all right well welcome everyone to today's process optimization Academy session focused on in-platform process mining for HR Service delivery and life cycle events um super excited uh for this session for a couple of reasons um first it's a topic that there was high demand for um so that's that's great and I really appreciate those of you who reached out via email and posted on the community and said hey would you mind doing a session a little bit more focused on on HR so thank you for that input and I'm happy to oblige there the second reason that uh um I'm excited for this session is it it does help to highlight the growth of our in-platform process mining solution in two ways one is our expansion continued expansion into other workflows so if you're not familiar process optimization was initially available for itsm and Cs and customer service management but as of the Tokyo release we've now opened it up and expanded it to cover both HR as well as Creator workflow us and the goal long term is to continue to expand and have complete coverage for all workflows on the platform with process optimization the second thing that's sort of cool about this session is it allows us to kind of highlight or kind of it allows us to highlight uh one of Tokyo's enhancements which is multi-dimensional mining where the ability to visualize interconnected processes in a single map so we're going to look at an onboarding example today in which you have a high level life cycle event and then there'll be subtasks that could be aligned to that onboarding case but before we get started and get to all the fun stuff we have to get to the academy house key and the academy housekeeping get that out of the way for the new the new newcomers and those who just happen to stumble upon one of our recordings for the first time and the recordings are are becoming more and more popular the views I'm just I'm very humbled by the amount of people that are interested in the topic and viewing it and I anticipate there's going to be a lot of new viewers um due to the focus on HR this time around so the purpose of these sessions are to help us all get better educated on servicenows in platform process mining capabilities will provide how to's practical guidance and use cases product updates and just as a group exchange ideas and collaborate so we all continue to get better how the sessions operate we'll have a scheduled topic that we'll cover usually for the first 30 to 40 minutes of the session then we'll open it up for questions on the chosen topic for that session then we'll open it up for general questions that go above and beyond the topic sort of an office hours style fashion the sessions are recorded and eventually posted on the community so just be aware of that fact when you go to answer your question you are going to be recorded and finally feel free to post your questions in the Q a section um I'm going to open the mics up towards the end of this but having the questions in the Q a are super helpful if there's times where I can get to them as we're waiting for something to happen on an instance I'll get to those questions in the Q a whenever possible and then for situations where we don't have an answer for your question immediately we'll follow up and post them to the process optimization product Hub on the new and improved Community Forum that's out there I mean maybe we'll use them as a topic for a future session and that's again a reminder if you've got ideas for sessions or things you want us to cover on these calls please post them on the community I'm always looking for new ideas this is the slide that says anything we'd say and do here today can't be held against me in the court of law it's also the slide that says it's part of these conversations if we make any forward-looking statements about what might be coming in future releases you should take them as forward-looking statements and make no decisions based on them because things are always subject to change I'm assuming you've read the fine print now for those that are new just a quick intro to me my name is Dan Grady I've been with servicenow for about six years now primarily focused on our now intelligence Solutions and helping our customers understand how to get more value out of the data that is flowing through the platform so things like performance analytics predictive intelligence the virtual agent and now I've got the the pleasure of helping our customers better understand how they could use our in-platform process mining solution to x-ray their workflows and find Opportunities to improve them what we'll be doing today um we'll do some process mining 101 as always a little bit deeper a little bit more time there today because again I anticipate a newer audience because of the HR Focus we'll talk about a few common inefficiencies that we see um when looking at HR processes but as you'll see they kind of apply across most workflows then we'll do a quick demonstration of how somebody would interrogate or use process optimization to find some insights as part of an HR case process flow and then this is the exciting part we'll create two visualized process Maps one will be just on an HR case flow and the second will be a little bit more advanced when we start to put together a map for a life cycle event so we'll have a life cycle case and we'll link it to two activity sets so we can visualize that we'll get to all your questions we'll talk a little bit about how you can get started once you get all amped up about what you saw here today talk about resources and then future sessions so as always the quick process mining 101 for our newcomers with everything we do there's a design and desired path in our minds for how it should work and how things should play out whether that be planning an event like these Academy sessions uh or a workflow with either an HR case or a life cycle event and when we design these things we design for efficiency as well as completeness to provide the best experience possible for everyone unfortunately what we design isn't always what ends up happening in reality the reality is not all work flowing through a process is going to take the optimal path and that's going to have a negative impact on the experience that both the employees requesting services are having as well as the people that are working at those processes behind the scenes and identifying what is actually happening within our workflows and then improving them isn't always the easiest thing to do for many of our customers the continual improvement process is still a very manual time consuming expense and expensive it it involves pouring through a ton of reports and holding virtual and in-person workshops with a bunch of sticky notes probably maybe it involves bringing in third-party consultants and then ultimately after they go through that they still have to get alignment on the go forward path at the speed that most of you are moving these days even when you get through all of this so much time has passed that the optimal path forward has likely already changed what process optimization or in-platform process mining offers our customers is the opportunity to do better in a few clicks we're going to be able to expose to the relevant stakeholders the reality of their workflows in this case it's going to be HR style workflows and that's going to accelerate their ability to do better for everyone and anyone involved and the beauty of this session today is that we're going to show you those two those those not two it's a few more than two but we're going to show you the few clicks that you need to start getting the level of visibility that's super helpful now some of you might be saying hey we already have a bunch of dashboards that help us answer questions about our HR workflows yeah you do and those dashboards typically help us answer very important questions like how many cases have we worked which type of cases are they how long did it take to close those cases and how did we perform against our targets and usually the most important question is are we improving our cycle times because that has a direct experience impact on our employees experience with our service delivery your dashboards might look like something like this now I was just mentioning to somebody earlier before we started this call I've been in the space for a little while when it comes to analytics and one of the things that I've learned is that information tends to be the most addictive and contagious things out there the first thing that happens when you share a dashboard like this of leadership starts seeing things that they haven't seen before and then they have more questions right why are our promoter scores going in that direction why are our closure times trending that way I mean sometimes the answers to those why questions are not the easiest to get that's where process mining comes in process mining helps us answer a lot of those why questions behind the kpis that we haven't been able to answer before things like where is the process stuck where is unnecessary rework happening where and why are cases going to on hold or awaiting uh employee information right hey if we're Outsourcing some stuff to some vendors which vendors are helping us run faster and which ones are helping us run slower and one of my favorite ones that always everybody always wants to know is who's holding on to the tickets the longest these are all questions that process mining is going to help us answer right you're going to see today that it's going to help us get to the why behind all those HR kpis that we might be focused on from an organizational perspective and we can get to those whys for a single table process like this one here for HR cases or uh we could start asking those questions uh in an interconnect of an interconnected process where you've got a number of different tables linked together on the platform like a life cycle event where you might have some onboarding tasks some tasks that are assigned out to the IT team um well we now have the ability to visualize those in a single process map to help give you that more complete picture whoops going the wrong way in the slides now since this is an HR Focus session and I've mentioned this already that I'm anticipating that this might be the first exposure to process optimization so I I wanted to make sure that I included this part of the conversation if this is your first exposure to what we're doing here if you think about it the process for improving or optimizing a workflow there's really four stages detect analyze improve and monitor I mean we have the ability to help our customers on a single platform move through all four of these stages so if you think about the detect phase that's all about uh identifying that there's something that's not going the way that you expected finding anomalies within your process and there's a number of ways on the platform we help our customers do that whether it be notifications on slas or you've got some performance analytics key performance indicators that have targets and thresholds with notifications tied to them number of different ways on the platform to help you understand that something's going bump in the nights and then once you know that's happening you move to the analyze phase and this is where process optimization really dominates it helps us start asking all these questions about why those things are happening and how to improve them so it gives us our visualized process Maps our bottleneck analysis our root cause analysis our variation analysis and a machine learning driven solution called automation Discovery to help us analyze why things are happening once we understand why things are happening and have an idea on how we want to improve them that's where the rest of the platform comes into play and it's all about improving things so maybe process optimization helps identify an intake challenge that we can improve by adding a virtual agent conversation maybe process optimization helps us identify a multi-hop issue and we can address that using predictive intelligence to help cases find the right assignment group in a much more efficient fashion maybe process optimization shines the light on a manual step that we're taking that we can automate away let me open integration with integration Hub or potentially using a robotic process automation to help us create an automation for repetitive task maybe process optimization helps us identify something that we can just stop doing from a human perspective this happens quite often where we're looking at a process and we say hey why are there two levels of approval on that thing there maybe we just need one level of approval and then we'll remove three days from our closure times or average floater times um so lots of different ways to improve on the platform some automated some human and then the last piece to close the loop is this monitor phase um so once you find an Insight it's important to make sure that we start tracking those insights and make sure that they get followed up on because hey just because we found an Insight doesn't mean we've improved we have to act on that Insight so we have applications like continual Improvement management and automation Center that help us capture the insights and the actions that we want to take to ensure that they get prioritized and followed up on um and then we've got uh things like performance analytics and benchmarks to help us start communicating the impact that those improvements are making all of these things on a single platform really helps us provide our customers um a pretty compelling story if you're thinking about to creating a closed loop continual Improvement experience right that that's the that's the key here is making sure that this is a closed loop repetitive thing right we want to make sure that process optimization becomes part of our routine and just not this one-off process mining science project that we're doing it should be a continual thing because we're always striving to get better the fact that you have all the pieces on the single platform is a big benefit to many of our customers and we think a pretty unique experience all right now one of the things that I'm asked quite often is hey Dan this is kind of our first exposure to process mining process mining is like a new muscle for a lot of uh the the customers that I'm speaking to so I always find it helpful to share um some common use cases or process inefficiencies and non-conformant activities just to get the juices flowing um you know thinking about hey you probably have some of these things you just you haven't been able to quantify them or maybe communicate the impact of so to help you prioritize Improvement projects inside of the organization the first is multi-hop issues or in some situations cases ping-ponging between different groups right reassignments are are very time consuming and in general reduce overall productivity and there may be any number of reasons why a case is getting reinsigned but in order to improve we need to get a handle on who is reassigning cases and and the types of cases that are being reassigned and to whom and once we get a better understanding of this we can take actions to reduce the amount of reassignments that are taking place and improves our speed and productivity so multi-hop issues something that process optimization is really good at exposing or providing insights around the next one is the misuse of the suspended state or I've heard this called some people call this how do we catch people that are gaming the system right the suspended State exists for a reason but sometimes it's misused and it's just some people are just using it to stop the SLA clock and obviously this has a negative impact on throughput times and ESAT scores so getting better visibility into how often uh the suspended state is used by who when and how long it's used is often useful information for organizations now sometimes cases require us to go back to the employee from some for some additional information and this might be part of the process but other times uh these situations highlight an opportunity to improve an intake experience and gather more relevant or necessary information up front to speed the closure process right um so you've got different parts of the process sometimes we we need to get employee accept part of the process that requires us to get employee acceptance before we can close the case I mean and that's just a natural part of the process but sometimes when we reach out to employees to get that acceptance the case then comes back to work in progress so there's something that wasn't right potentially we could gather more information up front to ensure that we we don't get it re-assigned back to work in progress or potentially we need to do a little bit of training now or create a knowledge article so anytime something goes out to an employee and then potentially comes back is an opportunity to improve and that's where this one comes in as well looking at solutions that have been rejected by by the employee anytime a solution is rejected by an employee there's a productivity impact both for the requester and the filler account it also has a negative impact on ESET scores again so identifying categories of cases or groups involved helps give us Direction in terms of training opportunities or maybe knowledge article quality improvement opportunities that are out there this one is always interesting to me because when you start digging in and asking questions questions about this scenario it varies the answers that you get back but cases that go directly to closed could be a potential self-service targets I had a customer the other day describe the scenario where a call may come in and the agent answers the question over the phone by sending the employee a KB article and then they just move that case directly to closed um and I love the way that they described it to me because they said we think of that scenario as having a human search engine because we already had the content the person just chose not to find it so getting a better understanding of these types of scenarios can can help them identify future knowledge articles or virtual aging conversations or maybe areas where can they can just Market the self-service options they have a little bit better I mean the last common use case is this Channel or Source performance so just getting a better understanding of the volume of cases coming in Via each Channel and the impact each channel has on velocity is is something that a lot of customers are interested to are there certain types of cases coming in Via certain channels like go to on hold or awaiting employee additional information from an employee more often than others in those scenarios can we improve that intake experience um and if we did what would be the productivity impact so that's another popular one all of those types of analysis or use cases they're they're pretty straightforward to to do with process optimization in fact some of them are things that we package up with our HR SD content pack for process optimization so what you'll find is with most workflows that we have out of the box we provide content packs that contain we call them finding definitions that help guide our customers to certain common inefficiencies and non-conforming activities some of the ones we just ran through there are part of that content pack to jumpstart you now last thing before we get into the instance uh we're obviously using process optimization here at servicenow and when we applied it to our own hrsd workflow uh we identified that we had 74 of our cases were needing to be re-categorized re-categorized sorry um and that had a downstream impact of 60 of our cases taking longer than six days to get to closure and as we started to dig in and look at areas where we can improve the intake experience to reduce these things reduce the times that things went to General or other like that's one of the areas that we focused on there was a lot of stuff that was going into general or other you know we were just being lazy employees so we looked at ways in which we can improve that intake experience and then we also started looking at ways in which we could apply predictive intelligence to head automate uh some of the categorization we calculated that if we acted upon these things there was opportunities there that we could that could make the process 46 faster and again uh it comes back to acting on these things right we're able to identify these things we're able to quantify these inefficiencies and non-conformant activities now with process optimization and then it's a matter of us taking those insights prioritizing them and acting on them with the rest of the servicenow platform okay let's get into the instance and start looking at process optimization in action now I'm going to start my journey here on a performance analytics dashboard just to be clear you don't need to be using performance analytics to use process optimization so we think it helps you create a more complete experience if you are using the two in combination yes but it is not a requirement to be using performance analytics to use process optimization but in many cases your journey might start on a dashboard like this or somewhere else where you're looking at some key performance indicators around HR case management like ESET scores and closure times and you might see that your closure times are trending upwards which is going to trigger the question of well why is that happening and right from performance analytics you can drill down and you can get to process optimization and the process optimization workspace again don't have to start this way but you could I mean what's going to it's going to take us to a landing page for our the mining we've done for our HR cases and in this case here we mined about 2500 HR cases those HR cases have taken uh 234 routes to get to closure and their average duration is six days in this case the first thing that you're going to see on this summary and insights page is these insights cards and these are things that come as part of those content packs we provide you also have the ability to create your own insights cards and we have some other recorded Academy sessions on creating finding definitions to generate these cards if you're interested in learning how to do that but it's going to highlight things like hey those solutions that are getting rejected after proposed closure and we're going to come back and dig into this because wow we've got 72 percent of our cases of these two 2500 that we mind that met that criteria or those cases that are slipping through the cracks and going directly to close from ready and then being reopened or those cases that are misusing the suspended state so these insights cards again highlight some areas of opportunity and we can drill down and narrow and just focus on these cases right from the beginning on this summary and insights page we also have our bottleneck analysis which is going to show us things like uh the which hops are happening most often or hey which hops in the process are taking the longest to complete on average we can start focusing on the the bottlenecks within a process we also have our variation analysis so whereas bottleneck analysis shows us the individual hops in a process variation analysis looks at the entire route that a case may take to get to closure so we can start looking at here's our 234 routes maybe you want to look at the routes that was the route that was taken most often by cases or maybe we want to look at the ones that have the most steps and this is going to look a little bit scary but I'm sure you have a few cases that have went this route as well and back at home or maybe you want to look at the the routes that are taking the longest on average to see if there's an opportunity and on any one of these we can drill in and just view the process map for the cases that took those paths for our analysis here today we're going to start with these solutions that were rejected right we want to get into our why our closure times are trending upwards anytime something gets rejected that's going to add to our closure time so we can simply drill into that and it's going to get us to the visualized process map for just those records that were rejected um by the employee we want to get some basic statistics about this map we can open up our model options here we can see that these 1800 cases that were rejected they took 203 different routes to get the closure their average duration is three weeks and their standard deviation is two months there's a lot of variability there now some of you might be saying hey Dan um that doesn't look like 203 routes in that map and that's because by default we're just showing you the top 20 in terms of volume but you could expand that out if you want to look at all the different routes that things are taking we could expand it out and look at those as well I'm just going to focus in on our 20 because that's going to be the easiest place to start from an analysis perspective now if I start looking at the map itself to see if there's any opportunity one of the things that jumps out at me on this map is this here and not just because it says 666 but I'm looking at these and I have a nice chunk of tickets that are going directly from ready to closed complete and that and then remember we started at these are being rejected and reopened back so going back to work in progress so what's going on there like why are we just going directly close complete and then it's being reopened seems like an opportunity or something from a quality perspective that we can improve um and there's five years of total duration packaged up into these 666 so that's five years of productivity that we can gain back if we can find some ways to improve or some areas of improvement opportunity here so what I'm going to do is I'm going to use this transitions filter to just isolate the things that are going from ready to close complete and then back to work in progress so this ability to start building out very specific flows or transitions inside of your process and analyzing those something that you can do with process optimization and comes in very handy and if now I've isolated it down to just those things that are taking that path and one thing that I'm noticing here is from ready to close complete there's three days built into that then when something gets closed and then reopened it's taking another three days for it to get reopened right to move back to that work in progress stage and then it's taking us six days to actually do the work and get it to to closure so there's six days in these two steps that are just getting wasted right because the the actual work is only taking us six days to complete so opportunity to improve there now if I want to start analyzing this further I can start using our breakdowns on the left hand side of the screen that's going to give me insight into different areas of these cases that are taking this path so I can start breaking my data or my process down by let's say service or assignment group or priority any structured data you have you can use so you can configure your own breakdowns these are just the ones that we we set up by default um so if I want to look at it by service I can see that the majority of these cases that are taking this path are in that employment verification area so I might want to focus in on just those and narrow it down and then if I want to look at which assignment group is involved most often I see that's my Global people care e area so let's focus in on those as well so done some analysis now I found a potential area that we want to dig in further it's our employment verification cases that are being handled by global people care e what's going on with that process I want to ask a few questions of some folks so I can come over here now and I can add a note that says hey please follow up with Global care e on how they are handling employment verification and I can add a snapshot I could add somebody in this note to make sure I can um they notify somebody I could share this project with somebody to kind of give them visibility into the insects are here so a number of different ways to kind of act upon this or ensure that it gets followed up on them we'll look at another in a second that's just one way to analyze the data I'm going to show you two others just so you can get a feel of the interface and how people would navigate it so what I'm going to do is I'm going to clear all my filters here and we're going to start again and I'll close this right the next thing I want to do is let's use that bottleneck analysis so you saw on the summary at insights page we had bottleneck analysis to help us launch into a map I have got that bottleneck analysis capability right here within the analyst workbench as well we've looked at some of these Bose completes back in work in progress already but the next thing that jumps out at me as this is this ready to work in progress and what I can see here is that it's taking us three days to just pick up the cases I mean it looks like I've got 11 years of productivity bottled up in the time in which we're is taking us just to pick up a case so let me just focus in on those ready to work in progress cases and if I start using my breakdowns again to analyze these cases I can see well while Global Talent uh is the one that's got the most cases in this bucket if I use the average duration it looks like this Global people care team um they've got something's going on there why is it taking one month and four days for them to to close a case so let's just focus in on that Global people care team isolate those guys and now what I can do is I can start comparing this team maybe I want to compare it to those other Global people care groups to see where in the process there might be opportunity or where what's where in the process they're taking longer than others so let's come in here and I said compare we'll say I want to apply filters and just apply it to let's say this Global care people e and a team and look at those guys now Alex oh I can do some comparison statistics here where I start comparing Model A to model B very useful we want to comparing teams different types of cases or different vendors potential side by side and I can also then just start playing with the map a little bit for let's make model A a little bit larger here and expand out model B a little bit just so we can start looking at the different steps and if I start walking through this here I can see for these guys over here it's taking them two days to move a case to to work in progress whereas the overall Global care team here is taking them one week so there's an opportunity and then if I start moving down the map more these cases going to a waiting acceptance it's taking two days to go here if I look at this map um for this team it's taking six days for things to go from for work and progress to awaiting acceptance and then another big area here is this two weeks so once they actually do complete the case it's taking two weeks to be reopened so this is not only their quality issue there's something going on in here in this reopen time uh whereas these guys over here are only taking four days so lots of opportunity inside of here so again I could come in here and say hey follow up with Global care team do they need help and I'll attach a snapshot and post that last use case inside of here that we'll just touch on real quickly so I'm going to come in here I'm going to close the comparison I'm going to clear all my filters and then just look at this uh priority piece so we'll clear all the filters again and I've got some priority breakdowns and what I notice here is I've got some p1s actually 30 of my cases have been classified as P1 but it's taking me 10 days to move through these p1s that's not a good thing I'm sure that's not a good thing so let's just focus in on those those P ones and if I come over here to the service it looks like my referrals right one month and one week for referrals and these are p1s what's going on with those and I can I can focus in on these referrals hit apply and start looking at the process itself maybe I want to expand the map and see where in the process there might be some opportunities um and if I start looking at this the fact that for P1 it's taking five days for these to get to work in progress is a problem all right and just so I can see I've got my histogram here that helps me kind of show the the distribution so it's not like one or two things that are skewing this um it's taking me seven days for things that are get reopened and again just making sure that most of the stuff is Falling Towards the front and it's not one or two cases that are out there skewing things just some opportunities to it to improve here especially these ones that are going to awaiting acceptance also an opportunity now if I bring this back and say let me just that's referral so there's an opportunity referrals but where else is there opportunity with these p1s well employee verification has the highest volume let's look at those and if I expand this out a little bit what we'll start to see is that I've got these four days built into this awaiting um acceptance Loop that's four days that we should be able to claim back I think maybe we need to put an SLA on that like if if this is a P1 we should get a response within the first day that will speed things up so what I can do right from here is you've seen me add a bunch of notes to ensure things get followed up on or let people know about something but I also have a linkage with automation Center and continual Improvement so I can come in here and I can create a continual Improvement initiative that ensures that people follow up on this employee verification P1 thing that's going on so like we'll say follow up or improve P1 uh experience for employee verification process and this will create a continual Improvement management record as you've seen in other class Academy sessions and then we can take these Improvement initiatives prioritize them share them with other folks attach projects to more tasks to them to make sure that they get acted upon so just three different examples of types of analysis you might want to do I wanted to kind of give you a sense or appeal for how you might navigate the interface and identify some improvement opportunities real quick before we get to the authoring piece just so we can look at let's say so we looked at a single process map let's open up an onboarding example and again real simple straightforward example well we kept it a little bit more simple one of the things I will say these interconnected Maps lots of power but just like with Spider-Man with great power comes great responsibility right we just because we can create these interconnected Maps doesn't mean it's always the best way to start answering the questions you have about your processes very often you might see something in let's say performance analytics or on a dashboard somewhere that tells you that you might have one part of the process that's a problem area if you already have identified the part of the process that's a problem area it might be easier to do your analysis with a single map instead of these interconnected ones but if you're just starting out and you want to dive into this you can you can do this as well so you've got in this case we've got onboarding the main parent case life cycle event cases over here on the left hand side I'm going to zoom in a little bit so you can just see and what we've done here is it walks you through the different activity sets that are part of this so I can see we've got our pre-boarding tasks then it's taking us four weeks to go from our pre-boarding tasks to our day one prep tasks and then one week to go from our day one prep tasks to our day one tasks and if I want to drill into any one of these areas let's say I've got my pre-boarding test I see that's taking me four weeks I might want to just jump into those tasks and start analyzing those so I can come over here I've got my pre-boarding tests I can expand this out and now I can focus in on those pre-boarding tasks use my slider to open those up and start analyzing that part of the process so these interconnected Maps again give you a nice holistic picture about what's going on inside of a process but again you may say to yourself this might be a little bit too much at first I want to just focus in on a specific task for a specific part of the process to do my analysis really up to you but it's nice to know that you have the flexibility to create these types of maps if you want and we're going to walk through the process of creating one in a couple minutes all right now most of you are sitting there saying Dan this is awesome like how hard is this to do and that's what we're going to do next we're going to create our first process visualize process map for HR case so to do that we're going to come here and just so you can see I'm on this screen if you ever wanted to get to the screen you'll say process optimization in your left hand nav and then you can get to the analyst workbench right from here that'll take you to this screen and we'll say create new project we're going to call this one Academy HR case and we'll say save and then we're going to be presented with the screen that allows us to configure the data that we want to mine um and down here we've got a related list for our table configurations we'll hit new and set up our table and the table that we're going to want to use in this case is going to be our hrk so we'll give this the name HR cases which table do we want to use well we want to go find our HR case table I always like to hit preview here to see the volume of records that I'm working with in this case I've got 15 000 demo cases inside of this instance now best practice dictates that we should focus on cases that are closed right from an analysis perspective we really want to be analyzing work that has completed the process remember we're doing process analysis here let's make sure we're comparing Apples to Apples and looking at things that have went through the cycle of closing so we'll say active is false and then the next question that typically comes up is hey like what is the right volume of data how much data should I be using or how much data do I need to use process optimization the answer to that question is what is a relevant data set for the analysis that you want to do and the questions that you want to answer you don't need to analyze two years worth of data to analyze how effective your process is in fact your process has probably changed a bunch over in those two years and it's really most cases probably best just to focus on the last month worth of cases or maybe the last quarter worth of cases maybe you want to focus in on cases for a certain category or service you just want to narrow the data down to what is a relevant set of data for you you could mine five cases if you wanted the mine five cases don't really need to have a large volume to do this so we're going to come in here I want to just so you would say if you were doing that let's just say things that were closed on the last quarter right if you wanted to do that my demo data a little bit older so what I'm going to do is I'm just going to use the criteria here to to isolate Just My Demo data so I'm going to say created by is po underscore HR underscore demo user and that's going to narrow me down to my cases that I know have the relevant audit data to do this and we'll say save and then once I save that I've only got one more step that I have to take and that's to create an activity definition activity definitions what those are those are the steps in the process that you want to analyze very often we're focused on the state transitions within a process so I can come in here and I can say hey you know what I want to look at how things are moving from state to state within the process um sometimes you're going to add assignment group maybe you want to look at how things are bouncing back and forth behind between the assignment groups you can have more than one activity for a mining project I always like to start with just state to get a feel for the data that I'm working with and the analysis that I want to do but you could have state and assignment group in the same map you want to start looking at when maybe a knowledge article was attached to if that's changed on the record because we're capturing that in the audit log you can add that as an activity definition you have to have one activity definition to run the project but you can have more than one if you want to and then the last thing that we're going to do here is we're going to add some breakdown definitions we'll just add one for purposes of this exercise but if you remember as we were doing our analysis we had HR service we had assignment group we had priority you can add up to 10 breakdowns per model in this case we'll just add our HR service here we'll say submit and now we're done I can go to my project definition and I can hit the mine button and what this is going to do now is it's going to harvest the relevant data based on the criteria that we just provided from the audit log tables it's going to take that data it's going to pass it off to the machine learning infrastructure so the shared machine learning infrastructure that things like predictive intelligence and the virtual agent uses we're doing all the heavy duty number crunching on that shared infrastructure so those process optimization should have no impact on performance of your instance because all the heavy duty stuff is happening in that shared infrastructure once the mining is done it's going to spit that finished model back into your instance and you will have your visualized process map now this is the part of the exercise where since we're using that shared machine learning infrastructure you're basically in step two here we're getting in line we're getting in a queue of others you know with others to say hey when is it my turn to get my mining done so this could take anywhere from 30 seconds it could take anywhere from five minutes or 10 minutes went pretty quick here today which is good we were first in line and once it does this the number crunching happens it's going to spit the model back in I'm just going to check here in the Q a while this we're waiting for this don't see any questions in the Q a as of now so that's good and this completed so we'll we'll Now open up our analyst workbench and we'll have our visualized process map it's just that straightforward right it's that easy to set up um it's one of the things that we believe give our customers a huge Advantage from when they start thinking about getting involved with process Mining and using it to find inefficiencies and non-conforming activities in that process most of our customers if you know how to build a report on the platform you can generate a visualized process map in a matter of minutes right you could go turn on the plugins right now in your subcloud instance do what we just did and start getting some insights into your uh your workflows and where there's some potential Improvement opportunities all right that was straightforward HR case awesome now we're going to take it up a notch we're going to do a life cycle event um and again this is a little bit more advanced so my we'll we'll go it's a lot slowly through but I want you to see the process it's not overly difficult it's just got a few additional bells and whistles so we'll call this Academy life cycle event and we'll hit save now in this case we're going to have multiple tables that we're linking together essentially we're going to be linking together the HR lifecycle events case table and some HR tasks so we'll hit new and we'll call this our life cycle event case table and the table that we want is our HR lifecycle event case and then the criteria that we're going to use again I've got my created by which gets me to My Demo data oops created by I don't really have the best demo data in this scenario for life cycle event yet so we'll say created by his Pia underscore life cycle events demo user I'll hit preview and then I'm going to add this condition of parent is empty should be good great now we're down to the data that I know I have for my life cycle of it again you'll probably have time frames and and kind of closed in your cases I just back to what is the relevant data for the analysis that I want to do from a demo perspective that's the relevant data that I have in this case and then we're going to add our activity definition have to have an activity definition Again State I highly recommend that you use State when you're starting out and then we'll add a breakdown just so we have one and in this case we'll make it let's say assignment group so we've got our parent table set up at this point now what we want to do is we want to link these lifecycle event cases to the tasks at each of the different activity set levels uh throughout the process so I'm going to go back to my project definition and I'm going to come in here and I'm going to add a second table and it's going to be my task table so we're going to call these uh pre-boarding tasks we want to link to this pre-boarding and those tasks live in the HR task table and I'm going to set this up to be I need to get them ones that are underneath activity and in the activity set called pre-boarding so we'll say the title is pre-boarding and just make sure that I got that right right and then if I add in My Demo data filter again created by is po underscore Le underscore demo underscore user I got that right perfect we'll narrow it down to that set of data so again isolated the data or the tasks that we want to to analyze here and then since we want to link this to our uh life cycle event case we have to choose the relationship in this case it's a child because we can have multiple uh HR tasks linked to that case uh typically I look at the say hey if these things could be living in a related list on a record for that parent I'm going to use child if it's something that I would link via let's say a magnifying glass to that parent case I'm probably going to be using the reference relationship that's just how I think about it now which table do I want to link this to I'm going to link it to that life cycle event case table that we created in the first step of this and then what field is going to bind these things together well I'm going to use the parent field um that's there so we've configured this piece and we'll say save actually we should have just said submit so we've linked our preboarding oh I need to give an activity definition that's right always have to have an activity definition so we're going to say State again here say state submit we got one table and then what we're going to do is we'll go back to our project definition and we'll add our second table I'm going to move through this I'm just being mindful of time I'm going to move through this fairly quickly same exact process as we did the last time these are day one tasks we'll come in here we'll say HR task we want to get our Activity Set so come here to activity activity Activity Set title is going to be day one let's hit preview to make sure I got that right yep I'll add my qualifier condition created by you know and then we'll set up a relationship again it's going to be a child which table are we going to be child of you'll notice I have both of these tables if you think about something that's a little bit more hierarchical where I would say like a request we have request requested item and then tasks uh you would you know you could Traverse that tree here if you were looking at doing a request a multi-dimensional model or interconnected process and again we're going to use parent say okay we'll save this we'll add our activity and then we'll hit the mine button so if you have questions Now's the Time to put those in the Q a uh just because we're going to hit that mind button again we're going to get in line we'll say State and I'll have a few minutes to answer some questions hopefully as we wait for the mining to come in we'll say submit great and we'll go to our project definition by the way this is an important piece if you started to fall asleep now as we dial back in look at this check box here so for life cycle events right each of these individual tables here pre-boarding and day one or activity sets I should say um we're going to want those to show up as the transitions in our parent table on the map so to make sure that happens we're going to check this box it says use child table configurations as the activities in the parent model so we'll we'll check that box we'll hit save um and then we'll generate the the full model here all right that's going to run I'm going to check for questions I got a question here so a question comes in if you were doing this for for itsm on incident trans or problem record where multiple tasks may be involved to certain groups I assume you could reference a specific CI in the C and D be in a similar fashion uh Phil I'm going to take you off mute just so I can understand that question a little bit better to make sure I know Phil you always love to have the CI in your question go ahead Phil I think I just took you off mute there just ask if you could ask out loud so I don't know if Phil Phil's not hit this again you there Phil ah there we go perfect didn't have the mute button there yeah um welcome back thank you sure sorry I missed the last couple of weeks but it's been busy um yeah I was just watching to you know you're lining up from a case standpoint and you know me I'm always putting on my cmdb side perspective as well because I'm looking at configuration management as a process yeah and um so as you were building that out you know I think to myself okay well if it was an incident or a problem you know situation but I wanted to hone in on a specific group of CIS for example yeah maybe there's some problem Windows machines or something and I want to find out you know the process minding of saying why are all the tickets you know going to this particular group that are part of a particular group of CI is always you know being a problem right and that's when I was watching you walk through signing the child reference versus the relational my question was about when you're specifying the reference would that be the place where you would reference the Ci's perspective yeah a couple of different ways to do that a approach that use case one would be maybe if you've already identified the problem area maybe you just create a model for just those you know use the filter criteria when you're building your model to mine the data up front so you're you're already isolating those incidents that are tied to that CI class potentially another way would potentially to use breakdowns uh breakdown on CI class um for for if you're mining the incidents those will be the two ways that I think I would start initially analyzing it um but we could always look at other other ways to do it as well but those those are really where I would start so my follow-up question is would it also support idynamics seem to be group I don't know that off the top of my head we have to dig in onto that one okay because that would be based off of a cdb query yeah I'd have to dig in I don't know I have not tried that yet okay thank you um and as we were talking about the cmdb with Phil look at this we've now got our our um visualized interconnected process map um so pretty straightforward to set up didn't take us too long uh to get to get this going again ask yourself the types of questions that you're trying to answer and make the right choice do I need to have an interconnected process map to kind of answer these questions maybe I start with an interconnected process map I see a subset of tasks from there that are a problem area and I create a follow-up a single table map on just those tasks to give me a little bit more flexibility from a breakdown perspective moving forward as I do my analysis um so that was the kind of quick demo setup we did our creating our first HRS team we did two visualized process maps in less than 15 or 20 minutes one of them quite Advanced so let's open it up uh any any additional questions comments thoughts uh that you you have out here um just looking in the questions here see what we got a good question so so um John's asking like hey who gets involved I'm gonna paraphrase John's question like who gets involved in these types of analysis like what are the typical folks that you're seeing um and from organization to organization it varies um in terms of who's using process optimization and very often it's not just one person uh right you might have somebody who's comfortable turning the knobs like we just did there to kind of create the map but then they might not be the same person as the person that's good at interpreting the map um that knows the process really well so you can start asking questions like hey in the map I'm seeing it's taking six days to go from this state to this state or I'm seeing a whole bunch of these things in this category that are taking five weeks compared to everything else the person who's configuring the model uh might not be the same person that understands the process well enough to say whether that's normal or not or that's out of the ordinary um so really it tends to be a combination of folks when we start working on using process optimization somebody who understands the data a little bit somebody who understands the process a little bit and somebody that can do what we just did sometimes in an organization that's the same person sometimes it's more than one person uh doing those things so it's a mix um and then in terms of kind of how this is used that's something that I'm asking every customer we get involved in right because again we want this to be not a science project but a regular routine thing that we're doing because we want to continually improve our workflows so one of the questions that I'm constantly asking customers is what is your continual Improvement or process Improvement Cadence do you have weekly stand-ups where you read the process are they buy with you are they monthly and I'm getting a range on customers sometimes it's quarterly sometimes it's monthly sometimes it's bi-weekly um in which this is involved as part of that kind of stand-up or that process review session um and there's some customers who are just starting on that journey and they're asking what is the optimal um check-in point and we think anywhere from two weeks to to a month or every month is a good time frame these projects that we just generated you can schedule them so it's nice to see side by side month over month and do two different visualized process Maps or are fun to analyze at least fun for me to analyze to see how the process has changed um so you know it's it's uh it's everybody everyone's a little bit different but those are the trends that we're starting to see John I hope that helped a little bit anyone else have any any questions they want to put in the chat um as you think about your questions I'll just move on a little bit in the slide deck here just to get this after so if you're one of those folks that this is your first time you're saying well this was awesome how do I get started what do I do next uh one is uh on the now learning site there's a process optimization Essentials course it's on demand self-paced whatever you want to call it takes about 60 minutes to go through and it just walks you through a lot of the the core components and kind of how to get going with process optimization the other thing is that you want to activate these plugins in your sub product instance so you can go you can activate the process optimization plugin and then if you're on the HR side of the house there is these content packs that get you those in insights cards that we saw you're going to want to activate that HR SD process optimization content pack that will get you the content for both hrsd as well as the life cycle event so activate those two plugins in your sub prod instance and Away you go you have the ability to mine up to I think 4 000 cases without being licensed in your subprod instance so that's that's what you do next there's tons of other content out there for you so I mentioned the on-demand training there's a great now on now white paper uh talking about how we what we've done here at servicenow around process optimization and it looks at five stories of different processes or five examples of processes that we've mined already here at servicenow I think we're up to about 25 or 30 that we're Mining and then presenting those results as part of our CIO and CEO dashboards as a great Forbes article written by our SVP of analytics VJ kotu who's a huge process mining believer so he's got an article out there around the transformative power process mining on employee experience you probably found this session via the community there was a community Forum which is now called a product Hub specifically for process optimization so if you have additional questions please go there and post them if you have ideas for future sessions please go there and post them it's a great way to start engaging more regularly than these two-week sessions so I'd love to kind of interact there if you've got the opportunity to do that or desire to do that there were some knowledge sessions that are recorded and posted there on that product Hub and then a very similar session to what we did today a little bit more itsm Focus but there's a live on now session that talks about getting started with servicenow in platform process mining that's out there again very similar to this session but if you're you're interested that exists as well last chance for romance Last Chance anybody got any last minute questions that they want to get out there and get answered before we wrap it up for the day all right with that appreciate everyone's time just so you know what's coming up next uh next session hopefully on November 18th unless we get a significant uh influx of demand like we did for this one we'll be focused on process optimization in the request process then as you've seen these interconnected Maps create a lot of give you a lot of insight and visibility but interpreting them is not something that most of us have been doing our entire lives we want to spend time focusing on how do we interpret these interconnected process maps and then on December 16th right now the target is to look a little bit more about how do we start using process optimization around the yes work that has slas tied to it a lot of other great Academy sessions out there if you like what you saw here today there's Academy sessions for things like the virtual agent the platform itself analytics on the platform and and the mobile Academy please check those out thank you all look forward to talking to you in two weeks so we're communicating on the community forum