritsky: And the one thing I would add is oftentimes, when we would speak to folks about the close, you oftentimes think about it, where I actually have all the data points that I need, but oftentimes accounting is responsible for those activities. I call them pre-close activities, if you were to include those pre-close activities, that could be two, three, maybe even five days, and they actually push you more into a 10-day close category. So all oftentimes consider that as part of those pre-close activities should be kind of factored in as you're evaluating how much effort you really putting into it, even if it's other teams that provide you with your data points.
Steve Palomino: All right. Well, we have two questions that actually play well into this, which is the impact of process improvement. If we think about where we started, where I started my current accounting, which is in the inefficient world, where production is where we spend most of our time. And when we talk production, the question comes up, and this is a very old slide or graphic of the world that we exist in. And nothing new to us, I don't know who created it, but someone did long time ago. But the bottom line is, the question I get is, hey, why does production and control go ... Why are they so large?
Well, we have people that are doing the work, and therefore have to have lots of control is around it. What we want to see in process improvement in through automation and robotics now is driving that from inefficient to efficient. So, if we think about right now, process improvements made some changes, it's made some impact by using Lean Six Sigma and other things, but we've never move that needle. And so, the two questions actually are very related. One of the questions is, at what point do you move from Excel to other products? And the other one is, hey, we need a tool more powerful than Excel to do our work. And we want something that has some controls and helps the Quality Enhancement. And again, that's where we want to go from inefficient to efficient.
And so when you ... I would say that if you're in a world right now where production is a large amount of your work, and the cost benefits of applying some sort of application that does this work for you, and you actually can measure it and prove it, you're at a point where you want to start doing that. There are a number of tools out there, as a matter of fact, Greg and I'd be glad to reach out and talk to you about it is to what's out there. There's no one solution. We're agnostic, we're glad to work with them all. But the bottom line is that, once you start to see that need, it is time to start to investigate what's out there. So, as we think about robotic process automation and how it applies, on the left, it's really what is it? So what is robotic process automation?
You have organizations, and I've worked with many that say, we're already automated, and they consider automation is that I get up at 3:00 in the morning and kick off a bunch of reports or I work all night long to get these reports done, or to try to complete my journal entries, and then, load them into QuickBooks. But really, robotic process automation is really the application of this technology that allows in essence technology, or this bot is we've called it to emulate the human and that's where we're seeing a significant shift in the environment.
The bot really can capture and interpret what's going on in the environment, how data and information is interacting, often describe it is, its very similar to where if I was going to sit down and teach a new account or somebody join my group or my team, here's what you need to do, here are the 10, or 20, or 50, things that you need to accomplish on a daily or weekly basis to complete your job.
And that's really part of that robotic process automation. And it is most certainly focused on standard activities like reconciliation, that's a great one. Applying cash to your GL, to your trial balance, doing three way matches on invoices, you would think that it already happens in SAP and Oracle super efficiently after all these years. The reality is it does not. And so that's in a nutshell, what RPA is.
Let's talk about the evolution of RPA artificial intelligence. On the left side is task automation, and this is where you have some of these applications like automation, anywhere Blue Prism, who started in that space. They've quickly moved to the right, where they're starting to consider machine learning and so on. But basically, what these machines are able to do is perform manual tasks. And again, I think this is where the largest amount of opportunity exists, this is where you'll reap your greatest benefits at most likely the lowest cost from a technology perspective, machine learning is there, you actually see it at some of these call centers. The phone companies have been using it for years, where if you call right now and complain about your bill, they actually have bots now in the background, that will look at your bill and see where you billed twice or you pass due whatever, and it automatically route you to the right group, that's a form of machine learning where it knows what's going on.
And then there's cognitive AI, there's different applications out there, that's definitely out in the future talking to some PhDs out at Rutgers their belief is it's five to eight years out before you really see it go mainstream, still very expensive, you really see that more in the military arena. So just kind of give you an idea of where we're evolving now. That's not to say that within the next three to five years that you won't see the costs come down, and maybe like Centers of Excellence, or call centers or software versus service type organizations offering some sort of machine learning capability, or even maybe accounting firms, because we're experts in processes, in accounting rules, regulations, we may start to apply that in to our own back office for our clients. So with that, we will jump to a polling question, and we'll give you a few minutes to respond.
Moderator: Polling Question.
Steve Palomino: One of the things to also think about too is that, oftentimes people will say, do I need an application, do I need a bot? So kind of are their capabilities where an application can actually have some of that RPA technology embedded, and they're all applications. We now work with a number of different vendors and technologies, and those capabilities do exist. But there is a point where Excel just doesn't cut it anymore, or QuickBooks Online and you starting to scale up. And there's that kind of view of where do I start? I want to eventually have capabilities, it gives me some cognitive abilities. But again, we're talking about finance processes, accounting processes, very rules based structured data, your Blue Prism, UiPath is technology we actually use here at the firm. And, there's a number of automation anywhere as well.
These are all good technologies, but you have to understand that there's no silver bullet answer. Oftentimes, it's getting a broader view of what are the use cases for why I want to do what I'm doing. And that'll sort of lead you to the solution.
Moderator: We're now closing the poll and sharing the results.
Greg Fritsky: Okay. So again, it all gets back to spending the time, doing a lot of manual, repetitive tasks, activities, that we continue to try to kick away at and try to get something in place to automate. But we're still a 57% complexity of business, lack of resources a little bit to a lesser degree.
Steve Palomino: And this is actually a great insight into what we hear all the time, which is, will walk in organizations, and I've interviewed hundreds and hundreds of individuals accountants, operational folks, looked at thousands of processes now. And one of the first things is, well, we're so complex. The reality is that even if your organization is complex, it is a smaller part of your bottleneck. What really is the bottleneck is the manual effort surrounding it. And that's why I'm absolutely confident that the RPA or robotics in this transformation in a way this is so hard, is because there is so much opportunity to focus on that.
And one of the learning objectives is identifying month end close processes that can be fully automated, this is where we're focused on. It's in this manual arena where you have individuals, even if you only have two or three folks in your shop, who are doing this kind of work, because reality too, is lack of resources is becoming more and more of a challenge as less and less individuals are interested in joining the accounting world becoming CPAs part of the cool kids, like Greg and I, it's become more and more of a challenge over years.
Greg Fritsky: I didn't know we were cool.
Steve Palomino: We're totally cool.
Greg Fritsky: So sort of illustrate this, what does that mean? What can I automate? Where's the capabilities can it be applied?
Steve Palomino: And how do you look at it Greg? I think to answer the point, change your mindset. If it's painful, that's probably a good place to start thinking about, it may not be the candidate. But that's definitely what you're looking at here.
Greg Fritsky: So here's just one of many many processes, just to kind of walk through with you, what are the steps? How do we actually ... how can we evaluate a use case? And this again, this is like, one of the top questions we get is, where should we start, what's a good process to pick? Cash comes up quite a bit from a controls perspective, as much as it's very much still manual. It's through the activities. And in this case, you could see, you could do some sort of Whiteboard exercise, build your boxes out on a flowchart of what are those steps that someone's doing? How are you doing it today, and that's a combination of running programs, extracting data, downloading information into Excel. In this case, there's extending manual matching with the invoicing, downloading the reports to Excel, extending that manual match in the customer requests matching info from vendors, sending emails, match that information, update the task list.
So this is just a fairly generic process. And the goal is to take that future state and just take the hands out of it. All those acts of doing where someone is sitting there going, I call it the swivel chair approach where you have multiple screens up, and I'm an SAP, I'm downloading some information from the bank, and then I have Excel over here, and I've got my pivot tables, and I do 10 steps to get the data where I want it to be before I send it off to Steve to get his approval, but he doesn't agree with it. So he's got his 10 steps. So it's all about taking those touch points out of it. And it's not just posting things. It's also having robotics and automation to do some of the workflow for you and capturing that information and the detail behind it so that you can provide it as audit evidence later.
Steve Palomino: Awesome. So let's talk about how RPA can be applied. And actually, we've got a couple of great questions that came in about the close process and automation, one of them is about, or the question is, do we see more private or public companies in the early closer environment. And in my experience, I've not, I can't say public versus private, I see more public companies that are closer books faster than private. But what I can say the difference I'm seeing is, does leadership see this is an important part of the business. And so I've seen private companies that have really focused on this, and made a difference for whatever reason, than usually has to do with operations are trying to respond to business environments. And on the public side, they're trying to get ahead of their colleagues within the industry, and they're trying to maybe produce the reports faster, or be able to respond to the external environment. So not seen one over the other, but more business needs, what's driving it.
And then the second question is around RPA, which is great. When we start to ask about general system platforms versus, for instance, moving from NetSuite to QuickBooks. Great question, the thing that I love about RPA, the software's that are out there is, you could actually stay within QuickBooks, if your processes could support it as long as the application could scale to meet your needs, you may not necessarily need to move forward, a lot of the male tasks surrounding QuickBooks may be executed by a bot, you may have to do some process changes. So, with that, let's talk about actual showing you the applications that are in there.
Ben D.: Thanks, guys. So for this example, the trade allocation exception process for the company is very manual and time consuming. So because of that, and the large volume of trades and allies on a monthly basis, potential investigations are missed and or overlooked leading to reporting issues. The company would like to make this process more efficient, while both catching and sending out notifications for any potential exceptions. So the assumptions for this bot are based on math rules, where all trades are allocated equally, and total 100%.
EisnerAmper developed an automated process using UiPath RPA technology to perform the following tasks. The bot applies business rules and calculations to check for potential exceptions on the entire data set across all time series. Flags potential exceptions and groups by time series, create a report detailing number of trades, investigations and notifications issued. It populates a bar chart for the above reports for each time series, and sends out emails, including the exception reports, analysis and bar charts for every point in time the bar is run, which in this case is on three months.
Steve Palomino: So let's talk about that a little bit from the bot demonstration. And as we lead into it. Think about again, from the art of possible, you're pulling data, which that's what we do a lot of that, you're doing something, transforming it, which is what we do a lot of, you're putting it somewhere, usually another spreadsheet or another tab in the spreadsheet, you're producing a report based on the analytics. And then you're sending it to somebody usually by email, you might drop it, put it in a shelf holder. So that's again, what we want to illustrate, but think overall from the closed process, all the activities that this illustrates with just one bot demonstration. And with that, Ben, I'll let you finish your illustration.
Ben D.: Thanks, Steve. So with that, we'll click play on the video. Instead of begin, we're going to look at the three original data sets just to show that nothing was tampered with before we actually run the bot. As you can see, we did three times series for each month just to kind of show a little bit that this could be run at different points in time. The last report we look at is actually the shell that will be populated once the information is done for the bar charts. And the last thing we'll look at before running is a test email account, which currently contains no email messages. Will now click run on the bot and see what happens.
So all calculations are performed on the front end for this example. Once January data is finished, the bot is prompted to enter an email address. Once the emails entered, the bot will finished running the calculations on the other months and populate the bar chart.
Steve Palomino: We also must note that we did slow the video down so that you could actually see the robot doing something.
Ben D.: Which it actually just finished execution. And here are four emails with all the reports that we just discussed. These emails all contain a customizable subject and body line depending on your firm's needs. And I'm just going to go and open all of these reports. So we can look at them at once to see all the work that was done.
We'll first look at January. And this is a summary of all the investigations from each tab and these are the group data, which again was flagged based off of the business rules and calculations we had given the bot to begin with. And after looking at all three months, we'll look at the report that is populated, and often will be sent out to external auditors.
Steve Palomino: Or business leaders or-
Ben D.: Whoever. Yes. So as Steve pointed out, this video was about three minutes long, but the actual demonstration from the bot was only about 20 seconds. So for this example, the bot was running a smaller sample data set, but could be run on thousands of rows and at any point in time.
Steve Palomino: Excellent. Thank you, Ben.
Moderator: Polling Question.
Greg Fritsky: Again, that was software that we use UiPath software. We work with a number of different RPA providers. But again, there's a number of applications as well, we're focus on, I don't know. BlackLine, Workiva, Central, are just some from a sample. I encourage you to go out, and you can actually view the demonstration that you just saw, we're going to have that out on our website. And we'll share that information with you at the conclusion of this event.
Moderator: Okay. And I will now close the question and share the results.
Greg Fritsky: Okay. So overwhelmingly, the answer is obviously no. Folks are probably thinking about it, but don't necessarily have a transformation strategy. For those that do, it would be interesting to know where you are in your journey. A lot of folks that I've spoken to it's, they've educated themselves, they may even be in a point where they've done some sort of proof of concept. And there are a handful that have actually successfully implemented, and those that have there's success stories behind that is actually amazing. I've been to a few conferences where your peers who have implemented have succeeded beyond their wildest dreams. So there's a real opportunity here to transform your organization, just using a technology that's fairly commodities at this point, meaning it's fairly priced as well.
Steve Palomino: All right. With that too, we have a couple of questions that I want to address. One of them since we had a chance to look at what the bot did and how it executed, there's question of what's the difference between a bot and a macro? And I guess that would be the same as asking, what's the difference between a modern vehicle and perhaps a horse. Both will pull the load, both will do it on command, but one does it way more efficiently and effectively than the other.
So macros, I've worked with him for years, and I don't dislike him, but they contend to get very complex. And when they fail, the debugging process can be very very challenging, where a bot will actually tell you where it failed, it can also tell you what part of the process it failed. For instance, it went to reach out to a URL to pull some information. And the URL no longer exists. And a macro just fail and you have to go through the debugging, where the bot will actually say, unable to find URL. And so it makes it a lot more efficient.
Greg Fritsky: And one thing to add on that too, is the UiPath really integrates well with a bunch of different software's. Whereas a macro, there's kind of been issues kind of whether it's pulling data or anything like that. So just the capabilities and kind of Steve said the efficiencies saved on both the development and on the end product, they're just a lot different with bot.
Steve Palomino: Excellent.
Ben D.: So again, folks are oftentimes asked, where do we kind of start and how do we get governance and control around this? We emphasis putting the folks who learn the technology that are in from the beginning become part of the program. But the nice thing about this model, is that it's really a convergence of different backgrounds, if you will. As folks, obviously from the finance and business should be absolutely engaged, oftentimes people say, well, as I robotize these processes, how did the folks in the prior roles, how can they get involved and engaged, and oftentimes, we say, they're going to be the ones that are going to designed the capabilities of future and be part of a robotic Center of Excellence, if you will. Designing with the flow should be, how to further enhance them, how to transform them, how to standardize and maybe you're looking at some sort of shared service model, so you can actually add to your scale. Include the finance and business as part of the program.
But just as importantly, is making sure those who are in the control side, the risk management, audit people, as well. This is a powerful tool, we were pretty early stages of when we got into the whole RPA space, we saw the capabilities around managing controls and automated controls, continuous controls monitoring, that all speaks well here because you're talking about automation. And a lot of those controls are embedded in your processes, this is an opportunity to enhance those controls. And I just want to add too to my colleague stated about the software's, the big difference between these software versus the macro, is the fact that also the enterprise versions of these softwares will collect and capture a lot of the audit information that can then be handed off from provided to audit and provide confidence that it was the activities were performed accurately and completely.
Steve Palomino: Exactly. So let's look at a simple one macro versus a bot. I have an Active Directory, and I have an HR system. And I have ADD changes deletes in the HR system, a bot, I mean, a macro could probably take data into a VLOOKUP and compare where a bot could actually go out and look real time at any changes in that HR database and compare it in real time to the Active Directory and be able to then send an email to Ben and say, "Hey Ben, Steve no longer exists in the HR environment, please delete him". Or that bot actually can be charged with deactivating that user out of the Active Directory and sending a note to the owner of the Active Directory, a macro and Excel could never do that. The bot could and also the bot doesn't have to API or talk to the back of the computer, it can do it as a user.
Greg Fritsky: And someone who has built both macros and these bots, it's just in terms of development time, it's really not even close on our end. So that's something to consider as well. Building the bot takes a lot less time just because they are more user friendly.
Steve Palomino: Which is hard to believe that the building these bots actually is easier. And as person is coding Visual Basic, I have to say that I like working with the bots.
Ben D.: I would also say, I have here analysts, that's oftentimes, the rules are changing. I often say to even our folks here, if you're an internal audit practice, audit practice, your roles are going to change, you're looking at data more in real time now. You're looking at predictive trends, what better way to manage risk on a continuous basis. So you become more of an analyst, you're more taking the knowledge that you have and an understanding of gap accounting or what have you and start applying these technologies to get insights.
So finally, I want to talk a little bit about this, how do I segment these processes and activities, and where do we start, and oftentimes, we look at it from a couple different perspectives. Oftentimes, we always emphasize to start taking activity that's causing a lot of pain, but it's usually, it sums up is where people spending the most time, you can kind of break it into four parts here. And those processes that involve the most people, and that are considered less complex, that's really a good place to start.
Oftentimes, people say, well, I want to throw the most complicated cumbersome process I can think of, and the reality is, yes, you can automate a lot of that, but the reality is, when you go down that path, and that project begins to take on a life of its own, it becomes a real challenge to win over people.
Steve Palomino: And the question earlier about the pushback, the thing that I found is success that overcomes pushback, that the faster an organization or leadership sees a result and more importantly, the process owners, the quicker they see, oh my gosh, that actually did happen. It did work, the adoption rate, it accelerates exponentially.
Ben D.: If you have a lot of manual journal entries, it's a good ... Where there's smoke, there's fire. Those are good places to look. Again, we saw reconciliations is a key area focus.
Steve Palomino: And a question is why all the journal entries? It allows you to now go upstream to the data, there was question about other examples for robotics? Well, instead of waiting for the trial balance data that you know is usually the middle month is going to be wrong. Again, using that bot to identify that problem further up the process chain, and be able to effect or impact that information, perhaps get better information. So you don't have to post a journal entry. So those would be all opportunities where you can leverage bots throughout that whole information chain, if you think of it as a factory floor. Instead of waiting to find out there's no tires to put on the car, you realize it when the car still at the point where it doesn't need tires.
Ben D.: That's right. That's correct. And then once you've learned from the technology, you can learn and expand from there. You'd be surprised how quick you win over people once they see how much work can be eliminated from their monthly close.
Steve Palomino: And then when we talk about the transformation journey. There was a question out there, do we need IT? Do we need ... is it finance owned? Most certainly is a journey, if you're in the premise world, and on-prem with legacy systems, IT will always be involved, but it may be more involved in then. The bot technology automation, and you were Blue Prism, UiPath, there's a number out there. They'll be more every day, and then eventually they'll probably be a consolidation in the environment. But the bottom line is that it's tends to be a business own technology with IT supporting is a move into that environment.
So you normally the transformation is going to be from an on-prem or premise technology, or people based to the next generation of cloud based technology, some sort of reconciliation tool or whatever. Like the question was earlier, what application would help us do our world better, you'd go to that, and then ultimately moving into the RPA. Now this can happen very quickly, but I still think you're going to have to go that journey. You probably not going to go from people to RPA.
Ben D.: That's right. I mean, it is a journey. Ultimately, you need to define what your outcomes are. You have to understand where do you want to end up as an organization. Most people just want to be able to drive change, have more value to their work, and honestly, just go home at a reasonable hour.
Steve Palomino: And also, one of the things that a question or comment that always is brought up to Greg and I with chief accounting officer, chief audit officers, is what keeps them up at night. I think what keeps organizations up at night is being in this premise technology, this people driven technology, they still can't look anyone in the eye and feel very comfortable that all things are working as they're supposed to.
Ben D.: That's right. It keeps you up at night. So at this point, we wanted to ... we have some time here, five more minutes, we want to take some questions.
Greg Fritsky: All right. Thanks, guys. So the first question would be, what's a good first step to establish an RPA program?
Steve Palomino: One of the things that Greg and I have done, which is proven very effective, is we like to do a session where we sit with the organization and find out what are your needs? What are your issues? Where are you in that journey? Again, we have organizations that come to us and want our RPA right now, and they're not ready for it. And so that's really the process. Greg.
Greg Fritsky: Yeah. I mean, look, at the end of the day, it's all about doing more with less so looking for activities that perhaps really kept you from expanding and growing. Where the people spend their time, where people are might even be complaining, at the end of the day, I don't feel like ... I went to school, I got my CPA, I'm spending enormous amount of time being a data entry person. And the reality is, I want to do more with analytics, help me eliminate that 80% of the process, and that's okay. Again, you don't need to sit there and focus on 100%. If you can even get to 60% 70%, you're already in a better place.
Ben D.: Excellent. All right, the next one will be how do you identify which process areas are ideal candidates for automation?
Steve Palomino: Well, Greg refer to that slide. Basically, the areas where you get the most benefit, and it's the easiest to implement. And Ben you and I have looked at that where organizations want us to do some very complex, bought building, and we channeling it into an area where they're able to reap more benefits with less effort, ie less cost.
Greg Fritsky: We love our Excel spreadsheets, but oftentimes I would say, anywhere where ... the more touch points in those spreadsheets, the more opportunity for error, as an ex-auditor, myself can attest to. I always get nervous around spreadsheets. So the ability to eliminate a lot of those touch points as much as possible. Obviously, not everything can be automated, but the reality is taking out key controls as part of that equation is very important. So, that would be my area of focus.
Steve Palomino: Excellent. And I think in our last couple of minutes, questions came in which I like again, I like all questions, but this is a good one, a cost of implementing RPA arrange, it can be fairly inexpensive with some of the applications, just a few thousand dollars. If you have someone in your shop, and you want to buy a license, and they're willing to learn it, to tens of thousands of dollars. It depends on the needs of your business. So probably 3,000 or 4,000 at the lowest ends, if you were going to buy it yourself and have someone in your shop build it themselves, all the way out to tens of thousands of dollars a year, it just depends on what your organization is looking to buy it off.
Ben D.: Great. All right. And the final question. So what are some key things to consider when trying to establish an RPA program in terms of people, process and technology?
Steve Palomino: I would say organization on people, what is your organization's personality? What processes existed or don't exist? What's there, what's not there around technology? What are your core technologies? How important are they in business? And what's the risk of messing or perhaps interacting with those applications? What's the risk to the business of failure around that? Greg.
Greg Fritsky: I would just add that the important thing is your people and making sure that you're transparent and clear as to what their roles are going to be. Make them part of defining what that solution is going to look like. And getting them involved early is important. Ultimately, this is about ... it is a cost driver, but it's also about enhancing value and making people feel like they're doing meaningful work in its simplest time. So definitely get people in the business involved the early in the discussion.
Steve Palomino: And don't leave IT out. I think it's important.
Greg Fritsky: And IT should not be left out. We've been on both sides of that equation, but IT can be your best friends. The focus is or the focus should be on, how are you improving the business and most importantly, this is business enabling software. So this is something that they also enjoy the idea that the business will manage it and they thought-
Steve Palomino: It frees them up. They end up with less work, not more.
Greg Fritsky: Yeah, and your auditors will be happier because there'll be less testing necessary because of the fact that there's going to be more automation and more controls-
Steve Palomino: More reliance on the controls.
Greg Fritsky: ... and you'll feel better and be able to sleep at night.
Steve Palomino: Exactly.
Transcribed by Rev.com
If you have any questions, we'd like to hear from you.
Explore More Insights
IRS Enforcement of Digital Assets is on the Rise | Tax Guidance and ExpectationsRead More
Receive the latest business insights, analysis, and perspectives from EisnerAmper professionals.