Transcription
Welcome to today's webinar, which is called "Benchmark Your JD Edwards Processes Against Your Best-in-Class Industry Peers."
Hi there, this is Rupert Fallows from Billasoft. You may have been expecting Yvonne Sims, your advertised host, to be here today. However, she is not feeling well, and we wish her a very speedy recovery.
Just before we begin, I'd like to do some housekeeping items. We have the ability to take questions from you, and if you see the chat box, you'll be able to put them in there, and we will try and get to them at the end of this webinar. If we are unable to do that, within a couple of days, we will send out a response to you. Additionally, we will be sending a link to this webinar and the presentation as well.
With all that being said, let's get on to our topic today. Many companies have struggled with core corporate systems that evolved over a number of years, and within that time frame, processes have altered, lengthened, and potentially become more complex. However, with the current inflationary pressures that most companies are being affected by, there's a huge need to increase employee productivity with faster processes and potentially reduce down the complexity of those processes and be able to take up some of the slack that maybe you're finding that you're not got the amount of employees that you need at this current time. We believe that process mining is a huge area of interest in this particular topic.
With that, I would like to introduce our two speakers today. We have Ben McDermott from Salonist, who is a Senior Consultant Manager, and we also have from Burlesoft, Saga Manvalker, who is a Global Practice Director for Burlesoft. Really, for the first question, Ben, it's over to you. Process mining has been around, um, the marketplace maybe for a number of years, but suddenly it's, it's got really interesting, um, for the JD Edwards community that may not be so familiar with this, uh, term. Could you say, in layman's way, what process mining is or process execution analysis?
Yeah, absolutely. I'm so really pleased to be joining you today to start with, and thanks, thanks very much for having me along. Um, I guess just to try and explain process mining, uh, from kind of zero level understanding cases, anybody out there is not familiar, um, you can think of process mining really as a method of visualizing how a process is actually running within your business today. So we often describe it as an x-ray of your business or a way to create a digital twin of a process or a number of processes. And what we're finding actually is that this visibility alone of how things are actually running within a business goes way beyond the understanding that most businesses currently have. Um, and actually, it really exposes the level of complexity that we just would never be able to reflect in the type of best practice process maps that people might currently rely on. I'm sure we'll see this in the demo later, but, but what we, what we normally find when we mine a process is that there's hundreds, if not thousands, of different variations of how a process is running. Um, we kind of think of these variations as execution gaps, and as I say, these execution gaps really aren't that well understood, uh, today. And while individually they might not be too significant, when we look at them cumulatively, they're actually costing businesses millions of dollars each year, and they're really the silent killers of business performance today.
What are the benefits of process mining? The slowness way, maybe compared to other ways of doing this? What I've described up to now is kind of the fuel process mining capability, but what we're really looking to do at Salonist nowadays is actually move into something known as execution management, whereby it's kind of the natural next step from process mining. So you can create solutions and really direct actions all within the Salonist execution management system. So what that allows us to do is go beyond the visibility we get from pure process mining, act on these insights to really maximize the control that clients have over their processes. So you can kind of think of the Selenus CMS is giving a 360 view of processes, almost acts like a brain for process execution, coordinating all the different moving parts across the system and really driving value quickly. Uh, because it's unintrusive on the existing IT stack, it sits alongside IT and really means that we can kind of orchestrate and work towards, um, improved outcomes on on all digital processes. And, you know, maybe some of your customers, what are you seeing out there? What, what type of processes are, are really put under the spotlight and where are they getting their kind of business benefits at the moment?
So it, it's really applicable to any process where we have a digital footprint. Um, I think it's fair to say that the, the most common processes we point at the moment are the kind of core finance processes, um, HR processes, procurement processes. Um, I'm, I'm kind of regardless of which process we're pointing it at, we can kind of think of the complexity that I've been describing in the three classic buckets. There's, there's people, process, and technology. So if we just drill into those a bit more, you know, businesses are expected to do more with less these days, kind of touching on the point you were making about the inflationary pressures. We've got, um, the process of having to run across often siloed departments and with teams that are struggling to find the resource required. Um, the processes also have to kind of flip to new laws, new regulations, uh, and also different expectations from customers, one and other other stakeholders to be more and more digital. And then, and then finally, these processes are running across a technology stack that's often been thrown together in quite an ad hoc fashion with systems that aren't really designed to work together, having to speak to each other. And, and all of these really create those complexities, these inefficiencies that Salonises is looking to solve. So ultimately, it's all about improving efficiency, um, creating savings, and optimizing performance, essentially.
Thanks, Ben. And Saga, what are you seeing with customers at the moment? What's the main drivers?
Yeah, uh, thanks for, uh, thanks for that report. Um, so yeah, so most of the time, what happens is, uh, most of our clients actually have, uh, as Ben was, uh, alluding to, uh, they have the technology stack which is across a rigid framework. So we have many different kind of ERP softwares put together, and they don't naturally work in tandem. So, for example, you know, in normal customers, we have a typical order flow coming from outside the system through ADI interface into a maybe a Salesforce or a Microsoft Automation, uh, then put into some sort of a trigger logic into an Oracle or a SAP ERP system. So what happens is, this results in a lot of, uh, inefficiencies and and friction, and also execution gaps. Um, how do we approach this in traditional, um, kind of, you know, approaches at clients? What we have done is, uh, we look at the, we, we talk to the subject matter experts, and then we do the interviews, that's pretty typical, and then we actually, uh, try to understand, uh, that that particular knowledge base, and that really is a very partial understanding. So what happens is, whatever recommendations we come to, they're not really taking care of the entire solution or the entire problem as a whole. Versus now, with the execution management system or the Surroundings process mining, execution management system, this approach is very data-driven. So now we have a very specific objective. We look at all the all the data that has been, uh, with the different kind of ERP systems, we extract that, and now we have a very high end-to-end transparency of exactly what's happening in your system. So a lot of times, what I've seen at clients is they think that their typical auto order flow is a very straight line, simple, like, you know, you start from the process, start the process, end, and there are maybe a few steps in between, and how they think that they may have some issues as well, maybe a couple of places there are some inefficiencies, right? Versus when we start seeing this process in the data, data model, when we start doing the extraction, we see that it's a kind of a very, involved spaghetti about, you know, so there are inefficiencies all across places, and, and that's where really the, the, the problems are lying. Versus you cannot really find out by just discussing with them with the team. So, so yeah, so what execution management system has helped us do is really get down to the, uh, to the, to the, to the transparency of the, of what's happening down in the, at the level of the process, at nuts and bolts level, and also it helps you realize the value by closing those execution gaps.
Thank you. You know, this is a very sort of visual solution as well. Maybe you could show us the, the software, and we could get an understanding of this, because, you know, it's a complex area. Uh, I think visualization is, is one of the benefits of this.
This is a depiction of the dashboard of the Sonus analysis that we have that the loss team has has developed. Of course, the back-end system is a JD Edwards Oracle ERP. As you can see, this is an extremely, like, you know, summarized depiction of all the KPIs that we have that we've designed for specifically towards the auto-cash cycle. Of course, the first one is rework. You can actually see how much is your company's rework case rate, how much of your users are actually losing productivity by doing orderly work. Second is on-time in full. This is a KPI to measure, how many, how many of your orders are actually falling, or coming in on time in full. So it again has comparisons to the industry-based standards. Um, and, and the third one, which is the most important one, of course, for a lot of companies nowadays, is the automation, right? How many of your activities are are automated? Um, so as you can see, this, this database is live. It is done through a continuous, what we call as a continuous integration. So whatever changes are happening back in the in the backend ERP system, and it can be multiple systems across geographies, that part has been extracted, and you, you are seeing these analytics on on your screen. Um, to expand a little bit further, um, that's where the my earlier point was, uh, once I, once I actually opened up the variant explorer. So as the second step, I can go a little bit deeper and I can try to show you, uh, how the, how the, how the overall process looks like. As, as you see this, this is how on a very high level, you visualize your process to look like in a very simple straight line, right? But as I start getting into the, the more details as to where the different changes are happening, um, as you can see, there are, once you enter the order, there are many changes been done to it through the quantities, there's some partial shipments done, as you can see, there's some changes in amount, for price, or maybe even the dates. So it is not a very straightforward flow. There are many changes been done that is affecting the overall efficiency of of the order flow. So either keep going and adding more of these variants to the to the screen, it's no more a simple straight line, as you can see, it is, it is spread across across the board. And that's, that's kind of the depiction that I was kind of talking about earlier, where this is where the, in in some form of fashion, in in some degrees, uh, your inefficiency lie. And now you can actually, uh, take one of each and every point and start talking to those, to those particular subject matter experts, specifically about those, those changes. Okay. Um, also at the same time, we can go to something what I call as a benchmarking. Let me bring it back to the, as you can see, most of the screens are very interactive. So benchmarking is, is one area where you can actually compare one type of business, uh, from your, your, your company to the other type of business of companies, and for the same process flow. So what it depicts is, um, in any inefficiency or some sort of a standardization that you want to do in one part of the business versus what's happening in the other part of the business. This can also help you in M&A activities. Let's say you are actually buying or you're actually acquiring some sort of a company, and then you want to compare your processes, your existing processes with that the newer company process. So this actually helps you compare side by side exactly what is happening, what's your throughput, how many orders are passing through, is there any inefficiency in your existing model? Also, that also can be reviewed. Okay. And at the same time, on a bit quickly, I want to show you, uh, the KPI, what we talked about earlier, called process automation. As you can see, this, this actually screen shows you specifically for that process, what's your automation rate, which is pretty low, and then, how is the automation of activity spread across different document types? For example, in back orders, in entering orders, warehousing, packing and parcel shipment. So all these different kind of flows, you can actually determine how much documentation has been, how much automation has been done, rather, in in these areas. At the same time, it also shows you a trend, you know, sometimes it could also happen that your automation could change by seasonality, depending upon the industry, you know, so, so that's on a very high level.
Thanks for that, Saga. Just, just on benchmarking, you know, this is an important topic for JD Edwards users. We, we do a benchmark between, you know, what we have and what we believe is, is, you know, the best processes out there. Do you feel that the benchmark is the way to sort of do KPIs for companies so that they look at those best processes and, and then trying to adopt them? What's your best practice here?
Yes, yes. So benchmarking is something that is extremely valuable and can be really leveraged at a lot of different kind of business scenarios. Mergers and acquisition was M&A was one thing I talked about. You can also, like, you know, a lot of times what happens is you, you start the implementation, which is very typical in European implementations, and you have a thought process in mind, maybe that was four or five years back that very implemented this, right? So you think like this is how you implement the process, but over the period of time, that has changed, you know, so you are not really sure what has happened. So what you do is normally in in most ERP systems and in more com most companies, you have like a model company. So what, what I would do is I would compare my model company or benchmark my model company to the existing implementation done at a particular plan site or or a particular company that you've just implemented. So now you know the exact differences that this is what the thought process was on the left side, and, and which is what he designed for, like to get all the gains and all the, you know, business value out of it, versus this is what has actually happened, you know, so now you can compare where the differences are, and you can go in different levels of, like, you know, detail. Okay. Versus at the same time, you can also benchmark against the industry best practice. So Strongest actually does something called as conformance. So it can actually help you compare what is your KPI with the industry best practice KPI, and that you can actually set up internally, not just comparing data, which is benchmarking, but also conformance. So are your particular KPIs, for example, your automation, your labor productivity, which is your rework rate, or your any other KPI that you have thought about for your specific process area, like order to cash or purchase to pay or any other, like HR, about what Ben was, other into, we can actually compare those KPIs to industry best practice KPIs, and those could be also changed, or rather, we can actually, they are flexible because we can actually go here in the conformance application setup and actually change them based on our industry best practice. So it's not very generic, you can actually customize them, very, very easily. At the same time, you can also, and once you do that, it will tell you like, you know, what, where are the violations? For example, what are the top most violations? For example, in this scenario here, based on the KPI, changing amount per price is the highest amount of violation that has been observed in the data over here, for some reason, let's say. And not in the scenario, this particular violation should not be a violation because, you know, as I was saying, you should be able to make it very flexible, right? So in in my industry or in my company, that particular thing is is a desired activity for some reason, let's say. So I can actually take this out, I can make this and make this desired activity, and then recalc or recalibrate or recalculate all my other, you know, violations accordingly. So, so it's very flexible. You can actually, um, compare with any industry best practice. You can set your own best practices. Salonist actually has defined on their knowledge base, what are the industry best practices. Their teams are very helpful in getting back to you about any information in that area also.
And, and Ben, what, what does a good project look like from your side?
A good project always starts with obviously ingesting the data. So [Music] one of the really powerful things about Salonist, I think, is is how it's a system agnostic tool. So we have nearly 100 pre-built connectors now, but equally, we can build connectors for systems that, um, we don't have a pre-built connector for. And what that really means is we can just get to delivering value for clients really, really quickly within four to six weeks. We can start to show some improvements on a process and have custom-built dashboards created. So, you know, just to, to maybe walk through the, the project plan from there, once we've got the data in, there's a little bit of a build phase where we create some dashboards based on the KPIs, the benchmarks that Saga was just talking to, that are really important for that particular business unit. And then very quickly, we're going to see where the performance is sub-optimal. We can do a root cause analysis on where the issues lie very, very quickly based on an objective and data-driven view of the process, which is something new compared to the traditional methods that Saga was talking about at the top of the session. Uh, and then we can start in this execution management facility to create some automated solutions. Um, and if we think about something like the regular change in price that Saga was looking at just there, in a typical RPA project, we might create a bot to automate the update of the price for us. Whereas in a Salonist engagement, what we can actually do is notice the pattern of changing the price over and over again and go and update it in the source system to get rid of that efficiency entirely at its, at its kind of at the very root of it. And actually back to that regarding the behavior patterns, as you can observe, you can actually set any automated emails also. There are a lot of standard APIs that Salonist has provided. What you can do is, instead of, you know, if there is a, um, not a very normally high interest on people going back and updating tables, we, what we do is, we will give you like an automated emailing script to the, you know, whoever's whoever's watching the pricing, the pricing managers, and they'll get they'll get notified accordingly, saying this particular price of this particular item type or this warehouse and for this particular item has not been the price is not correct because now we know the behavior patterns, right? So, so that kind of root cause analysis, what I think Salonist actually has huge value using the ML engine. They have an ML capability internally, and they also have opened up a lot of the machine language APIs for us to use. So it's not just that it's given one way to work with it. We have actually leveraged those machine language APIs internally within our development team to actually develop these these particular specialized, you know, behavior pattern analysis. And not also on the earlier part, what Ben was adhering to regarding the, uh, the analytics that we are showing, there's this AI engine that's that's run behind the Salonist systems. So once you connect, right, the initial connection phase, and then then we use the transforming the data into a particular data model that we, that we talk about regarding order to cash or purchase to pay cycles or any other particular, you know, data model. Once that is done, as you're visualizing the process and enhancing it, we use what, what Salonist has provided, like an AI engine, you know, so that AI engine plays a vital role to give that value proposition behind the scene.
Saga, we were interested in this topic. We wanted to run a project. Just give us an idea from an SI's perspective of, of how you would engage and what, what the usual items are in the proposal you give, etc.
As initially introduced by Ben, you know, normally we can very quickly, it's a very quick ROI tool. We, we can gain value by doing a quick project around 8 to 12 weeks. That's a typical project cycle that that we have seen for a particular process area, maybe for three KPIs or four KPIs, if you can define them earlier on. But of course, the project goes normally through to four phases, typically to four phases. One is, of course, you do the connection, where you normally talk about the data connection points and what kind of databases you have and what's the backend systems and all. Then you, of course, go to the discovery phase, where which is basically the build phase, but we cannot start talking about how do you visualize the process, what is the root cause analysis, not that. And then of course, we help the client to enhance the process, right? So we put in some automations, we put in some optimizing actions. It could be inside the system or it could be outside the system. You can, I mean, after you discuss the root cause analysis, there could be a problem in the setup itself, or there could be a problem in the behavior pattern of seasonality and whatnot, right? So that enhancement has to be done specific to the particular process. And of course, the fourth and the most important phase is the monitoring phase, right? Which is after all said and done, we have to still confirm that whatever metric you were trying to track has actually improved, right? Whatever benchmark you were trying to do has actually helped you. Whatever task you did for optimizing this particular problem or the inefficiency has actually helped you. So doing that monitoring task also, Salonist has monitoring engines where we will tell you exactly where the value has has been gained, and it will give you a comparative analysis as to what, what would happen if you're not implemented the process mining, you know, engine, ML root cause analysis answers versus what is what is going to happen if you've done it. So they show your curve, you know, where you can literally see a big difference as to what if you're not implemented, you will lost so many millions of dollars. Having said that, of course, while implementing, we are to this is the cloud software, so all this, all this particular extraction, encrypted data goes in the cloud. So it does take licenses. So once somebody is interested, we can get into a little bit of our discussion as to how many process areas and what kind of data size we're looking at. So there is licenses, data sizing for the cloud, and of course, there'll be a very minimal team from our side, as I was saying, we, the quick value we can gain by doing a project within eight to twelve weeks, you know, so we're talking about probably a mid-sized team of three to four people with the data engineers, Salonist expert, and SME, and of course, the project manager, compared to the licensing and other costs that are also minimal, you know.
Thanks for that, Saga. Is anything else, Ben, you'd like to raise?
Just to highlight a point that Saga made, that really it's, it goes beyond a one-time project. Really, the beauty of Salonist is that it can then, following that, move into a really operational tool, so that the teams that benefit from this can maintain that optimized performance over a longer term, and which I think, you know, was the, the monitoring phase that Saga was talking about. And then, you know, once you've done that in one area, moving on to the next processes is the natural next step to it until it kind of has optimized processes across across the back office.
Brilliant. We have run out of time. Thank you for that, and thank you for attending this webinar on benchmarking your JD Edwards processes against best-in-class industry peers. If we did not get to your questions, someone will follow up with you shortly. If you have additional questions, you can reach out to us on the following way. Thank you again to Ben and Saga. We appreciate your time and your insight into this really interesting topic, and for everybody else, have a wonderful day ahead. Thanks a lot. Bye. Thank you. Thanks.