RPA Bots: Bogus or Benefit? Bringing RPA to IBM i/AS400

Pat Cameron:                   Thank you for joining us today for today's webinar on RPA Bots: Bogus or Benefit? Bringing RPA to IBM i. Today's webinar will be recorded and you'll be receiving a link after today's session along with some additional resources and information about RPA. Our subject matter today is a little bit of RPA in general, what is it? What are the benefits of using RPA, and specifically, how can I use it with my IBM i/AS400 in that environment?

                                            My name is Pat Cameron. I'm a director of automation technology here at HelpSystems and I've been working with the Automate product for about six years. Previously, I worked with some of our IBM i specific products here at HelpSystems as well as I was an operations manager at a hospital here. So I've worked with the IBM platform for about 20 years, I guess. Currently I work with customers and help them to implement and train on several of our different product lines. I'm joined today by Roger Hammer from CM First. Roger, welcome and tell us a little bit about yourself.

Roger Hammer:               Thank you, Pat. Sure. I'm the director of software development and services at CM First Group. I manage the internal software development for a few products. CM evolveIT, an analysis product in particular, and also manage our services that we provide for customers in the software development area, particularly in the mainframe, and in IBM i. The Broadcom CA Plex, and CA 2E areas specifically. I've been working in this area for about 15 years.

Pat Cameron:                   All right, we've got some experience here, hey, Roger? So today's agenda, we're going to cover a little bit about our companies, what we do and then we'll discuss RPA in general and how companies use it to improve their productivity, and specifically how RPA benefits your IBM i operations. We also have a few use cases that we will review with you to show you how some of our customers are using RPA with IBM i. If you have any questions during today's session, please send them in the box on the side of your screen, and we'll answer them at the end of the session. If we do run out of time, we'll get back to you after today's session with the answers to those questions.

                                            Today's webinar's planned for about 45 minutes and we'll leave 10 or 15 minutes at the end for questions. So a little bit about HelpSystems. I see some familiar names out there who are customers. Welcome. It's always nice to see some familiar names at any of these webinars. So HelpSystems now focuses on two areas in our product portfolio. One of those areas is security.

                                            We protect business critical data with automated security solutions that help you stay ahead of today's ever changing threats. We have security products for IBM i as well as products for other platforms. Then our other focus is automation. Automate processes with RPA, workloads automation, and file transfer automation to increase productivity and add value. Our products are more considered IT utilities and middleware rather than applications.

                                            We have several system schedulers, system monitoring products and file management products. Because the company was started over 30 years ago and focused on the time at the System/38, if any of you remember that platform, we really are experts in IBM i platforms as well as distributed systems. So please keep HelpSystems in mind for your security and automation projects.

Roger Hammer:               CM First Group is headquartered in Austin, Texas and we have a number of partners around the world with offices in France and Switzerland. We partner with Broadcom CA, IBM, and of course HelpSystems and we have a number of enterprises, as you can see, that we have worked with over the last many years. We provide analysis tools to the IBM i and mainframe markets. We provide change management tools, and also web extension capabilities and mobile extensions to the IBM i specific spaces.

Pat Cameron:                   So before we get into the nitty gritty, let's get a little bit of information from you. So I'm going to open up a polling question here. If I can figure out how to open it. So hopefully you can see that on your screen. This is just a question about where is your organization in the RPA adoption process? Are you just getting started by doing some research for RPA? Are you evaluating some RPA solutions and vendors? Are you in the middle of implementation of RPA? Do you have an RPA solution already in place, and are you expanding or scaling your current RPA products?

                                            I'll give you just a minute to put your answer in there, and then we'll share them with the rest of you. So it's always nice to kind of know who your audience is when we're doing one of these webinars, kind of see how relevant RPA is to your operations. All right, thank you very much. That wasn't what I expected to see. Oh, let me close this poll. Bear with me a minute. Sure.

                                            So it looks like 58% of you are just getting started researching. This will be a great kind of an overview for you. Some are evaluating, implementing, some people have an RPA solution in place and some are expanding. So we've got a nice kind of a wide audience here of where you're at with your RPA solution. Perfect. Thank you for that.

                                            So when Robotic Process Automation or RPA first arrived on the scene, The term was almost exclusively used to refer to the manipulation of systems and applications at the user interface level. That's where the robotic part comes in. The robot interacts with systems the same way that a human does. Clicking buttons, typing login credentials into a website, reading words off of a PDF file and so on. This definition of Robotic Process Automation is still at the core of what RPA does.

                                            However, as the technology has matured, it's increasingly being presented as more of a comprehensive enterprise automation solution, one that can interact with a GUI, some type of a graphical user interface, but also one that provides deep integrations and sophisticated enterprise class features. This shift in focus also expands the benefits of robotic process automation software, from shaving a few hours off a user's desktop tasks to powering an enterprise wide digital transformation.

                                            A couple of terms that you might hear are software robot, and that's usually considered a piece of software capable of completing complex repeated processes, typically performed by a human. Then, with those software robots, you can create a digital workforce, and that's a scalable team of software robots working together with the human workforce to manage business processes. Business Process Automation and Robotic Process Automation are often used synonymously. RPA puts a larger focus on automation at the user interface level, but both BPA and RPA emphasize ease of use for non programmers and interfaces with cloud and web services.

                                            RPA is an important tool for digital transformation that helps your organization work more efficiently. It's important to remember that RPA and digital transformation aren't meant to replace human employees, but to help them work more innovative ways. By using a team of digital assistants to help employees with their most repetitive, time consuming manual tasks, we're free then to focus on more value adding tasks and more strategic projects.

                                            RPA bots not only streamline your processes to save time, but they also increase the accuracy of your work to cut down on costly errors. With a bot working alongside every employee, the possibilities for innovation are endless. Once you put RPA to work as a tactical tool to generate value, move into using RPA to provide reliable connections between disparate applications and systems. I see RPA as the great integrator. As much as digital transformation typically means new tools and technologies, there will always be legacy systems and applications that will stick around because their business critical. Or maybe a new system is purchased that doesn't integrate easily with existing systems.

                                            Either way, most organizations tend to have scripts and manual processes attempting to maintain continuity between applications and systems that can't integrate natively to send data back and forth. The right RPA tool can use an API to connect to any system or application, ensuring that data continuity.

                                            So how does RPA benefit users of IBM i? In general, Robotic Process Automation is a type of automation technology that transforms the way businesses operate. RPA boosts productivity, improves accuracy, reduces costs and optimizes resources by handling those tedious, time consuming processes and hands those over to software robots. These RPA bots mimic repetitive, rule based human tasks and communicate with your IBM i and other business systems to streamline business processes.

                                            So as far as boosting productivity, it frees up your staff for more strategic projects. It eliminates daily data entry tasks, Excel management, email management, and then the human only needs to get involved if there's some type of an exception or some type of an error. It improves accuracy. There are no data entry errors. There are no typos or fat fingers. It can verify report totals across different systems, and it always enters the correct date. Always.

                                            So to me that accuracy improvement is almost even bigger than the time saver as far as RPA. Then it also optimizes the resources that you have. Workflows execute without delays, such as a human needing to verify a report total, before you can go on with the next step. We can automate that piece. Dependencies are met and workflows continue and there's no need to schedule hoping that that prerequisite is met. Wherever that prerequisite lies, if it's on a Windows Server and you've got to wait for that to complete before an IBM i program is called, Automate can link those two together very easily.

                                            Then, in addition to optimizing resources, RPA will reduce costs. You can get more done with less staff, your bots work 24/7. There's no sick leave, there's no vacation and digital assistants are available for staff, again, to free them up from those daily kind of mundane, to me, I always think of data entry tasks that most of us are not really excited about doing. Several different approaches to RPA.

                                            For those of you that are researching and just evaluating, kind of keep this in mind. So depending on what it is that you want to do, will determine the approach. So in general, you need to gather your process requirements. Create an outline of the manual steps. A simple outline will do. What does the human do? You need to get that documented. Where's the input from? Is the input coming from Excel, from an email, from a database, from a file? Is there a keyboard entry that's required?

                                            Is the process consistent and repeatable, and does it follow kind of a generally predictable pattern? Doesn't have to be exactly the same way every time. Automate is smart enough to build logic and to go in different directions based on the result from a previous step, but if it's documentable, consistent, repeatable, there's no doubt that it can be automated.

                                            So kind of on the left side of the screen, if you look like at As Is. So what is As Is approach to RPA? What that means is I'm going to take that process that I have, identify the steps, get a subject matter expert, somebody that's doing this task now, and then automate it with interactive robots that mimic those humans steps. This method is often used for a proof of concept. It's very quick, we can get it done easily. You don't have to make any changes to your programs.

                                            There's no programming involved typically. So the benefits of that As is type of approach is, the time to completion is quick. Validation of automation, so we can prove that return a value, we can say, yes, this task can be automated. We might be able to do it in a little bit better way, but absolutely, we can automate it. Then it's easy to validate the requirements that you have and identify that process variability.

                                            So the method for kind of As Is RPA is just interactive, mimic those human keystrokes as they are. As we move over to the right, it gets a little bit more complicated. So more advanced RPA is optimized for automation. The first step is always the same. Identify the steps from a subject matter expert, and then map out the process via maybe there's some back end type of automation that can be involved in this as well.

                                            Maybe there's an API available or some type of a web service that can be requested. Or even direct access to a database through some type of a query or a connection to the database, not needing to be done interactively. There may be interactive steps as well. For example, maybe there's some jobs setup before you submit that job. You need to enter the date range, for example and maybe that piece needs to be done interactively, but the rest of the tasks can be done in the background.

                                            Often this might be maybe more of a second phase of automation, but sometimes it's done upfront. It's going to depend on the process that you want to automate. The benefits of that type of optimized automation are, you'll have improved stability and extensibility. Because we're not dealing with keystrokes on the front side, we're dealing more with an API, often the speed of execution is much quicker. You can also leverage reusable components, you can set up that API, and then you can use it over and over and over again.

                                            There is robust error handling and auditing and often allows for improved security because things are done in the background. So often for that optimization, it's a combination of interactivity, and what we call kind of back of glass automation, sequel automation, API automation. Then way over to the right is a re-work process and this will include process improvement and whatever methodology that you use. Six Sigma or Lean et cetera.

                                            Again, that first step is review the process with the subject matter expert, identify areas of inefficiency, map what you want that process to be, as opposed to what it is right now. Then adjust an update based on those automation capabilities. So the benefits of reworked processes for automation is you are able to address business scale impediments, gain benefits and have improved process and automation. It has a longer cycle to build, but it may lead to higher ROI and waste reduction on the other side because you've eliminated a number of steps.

                                            Again, that method might include a combination of interactivity and back of glass or API type of automation. So a few different ways to take a look at automation and how you can start to implement. Let's get a little bit more feedback from you. Where does your organization have the most need for automation? Let me find this guy.

                                            So where does your organization, as a whole, have the most need for automation? Is there a specific team or do you have a single use case in mind? Are there multiple areas within one department? Are there multiple areas within the organization? Often we'll see like accounting has some automation needs. They spend a lot of time in Excel. IT absolutely has some needs for automation. Those are just a couple of examples. I'll give you a couple more seconds to plug in some answers, and then we'll kind of take a look at where you think you are.

                                            Of course, that's going to depend on where you're at with automation. All right, cool. Thank you. So we do have 6% out there that are looking at a specific use case. Most people are not sure or I guess most people can see multiple areas across the organization. 47% as opposed to 35 are not sure. That absolutely makes sense when you're just starting out.

                                            Cool. Well, thank you so much for that input. Oh, let me see if I can get to the next slide. Sorry, these polling questions kind of throw me off now and then. Let's take a look at some use cases. How are people using Automate in the real world? We'll take a look at a couple of IBM i applications, and also some operating system commands that you can automate.

                                            So one of the latest requests that we've had is the single sign-on. We do offer a service through our security services to assist you in setting up your EIM or your Enterprise Identity Mapping table, and that's set up on the IBM i so that you can use your network logins then to log into the IBM i. Don't have to set up separate users and profiles and passwords.

                                            This makes it a lot easier for users because they just have one user and password that they have to manage. Once you get logged into the network, then you can access your IBM i, start up a 5250 session if you want to, without having to log in again. So the Automate piece of this service syncs the Active Directory entries or an LDAP server with your EIM table over on your IBM i. Let's take a look at this in action. So I've got a little video here.

                                            So this is my single sign-on. So I'm trying to start up a 5250 emulation session, you see I get an error, because I'm not in that EIM table. So this example shows adding a user to my Active Directory, Automate could absolutely do that piece as well. Then we run an Automate task that does that syncing. It takes the users that are in Active Directory, and it populates my EIM table. So the next time I go to login, I can just log in with my network login. So here you see we've added Alan Bentley, and if we take a look at the table, you can see that the association for that user is now his user over on the IBM i.

                                            He's logged into the network, we can log right into the IBM i without having to log in again. So that's a great use case. Helps out IT certainly and every user that needs to log in and doesn't want to remember multiple users and passwords as many of us do these days. Any type of data entry can be automated with Automate. To me, I think it's the biggest use case overall for automation. I really think that automating data entry will get you the quickest ROI.

                                            Data entry or data extraction. So we can write to a screen or read from a screen. Can be automated using any type of interface. A graphical user interface, maybe you've got a client on your workstation, a web browser interface or a green screen. So keep that in mind if you're doing data entry, if you're automating it, it never makes a mistake. The fact that automation eliminates errors to me is a huge benefit to automation. Like I said earlier, bigger than the time saving.

                                            Eliminating errors also eliminates having to rerun jobs, restore files, all those awful tasks that are associated with data entry errors. Automate can be used to extract data from websites and send that data to the new target. Doesn't really matter if that target is the IFS or a physical file on the IBM i or a SQL database or a spreadsheet. Those can be sources of data as well as targets of data. So in this little video, I've got some terminal sessions set up and Automate is running them and doing data entry.

                                            This one has got multiple 5250 sessions and it's going to request some reports. As you can see, these sessions can ... I can run multiple sessions at the same time and actually, they don't have to even be up on the screen. They can be done in the background, but then you won't have anything to look at, and you can see it run very quickly. So this is an example of logging into those sessions and submitting reports, entering a date range for those reports. Instead of having an individual request those reports one at a time, Automate can request them all at the same time, no errors and get work done a lot faster.

                                            Then my last example is how Automate can work with different applications to integrate them and eliminate the manual steps that might be needed for that integration. This task receives the same report from two sources. One is distributed in an email, and the other is a spool file that's placed in a network shared folder. Automate monitors the inbox for the spreadsheet email, and monitors the Network Folder for the spool file. Once both of those reports are received, we've got a workflow then that processes them.

                                            As you can see, data integration from multiple sources is an easy task for Automate to process. Automate can get the data from the source, parse through it, looking for key values, check or verify the data on the reports and then send it over to the new target. So let's take a look at this workflow. So this is multiple Automate tasks. So here's my two reports. One's a spreadsheet, one's a spool file and these totals match at this time.

                                            So I've got a workflow here that monitors an inbox, also monitors a folder and then it executes tasks that will read those reports, check the totals. If they balance, it will go ahead and run a task that will process the invoices. If not, it will notify the help desk. So you can see here they balanced, goes through and runs the invoices. So let's change one of these totals so they don't match anymore. So now they don't balance, and we'll run this report again.

                                            You can see I've got this weight built in so that the next step to check the totals isn't going to run till both of the reports are here. Those reports don't balance right now. So we're going to notify the help desk. This is when a human needs to be involved to see what's the deal. Why don't these reports match? All right, those are my examples. Roger, talk a little bit about automation.

Roger Hammer:               Thank you, Pat. So we're going to talk a little bit about some additional use cases, user provisioning in particular. Pat mentioned single sign-on and Active Directory integration, change requests. Obviously, migration of users or even removal of users are certainly good areas to work with. You may have onboarding of customers, you may have onboarding or provisioning of your own employees that can work in this area. So just a comparison here of a manual approach where the human is doing all the work.

                                            So receiving data from the employee or customer, sending that out to the appropriate person to be entered. That person has to open that up and figure out what they've got. Maybe identify the attached file, get that data entered into the system accurately, hopefully, and then somebody's going to validate that, check for errors and of course, if there are errors, then you've got to go back and correct those errors.

                                            In the green case here, in the automated case, there's no onboarding data management, there's no one needing to manage that information and data is going to flow directly into maybe a location, as Pat said, and then the data is going to be automatically read in by Automate. That information then will be completely entered in an automatic way with no errors, and then no corrections [needed 00:27:06] of course, if there's a fully automatic process.

                                            So we're going to talk about specific use case here demonstration. So this first case is a native application, running through and looking for an input file in a specific network location. Then it's going to walk through a number of rows of information, it's going to import from Excel, and then delete those records, those users from the application. So you can see there on the right, the just a simple copy of the file in this case, which could be an FTP, could be any way that that file is moved to that network location.

                                            Then once that starts, my Automate task is going to recognize it, pop up the application, open up Excel, read that data and begin walking through the process here of identifying that person, deleting that record for that particular user. So as we walk through this, you'll see three different users, in this, case being deleted from the application. It has been slowed down a little bit so you can see it working rather than running quickly.

                                            Then we will, at the end, once this final user is deleted, you'll see on the left in this web screen, a Trello card will get created through an API. So that card will indicate that we've completed this C++ native application task for this delete capability. So now we're going to move on to pure green screen. This was developed using CA 2E. It's going to do the reverse. It's going to take that input file, and now it's going to create new users, looping through all the data, again, from an Excel file.

                                            Recognizing the data flowing into that network location, it's going to run the 5250 emulator screen here, pop up, and you'll see that entry commands. So it's working directly in the entries to get to the right location. Then it's going to pop up the Excel sheet, read the entries, and then enter those one at a time into the fields, and cause those entries to be added into the application. So, as you can see here, we're going to talk about the native app, the 5250, and also web application.

                                            So all of these can be integrated using the Automate RPA tool, and with the ability to generate results for all of them. Again, final result, we get a Trello card, saying that we've completed the task of entry for this green screen. So the last task is a Plex Java web application, doing basically the same thing as the last. It's going to walk through this web application, reading the file input, providing the input for the entry of the users, and then it's going to walk through the web application, providing the same clicks a user would, providing the same data entries that a user would.

                                            One key thing here you'll notice is that screen, web screen popped up kind of on the left, and in a moment, I'm going to move it over to the right. Because you see that I move that screen, you notice that these clicks are not dependent on a location on the screen. It's actually recognizing the location within this web application that it needs to create the entries, and clicking and doing the entries, no matter where that is on the screen.

                                            So we've found the right location. We're going to begin and pop up the command to enter the data. Read it from Excel and begin the entries on this web screen. This is going to walk through each of the entries and then one of the other things you may think about is this is an application for data entry, but you may have backend processes to integrate between different applications both on your IBM i and your native applications or your web applications also.

                                            So all of those can be integrated and it may not be only a data entry. It may be backend processes, reading databases, moving data, all those kinds of tasks that may be needed in your particular instance. So entering the final user, and then we will move on to, of course, creating the Trello card once we've completed this task for the web application. You can see that on the bottom right, all the steps are executing. This is actually kind of in a de-bug mode, so we can watch those steps happen and see what's actually taking place there. There's our third Trello card from the Java web application.

Pat Cameron:                   Cool.

Roger Hammer:               Thank you. So let's move on.

Pat Cameron:                   All right.

Roger Hammer:               All right. So just a little bit of detail here for the web browser automation. You can basically do anything. You're going to be able to open a browser, navigate pages, extract data, you can interact with automating elements that are not HTML based. The Save As, menus, dialogues, any can kind of flash elements. Also, you can input the standard keystrokes as you saw there. You can do mouse movements, clicks, copy/paste, and any keystrokes that you need to and very flexible capability.

                                            One other area, generation of reports and distribution. This, I think has a lot of benefit for most companies. A lot of people have internal reports, management reports or project reports that need to go out to their users, and many times those are generated somewhat manually through Crystal Reports or some SQL process, or an ERP system.

                                            So all of that process can be automated, whether it's multiple different inputs from Excel to those different inputs, and then merged together, maybe in an even better report than is available today with combining the information and then can be automatically emailed or dropped into SharePoint or for our company, at least, we use Teams which is a SharePoint backend. So we can drop those anywhere needed, once those reports are available, and anybody can access them.

Pat Cameron:                   Very cool. All right. One more bunch of questions for you. So now that we gave you a few ideas, which processes do you have the most need for automation? We've got a few examples here. If you could share, that would be great. Application integration, I really think that, that that's what Automate does. It will integrate disparate applications, no matter where they live, get that data from one place to another easily and quickly and securely as well. Data extraction and data entry. Of course, I beat that up pretty much, and user provisioning instead of all those individual steps. Roger, I really liked your demo there showing that onboarding in lots of different applications.

Roger Hammer:               Thank you. So we've also had customers who are maybe bought another company, brought them in, and they need to integrate their old applications or even bring the data over to their current application. So some of that migration functionality.

Pat Cameron:                   I'm doing a webinar this afternoon for a banking user group, and one of our favorite use cases was a bank out in Virginia that acquired some other banks in that migration. He said they never would have gotten it done without Automate. So data migration is a huge application as well. So yes, application integration and data extract and entry, that application integration without having to change your existing applications. I think that's so huge. So thank you very much for that. Hope we gave you some good ideas. Let's talk a little bit about ROI, et cetera.

Roger Hammer:               Yeah. So two areas here we're going to cover. One is understanding what processes or tasks are high value to your RPA modernization or automation. Then once you understand those tasks that are high value, what kind of an ROI might you expect and how to calculate that ROI. I will say that we at CM First and HelpSystems, we can work through this process with you. So the first one here is the onboarding provisioning. You can see we have a grid here with time savings, how critical is it? What's the frequency of it? How much effort to automate, and what's the lifetime of that automation?

                                            How long is it going to last and be used? We can calculate those and now begin to come up with all three of these with a help desk automation and with maybe end of month reporting and see where do these numbers come up? Where do we find the highest value? Also, I would say you want to consider where is the quick, high bang for buck? Where are we going to get started that can provide as we get into the ROI, and provide a lot of value to your business from the very beginning.

Pat Cameron:                   I think this is really a great service to offer people because I think you'll go back and you'll see a bunch of different things that you could automate, but how do I prioritize these? So this can certainly help with that.

Roger Hammer:               So the inputs obviously, are the hours people spend, and how many people are going to do that and what are the hourly rates those people are getting. Then of course, your time to develop. How much time it's going to take to put the application into place, cost of the software, time to maintain and automate and you'll find that that development time is usually really short. It's pretty quick to get these up and running. Particularly those first tasks you'll want to automate. I heard a story where there was a automated task that was very, very repetitive, and had a lot of people doing it, and it was only 12 steps. So it only took a couple of hours to automate it, and it took over 50,000 actions a month in order to ... And now requires no human interaction to do that.

Pat Cameron:                   Awesome.

Roger Hammer:               All right, let's move on to ROI. So from an ROI perspective, we want to understand what's our payback? How quickly are we going to pay back the cost of implementing this? Of course, that is the savings that we're going to see with the automation. So we're going to start with the average hours to perform the task. In this case, we're saying 10 hours for user provisioning. How many people there, what's their rate, and then what's the average ... What do we think cost reduction is?

                                            So in this case, we're saying that 75% of the effort can be reduced through automation. With those totals, we end up with $2,000 a week or 104,000 yearly, and the total cost after that task is only 26,000. So there's a nice 74,000 or excuse me, 78,000 savings with that task. So a nice improvement.

Pat Cameron:                   Nice.

Roger Hammer:               We have two additional examples. The help desk 30 hours a week, in this case with a number of people working that task, a little higher rate. Then half of that time, some of those tasks may require a lot of user consideration. So maybe not all of them can be automated, but many of them, the requests that come can be and so half of them in this case are being automated, and we see a really large, $270,000 cost per year, and with a savings, now that cost goes down to 136,000. So very nice savings there. Third example, the report generation is kind of unique. It's only 10 hours a week for a few people. A little lower rate, but we can automate this entirely.

                                            So we see $20,000 annually saved through this task automation. The final slide here is going to give us kind of a comparison and say, okay, what if we automated all three of these tasks? What kind of a payback are we going to see? So we see a full 397,800 there. We see a very good sized number, a yearly cost and only 162,000. So about a third of the cost from the original, and then we start to see the actual percent of return and we're seeing in the first year over 200%.

                                            It jumps up significantly without the additional costs in the second and third year. 600 plus percent and over 1,000% with very large savings. Now, of course, your tasks are going to be different than these. You may have some tasks that are highly automatable, others that may not be completely able to be automated, but the users can still interact with these automation tasks. All the simple entries and all the tasks that are a little redundant, the automation can take care of while the user completes any consideration, or decisions that are needing made.

Pat Cameron:                    =Awesome. Well, thank you, Roger for all that good practical information on how to prioritize and how can I justify the implementation of a new product. If we can see a quick ROI, that's always a good...

Roger Hammer:                ... to be a part of this webinar. We're very happy to have been a part of this.

Pat Cameron:                   Well, you're welcome. It's a pleasure. We've done a few of these and they've always turned out well. So if you have any questions, we have a few minutes. I think we're kind of on time here, and we can answer them. Oh, let's see. Somebody just sent one in. For single sign-on, how do you handle legacy systems that has a max profile length lower than others? I.e., windows can be up to 20. J.D. Edwards cannot be over 10. I believe that's mapped out in the EIM table. To be honest, I am not really familiar with that piece of it, but that's what that mapping does in that table. We can get you some information on how that works exactly as well.

                                            Here's another one, do I need to install anything on my IBM i for Automate to work with it? You do not. So you don't need to install anything on the IBM i. In the examples that we showed you, we're just using a terminal emulation to enter those keystrokes. So you don't need to install anything on your IBM i for that. If your application provides some type of web service, or an API that we can use, and we also can use the APIs within the IBM i, there's no need to install anything over on the i.

                                            You would just install Automate. So Automate is installed on a Windows Server, and then you would execute your tasks from there and it would communicate over to the i. Let's see. Another question. So what is the ramp up time for a user that's getting started with Automate. So I think it's pretty quick. I know Roger, you might have some other ideas about this, maybe more specific, but I am not a programmer.

                                            I'm an operations person, and I was able to learn how to create tasks within Automate very quickly. Within a matter of a few hours, I could create one that was very simple. Services and training are offered that will help you to get a little bit deeper into it, but you can get up and running, even without that programming background, I think very quickly. Within a matter of days as opposed to months, certainly.

Roger Hammer:               Yeah, I agree.

Pat Cameron:                   Roger, what do you think you?

Roger Hammer:               When I got started, and built our demonstration, for example, frankly, I was shocked at how quickly I could do something. I have a programming background but from many, many years ago, and I've been in management for the last 20, 25 years. So I kind of understand the concepts of development and software, but just the drag and drop interface and the ability to bring in the different tasks that you need to complete and to build that kind of high level picture of the process that you're going to automate was very, very helpful to me. Then just to basically enter the information into the fields of the pop ups, makes it very easy to get started.

Pat Cameron:                   Yeah, I agree. Automate gives you kind of a template to fill in for every step. I also did a little bit of CL programming back in the day. So if you understand the logic, and I think even more importantly, if you understand this steps. A couple of questions here, but what's a good size project for a company to get started using RPA? Here's my experience in that is, take a look at your prioritization certainly.

                                            Years ago, I took a class from a master gardener and the first thing she told us was start small I think that works really well for Automate as well. Start with a small project, with kind of a limited number of steps. That's going to start to get you familiar with how the product works and how easy it is to create those tasks and then build from there. So I would say something in IT, something in one department that a couple people are doing. So again, I would say start small and build from there.

Roger Hammer:               Agreed. I think small and with kind of high value and give yourself the ability to build some momentum. You need to also typically share the knowledge of a new tool coming in, and the ability to really show that capability and highlight the value early is really important. So, find the simple task and automate it and something is very helpful to yourself or others that can kind of be shown off and say, "Hey, look at this capability that we can now share within the organization."

Pat Cameron:                   Right. Exactly. You want to get some people on your side, you want them to see the value, so that they can carry that forward. [Gus 00:47:21] does that help with your where and how we get started? You can certainly download a trial and start playing with it. We offer that from our website.

                                            So a couple of things. We'll send you a recording of the webinar, and we'll also send you the links to this infographic about 10 Ways to Bring Power and also request a demo. Happy to provide a live demo, take a look at one of your processes and see how we can automate it. Happy to help you with that as well. So thank you for joining us today. We appreciate you taking the time to take a look automation. Thank you, Roger. It's always a pleasure working with you. Hope we'll do this again soon.

Roger Hammer:               Yes, thank you, Pat, and we look forward to hearing from folks and getting started in helping them.

Pat Cameron:                   Thanks, everyone. Have a good day and we'll see you next time.