welcome everybody to the Digital Services Forum call we have a couple topics today they're they're different topics they're not really related so what we're going to do is we're gonna break it up into 30 minute chunks and talk about these two different areas and um I'm going to talk about some of the stuff in the beginning that I'm working on for no code low code or what some of you might be referring to as citizen development and then we're going to have um we're going to have a second speaker Dany come in and he's going to speak to us on the second half so if today's your first time we always like to keep a really active chat in the web meetings so if you go ahead and just go down to your chat and introduce yourself to the group we have a a really active chat that we always go back and look at afterwards to make sure that we're getting the right topics and we're asking answering any lingering questions that we may have because some people don't really want to chime in on the group so definitely go ahead and introduce yourself where you're from and um if you're here for your first time what you're here for that' be great today just to give you a quick look at the agenda I'm I'm G to be the speaker for the first half and then Dan Grady for the second half and I'm going to go right into a no code a low code no code workflow that I'm working on with a few of my customers today and for those of you that don't know I'm a senior executive Enterprise archit arit and I I work in state local government so the examples that I usually use are with state and local government or higher education sometimes without further Ado I'm going to jump right into the topic so we can get rolling because it is 30 minutes I'm G to pack a lot into the 30 minutes the part that we want to look at is not really no code low code right but the reason we're doing this is to be able to get Services out quicker and be able to expend expand our our development capacity so if we look at deploying a digital service and again I I come from the side of this where I'm looking at in the context of a government so a government for constituent Services I'm going to use that as an example so a state government might have anywhere between 800 and 1,400 services that they offer to their citizens okay so in the beginning you have to conceptualize it design and create it launch the service and then make sure you're assessing it and improving it as you go so that's the value stream that we want to get it's really relevant to the product cycle right we want to keep going and improving on that product as we move through now as we go in the initial the way we we usually do it today is we request that service right so a lot of what I've been looking at here is how is this done today where are we getting stuck is is it the the demands or the ideas coming in where are we getting stuck in this area and then on the design side we come down and this is when development has accepted that request and then we come in and we start working on it so we start asking a lot of questions around you know how long it takes to get to this point till you get a developer to pick it up and start looking at what you need and suggesting technology then we go ahead and launch it right and when we launch it we basically have to get it through that release cycle and test cycle and then into the MV into the um you know the production environment and then we're kind of starting phase two of it ideally right where we're looking at what do we do next with this how do we improve on it so the part I want to talk about today is it takes a long time and a lot of writing and a lot of different records to get here to get to a point where we have a prototype of the service and some of the low code no code what I want to do is look at a suggestion for getting there faster right so before we go in I want to show you the use case that I've been working on and then kind of talk about where we are with it a little bit so we're looking at instead of this like I've seen this take six or eight months sometimes uh depending on the sides of the service it might be warranted other times it's just a little silly thing that they people need with a form in a workflow in order to get there and um it's just a lot of paperwork right so to get to this phase I can't tell you how long it is in your area or you know it's different where we go but to get to that point is really tough and I think this is a good point where where we can bring in citizen development or low code no code so the idea here is to work towards a 15minute prototype and that's what I've been working on a little bit and I'm going to demo this to you in the environment where I am with it right now but basically what it is is coming in and being able to request a new service and I'm going to use the citizen example right so there's a an agency coming in in a state government and they they need a new service for their agency so they're going to they're going to be able to ask for it but I'm going to ask for a little bit of different things here because of where I'm heading so I'll show you that in our environment so if you want to go into the forum instance right I can get you um you can go and look at this stuff that that I have today so I'm I'm caught up in this area here where after that request comes in I want to do two things I want to make sure that I come in and I say this is a new service request or like in government you might have a license permit or an information request and then kicks off a workflow like any request does and that workflow I want it to build the actual application uh for that that new request that they're asking for right so basically what it would do is a kickoff here and it would run for those of you haven't seen it I'm going to explain a little bit about what an application te is uh but right now I can't get the flow I can't have one workflow that would go down and actually build an application for me that I would be able to hand off to a citizen developer okay the idea here is after this workflow kicks off and it it builds it builds the the application template so it gives that citizen developer a place to work in an environment that's completely confined so we're not just giving them open access to a Dev environment we're giving them access specifically to an application that they're able to work on with guard rails around that application so it's basically scoped app right for those of you that've done development before so after that flow kicks off what'll happen is I'm in this I'm asking who the developer is right who is that citizen developer that wants to work on this new thing and then what we're doing is we're going in we're sending out a welcome to the team notice to that person that requested that they they' be able to develop their own constituent uh service and after that happens they'll get access go to your app they'll get access to an environment where they can start to develop things like their record producers or or their workflows right they get access to the app engine Studio which is meant for these low code developers or or citizen developers okay but the the thing that scares Dev on this or it scares this service now administrators right is just giving them access to a Dev environment and that's what we want to we want to avoid here we want to put a little bit more guard rails than that so in this in this DEQ permit test one this is actually an application that I built with this workflow so that I can really confine that developer but still give him the freedom to see what's going on now because the template when they deployed the template and I know that a license and permit needs a record producer what you could actually do is you can let them see what that looks like in prod so you can say look here's the start of your licensing and permit request which is this right here you can see it in test uh but then you can come in here and edit it you can edit it in app engine Studio but again they're they're all they're only editing within that scope and they're only editing in a development environment so the idea here is that we don't even bother development or Architects or anybody at this point we just give them access right away to start building their prototype even if it's just the form that's fine right if it's just the form for now and then later you want to train them how to do some basic workflows uh but this is how we enable some of those other people to come in and do this with a controlled scope okay so some of the benefits that we've been looking at right if we can get to this point uh departments they get a lot of autonomy like a lot of people get frustrated waiting frustrated filling out lengthy lengthy forms on hey let's build user stories and do all these things so a lot of times you're able to get this to them right away and what Gartner says on on citizen development is that we can increase our Dev capacity up to 65% and that's just a number I pulled off the Gartner site so I don't know how true that is or if anybody's actually realized that benefit but that's what we're really after right a big Dev capacity Improvement the other thing is we avoid frustrations we're giving them immediate access access to Common prototypes so there's no waiting when we give them in this case it's a very specific thing it's a it's a government constituent request that I'm giving to a citizen uh the app is easy access by development so what what could happen here is development could come in and they already have the scope right so they could see what everybody did they don't have to go chasing update sets and say oh what did they do in that Dev environment because I gave them access right you're not trying to figure that out anymore because you're controlling the access they have to that specific scope you also have a nice package right so you can deploy this either using GitHub because of the scope you're able to push it up into Source control um you're able to do all the things you need after you develop it to make sure that that's controlled and not only is it controlled here but it's also controlled here like if the developer makes changes but then that citizen developer that person in that other department wants to come in and make some changes they're still able to do it because you still maintain that scope as you go okay we can also automate Beyond these things we can automate some of the architecture and um and really improve what we're looking at here so what we're really looking at the parts on this release this released new digital service the parts that we're looking at improving are from when you conceptualize it right instead there's a lot of stories and demands and all those things what we let people do is go right to that catalog and usually through a mediary like in the states we have these people called like it liaison they go out into all the different agencies and they say look this is what it can offer you right so this would be a liaison explaining that this is a good idea if you want to do your own development you can get started right away and so what we're really trying to do is shorten the time to hear to that prototype which can be developed by a non-developer you don't have to wait for development anymore so this is where you see the real big improvements where some areas um and again I work in government so a little bit of a caveat I mean this could take like six or eight months to get to a prototype because of all the paperwork required up front but if people could see a prototype very quickly uh you're really going to increase their confidence on the ability of that it organization to deliver what they need the other area where there's a significant Improvement is down here because now a lot of times if you do give them control in the beginning and it's a non-developer doing some of that even if you just say you can only edit the form right which developers don't really want to sit there and build forms anyway right in the beginning you can do that piece but then here if they want to go in and change oh I named that field wrong well they should be able come in and still do that here because you're still controlled in that scope and they still have access to that scope and development all right so looking at it a little bit more specifically I'm going to go actually go into the environment and show you because I know some of you have access to our environment and if you don't have access you can send me an email and I'll give you access to this environment that I'm going to demo on today so this is the DS Forum instance and I we probably have about 175 admins on this now so I just let everybody in so that they could actually use what what we're showing if if anything's of Interest right so where I started on this is I started developing that 15minute I was a little bit Bolder in the beginning I said um I'm going to develop a 15minute MVP so I'm not really calling it an MVP anymore I'm calling it a a prototype and this is a scoped app in itself So eventually I'm going to push this scoped app up to a git repost so that people can get started with it if they want to so basically what this contains is it contains the catalog item for constituent connect that one I showed you in the PowerPoint here and it contains um the the basic fields that we need to get started so in my case I have a couple different templates because it's going to depend the application template that I use when I get started it's going to depend on um let me show you in here it's going to depend on what type of constituent service they want so is it a service request is it a request for information like I need a a copy of my marriage certificate or is it a license or permit or there might be other types of services that I let them build here depending on which stubs I have and so it's a service request um I can do different things uh to pull those service requests down I would be able to add different fields here that I would be able to put these fields directly into their catalog item right from the beginning so they could see that in prod I can also add things in here like on my template um I could say add a map into my catalog item or add a payment right because in in government most of these things if it's a permit or if it's an information request it might cost you five or 10 bucks right uh to get that information request or as we go on there's different things so the state may have an integration with a certain payment provider or a certain GIS provider or maybe even a digital signature so what I can do here with my app template is I could say as you click on these different things I'm just going to do that for you up front like by virtue of you asking me for a permit with a payment in it or for a permit with a map in it so I can tell you where the map is I could build that for you I could automate that that build for you so when you go in it's already started and so you want to get as much of the information up front so that you can actually build that prototype for them before they start now the next thing the part that I haven't figured out here I just keep track of time for Dan is um when we come down here I want to kick off so once this request gets submitted constituent connect and now I know the type of permit they want um I know a little bit that they want a payment or they want a map in there the next thing I want to do is these workflows are going to run so it's going to build the app is the one workflow it's G to it's going to look and see is it a service request an information request or a license permit so it'll know the request type and then it needs to kick off this workflow so there's there's one application template that builds a information request one that builds a license and permit so I built those templates out this is manual right now I still can't get this automated uh so I'm still working on that piece but so I'm going to show you the one workflow which is really basic and then I'll show you this second workflow that um that comes from an app template has anybody on the phone used app templates before if you just look in the comments or raise your hand has anybody used these okay all right so I'll show you that real quick and then um let's go back to the environment so I go back to the environment the um the workflow that I have is pretty simple for get a new app so what I'm doing is I'm just taking that catalog variables off from the constituent request and I'm able to say okay what service type is it and then from the service type I'm able to make a decision on it's a service request information request or license and permit and then um I'm able to what I'm trying to do here is execute that application template depending on the type it is so let me let me talk about application templates really quick because I'm just going to introduce this so the application templates are um in app engine Studio and what a template is here is I can use this template um for a a new government Lo so this is a new license and permit so I can use this template and I'm going to kick it off manually and I'm just going to call this one two three four and then what this will do is it'll go ahead and create an app for me uh again so it's creating its scope and it's taking all of the things that I put in the template in order to build a new license or permit so now what I can do is in this app home uh what I can do is I can these are all the things that it created so now what I can do manually or automated right ideally automated is I could start building out the people so who are going to be the contributors on this team so I can add somebody um as a contributor and I I could say they're an editor for this app and then I could send this and it would act they'd get the notification um so L if L on the phone he might have just got an email from me saying hey you could come in and work on this app right now right so I basically assigned Lou permissions to work on the one two three4 licens in permit app that has all these things built out already so when he comes in he already has an environment that's working he already has six experiences built including playbooks and um and record producers right what whatever you put in there that template would deploy it for you okay so this is really the meat and potatoes of it right is that you can get them as a developer with a developing a template if you can get them 60 70% of the way there so their prototype is started when they come in and you're not asking somebody to come in as a citizen developer and create their own app app and do their own work right you're not I'm not even hiding this from the citizen developers I'm not letting them create their own app they can only ask me for an app through that liaison and through that request and then I'll give them permission to edit the one 1234 app right like I just sent to Lou but I'm not going to let them those non um service now administrators or developers I'm not going to let them do their own thing and so it's it's a really good way to allow citizen development in a in a very controlled manner so now at some point uh Lou might come in and change the form he might add a form he might do a couple things that he learned how to do in citizen development right so he'd be working in this environment here which is the low code no code environment but then L might say at a point I want to promote this right so the the part that I haven't worked out yet but at some point he might say look I got this where I want it John you know I want to promote this now to a developer so I can get it released so now what would happen is he could push that code now back to me right there'd be another request where he'd come back and and send it to the development team and then the development team is not going to use EP engine Studio likely what the development team is going to come in and they're going to use more of the traditional um development environment right because this gives you a little bit more flexibility to do a lot of different code so they might come in and they might say okay let me go ahead and look at that what did Lou actually do so now they can go into the one two three4 app and actually see that Lou added these forms or Lou added a a record producer right they'd be able to do all the stuff and say is this thing really worthy or is it close to being what we need to promote it into a production environment because the idea here is that only the developer doesn't even see it right until they get to a point where they're serious enough and maybe that liaison like in my case I have liaison from Central it going out to the agencies so maybe in my case I say when that liaison says it's at a good point then I want to get development and architecture eyes on it right only at that point do I want them to come in and look at it if they're do if they choose this path right and not the traditional path which is you know give me a demand and then I'm going to get you a developer to come in and look at what this might look like in production right so I can land it in development in the right directory right so that I could actually see my catalog item um and I could actually have the workflow so we're not engaging development at all in the beginning part of this cycle we're bringing them in later after the Prototype is developed hey I got a few things here and uh oh good Dan's on too so I'll hand it to off to you in a minute Dan sounds good and I I realize I do have the share option I thought I might need to be promoted to presenter but I'm good you're good to go okay I'm just finishing up on the last one here so um the last piece I I I have here really is step four uh part of that template which is really unique to these app templates right is that I will give a catalog item that's actually in production so you or not in production in in the the service catalog so in this case this is the default government portal so just like there's a marriage license here they would have that permit that new permit they'd be able to see it in the portal and they'd be able to go and actually access the the form that started so even though they might not have put anything on the form yet a couple default Fields might be there or they might have clicked on map in the catalog item so we embed the map for them this way we fill out the form of the address they're clicking on underneath so a lot of these things we could pre-build and the three that I see in government a lot or one you usually need to tell me where you need service or where you're going to ask for that permit uh two you need a way to pay for the permit and three you need a way for for a signature in a lot of cases a digital signature so that might be you know any of the digital signature providers so these things you can pre-build into the temp spit so that they don't have to say you know you don't they don't end up putting name address fields in there or anything like that they just use the map instead so as much as you could do to simplify it in the template as much as you can pre-build because you know 80% of what they need in a template right it's the other 20% that they're going to be building later on or ask for your help with right I'm GNA go through um so again the benefits are in here I'm going to share this deck out this is a quick overview uh I have some more on this that actually talks about the team so this goes back to last week there's actually a product manager that would be doing this right a product manager would be managing constituent connect and saying how do I get more people to use this prototype feature the developer that uh product manager rather of constituent connect they work with the liaison team who's the one that promotes it who's the one that says hey we got this really cool thing you don't have to wait six wants for Prototype I can get you one this afternoon right we need somebody out there telling that story about the features because that's what you want to do you want to get a lot of people using a common platform so they're not out there buying apps like crazy like we have uh the problem with today so all that enablement and all those people that would be involved the developer would be involved because they would probably develop the template that I showed you um so they would be involved than doing all that piece and then the architect would be involved we can even automate with csdm right we already know that they're developing a mat right so they're using ezri in this case we we so we know they connected to ezri on the app service right so we know this is a license and permit that uses ezri we know this is a license and permit that uses service now so think about automating the building of the basic architecture that we would need to support this service ongoing so not only can we automate that development piece and get to that prototype but we can also give you a stub of the architecture and you saw where we gave you a stub of what a developer would need to actually carry this into prod that's a quick overview I'm GNA hand it over to Dan now but uh this is something I've been working on if there's any interest in this I'm working with some other other Architects on it too so we can get connected and start talking about this but that's really the phase we haven't implemented this anywhere yet and as you can see I'm still working out some of the things with the workflows to get those app templates kicked off but I think it's a really nice way to deploy that that citizen development or low code no code without being too afraid of just letting people into your Dev environments so any questions I'll be on uh while Dan speaks I'll be in the chat and uh if you have anything that you want to ask on this or uh get in touch or download this I'll have it up by Monday I'll have this recording up with this presentation so that you can use it yourselves to kick it around and see if it's of Interest internally thanks everybody and I'll hand it over to Dan and he's going to talk to us about process mining so we're gonna take a pretty big shift here well actually process mining it can be applied to Applications built with Creator workflows on the service now platform so you can identify an efficiencies your custom applications and continually improve them actually funny story about that um well not funny but true story Chris Betty actually has uh made the statement that process mining allows us to innovate faster uh because he he's not worried about getting the process 100% right the first time now that he knows he's going to have visibility into kind of How It's performing and and where it's going off the rails from day one with process mining in place so it's it's actually a very good tie-in to there's a nice tyght relationship between process Mining and and custom applications potentially created by citizen developers um so I don't know how how you set this up John but I just to introduce myself to the group uh my name is Dan Grady I'm part of the the product team here at service now focused on our in platform process mining solution but I have been with service now I just got my bill mcdermit uh email today for my 8year anniversary I and the majority of that time has been spent working with customers uh when it comes to anything analytics AI machine learning on the service now platform and I saw some familiar names in the the attendee list what I my goal today is to just run you through what what process mining is part of my responsibility is to kind of make sure the service now Community is educated on the offering why it's important uh what makes it different than some of the stuff you may be doing in the past um and I always like to start these sessions with this quote uh do the best you can until you know better then when you know better do better and I love the quote because it usually aligns to the reason that we've all gathered here today right to learn a little bit more about service now platform take that knowledge go out apply it get more value out of our customers investment with service net but it also will align to the solution that we'll be talking about here today process mining which is designed to help us uh automate that process of x-raying the workflows you have running on the service now platform and start to show you where and how you could be doing better for everyone involved with them now what I planned I got a little bit of a why and what of process mining just to kind of level set and get us all kind of on the same page on what it is then I'll jump right into an instance and we'll do a little demonstration and if we have time I'm going to do the demonstration in two phases one is I'll show you how somebody would use what we call the process mining workspace to dig in and analyze and understand and identify processing efficiencies and then one of the biggest benefits to our in platform approach to process mining is the speed in which people can get started with this I always joke that if you can dodge a wrench you can dodge a ball or if you can build a service now report you can visualize your processes in a matter of minutes so most of our customers already have the skills they need to do what you're about to see here today I'm going to show you how we've done that then uh you know it's always good to get a sense of once you see this like what's coming next so I figured I'd run you through a little bit of the what's coming in zanado do and then there is no shortage of opportunity to learn more about this out there on the service now Community page uh for specifically for process mining so I'll just run you through some additional resources if you're interested as well that's the plan and then obviously I'm always hope happy to take questions and answer as much as I can can for you here today um because we'll be talking about zand do I do have to throw the slide up there I didn't know what the rules were John for the call I'm assuming but this is the slide that says hey anything I say and do here today can't be held against me in the court of law it is also the slide that says hey if we happen to talk about things that are coming and we will be doing that make no purchasing decisions I'm assuming you've read the fine print awesome so like with everything we do in life whether it be a meeting like this one or a business process we have a vision in our head a plan for how it should play out and when we design processes we design them for both completeness and efficiency to provide the best experience possible for as many people as possible now unfortunately what we design isn't always what plays out in reality uh the reality is not all work is going to be flowing through the optimal path and that's going to have a negative impact on the experience both people requesting a service or having as well as the people trying to deliver that service the challenging part is actually seeing the reality of what's going on and getting an understanding of it so we can actually improve that process make it better over time if you think about it for for many continual Improvement is a very manual timec consuming and costly effort right it usually involves running a whole bunch of reports and some people still like to print those reports out and go through them line by line with a highlighter don't know why they do that but they do it it may involve running imperson or virtual workshops in which we take a whole bunch of sticky notes and we put them up on a wall then of course somebody's got to snap a picture of those those sticky notes on the wall and post it to some internal site because like hey did work really happen unless we have a picture of sticky notes on a wall somewhere it may involve bringing in some third party to help us out uh super helpful but they're not free um and then once we get through all that we still have to get internal alignment on the optimal path forward and at the speed in which most organizations are trying to move these days by the time they get through all this the optimal path forward may have already changed and this is where process mining comes in process mining automates that ability of showing us the reality of what's going on inside of our workflows showing us where the bottlenecks are the Ping ponging situations the rework in the process where all the inefficiencies are and now that we have that accelerated visibility into the reality that's going to speed up our ability to make the necessary changes to improve that process to drive more productivity back into the organization now for some like I mentioned I've been here for about eight years now I me most of the time been talking about dashboards I have some customers say whoa whoa whoa Dan how is this different than the dashboards that you know we have with performance analytics and uh reporting they give us visibility into how our processes are performing and they're right and I would hope that our customers are using our performance analytics and reporting to answer very important questions like how many tickets cases incidents whatever it might be did we work which types were they how long did it take to close them and probably most importantly how are we performing against our targets and are we improving our cycle times all super important questions that we hope our customers are using dashboards just like this to ask and answer but here's the deal been doing this a little bit a little while now and I've learned one important thing information is one of the most addictive and contagious things out there you give somebody a little bit and they always want more so the first thing that happens when we get a dashboard like this out the door before we can even high-five each other that we got it out the door somebody in leadership is asking us questions like well why are we missing our cat or ESET targets why are our closure times trending in those directions like why why why and the answer to those follow-up why questions they don't always live in dashboards like this this is where process mining comes in process mining allows us to start asking and answering all those follow-up why questions about what we're seeing on those higher level dashboards things like where is the process getting stuck where is there any sort of unnecessary rework happening where are we spending or wasting the most time or which vendors are speeding us up or which ones are slowing us down or my favorite one I've been asked for this report so many times over the years here at service now I've actually given it a name I like to call it the finger point in report everybody wants to know who's holding on to the tickets the longest I need to know who's holding on to the tickets the longest or if you're a more glass half full person we call that the opportunity to improve report these are all the types of questions that process mining allows us to start asking and answering essentially we' like to say that process mining gets us to the why behind our process kpis and we can apply this process mining solution to a single table process like this incident example here where we're looking at how incidents are moving from state to state within the process and the volume and velocity there or if you happen to be using us or a customer happens to be using us for a more robust workflow like an HR onboarding case a life cycle event with multiple activity sets and sub requests and tasks that come off of that we can apply something we call multi-dimensional mining that gives us endtoend visibility into that entire process these multi-dimensional Maps super useful when we start looking for bottlenecks in our request processes if you think about where the two primary bottlenecks are in a request process is really two places the approval step and the Fulfillment tasks themselves and if you think about that those two things live in two different tables on the service now platform the approvals typically live at the requested item level and the the Fulfillment tasks are in the SE task table reports can show us the overall time spent on a request but process mining allows it to look at where is that time being spent is it waiting for somebody to approve it is it in the Fulfillment test themselves being able to put these two things side by side on a map really help us understand where we're leaking air from a productivity perspective and where we can start plugging some gaps and improving that process for folks and then lastly as of the the Vancouver release we we've opened this process mining capability up to allow customers to import external process data into service now and apply process mining to it so the example that you're looking at on the screen here happens to be data process data pulled from Smart recruiters from a hiring process process and we're able to apply process mining on top of that to start looking for bottlenecks in that process here at service now actually um but that ability to open it up is is something that's available in the Vancouver release and then in the Washington release we actually just open it up even more to allow you to create those multi-dimensional maps with both internal service now data as well as external data uh for situations where you want endtoend visibility into a process that might be touching both service now and someplace else um I'll just pause here before I jump into a little a little demonstration piece any questions comments thoughts about anything that we've covered so far actually I should ask I don't know John how these calls typically go do we do we open it up for questions or do we just put them in the C the chat yeah I'll watch the chat for you so if there are any questions that come up anybody using it or um want to come off mute that's fine too awesome all right I'll jump over here to a little bit of a a demonstration for Dan sure go ahead can can process mining be applied to any Tas table Yeah so the the technical requirement uh for process mining is in service now at least is audit log data actually I'll take one step back technically um there's three key things from a data perspective we need for process mining you need these three values you need a unique identifier uh an incidence right uh or if it's an external system maybe a claim ID a patient ID or something like that a job requisition for that hiring example you need a unique identifier you need something called an activity or in some process mining circles they call it an event like what is the thing that's changing on that unique record over time very often we're just looking at State changes but I'll show you an example today when we look at assignment group changes uh then the third thing you use is time stamp uh for when that change took place now Lucky For Us in service now all that data lives in the audit log table so every time a record on service now changes like say the state changes the assignment group changes we capture that data in the audit log table so for service now we're just harvesting data from the audit log table auditing is turned on by default for task and I think something like 150 plus tables across the platform um and then you know for most customers they specifically have to go them turn to turn auditing off if they don't want it on uh for your custom um example like you're building a custom application that's one area if you're thinking longer term and you think you're going to be using process mining on top of a custom application just Ensure that you have ating turned on for that the tables in that app and you can apply process money to it um so yes it works on tasks um one more for you yeah is is it a Plugin or it's its own scoped app or how does it like yeah uh so the way that it's packaged um it is a plugin you can go out and get it on the store um as a core process mining plugin you can install it in your sub production instances and you can uh mine up to 3600 records in with in your sub production instance there's also content packs for the individual workflows on the platform so for itsm CSM and the industry flavors HR uh field service management strategic portfolio management uh we have content packs and you'll see this in the demonstration that just give you some common process Improvement or inefficiencies that we see they come bundled in these content packs so those are usually most customers turn on two plugins the core process mining one and then there's a Content pack for whatever workflow that they're they're entitled to mine okay um and from a just a packaging perspective process mining is bundled with itsm Enterprise and CSM Enterprise and all the industry flavors of that enter CSM Enterprise uh but we also um have broken process mining out as its own thing for those customers to say you know what I don't want to go all the way to Enterprise but this process mining thing is super cool super valuable I want it there's an an add-on skew that they can use to get entitlement to process mining excellent so Sarah Allan Mike did that get all your questions answered I'm not going to answer the pricing one so okay Sarah says yes Mike you good and Allan's good he gave a thumbs up okay perfect all right so I'd like to start these demonstrations inside a performance analytics dashboard you do not need to be using performance analytics to use process mining but they are very complimentary Solutions right the what and then the why so you might be on a dashboard just like this and you might see let's say your resolution times on the rise and then in the single click you'd be able to drill down into the process mining workspace to get a better understanding of what might be causing that right and I'm going to give you a little bit of a a harbor what you're seeing here in the the process mining workspace I'm just going to move people's faces out of the way for now um in this case here what we've done is we've mined 21,000 uh incidents 21,400 incidents actually I and they've taken 1,500 different routes to get to closure now what we're looking at on the screen here is we give the the person who mined this data inside of this process mining project a view into kind of the average completion time of those 241,000 records over time and then a distribution of kind of the different buckets it took from a closure time perspective this is the out- of thebox uh dashboard that we provide or or visualizations we provide but then customers also have this ability to start to add their own metrics and kpi to this same view the idea being we want to keep the person doing the analysis aligned to the higher level kpis that we're trying to influence um over time and for incident obviously you're going to have your mttr your backlog and your average age most likely but completely configurable you want to tell that complete story to the person doing with to the person doing the analysis now as we move down the screen we get to what is my favorite part of the solution everybody loves the visualized process maps and they're awesome and we'll look at that in a second um but we also package up as part of those content packs that I was mentioning these uh preconfigured Improvement opportunities or we call them under the covers finding definitions and these are just common inefficiencies that we see across customers and they come in two flavors uh rule-based and an automated and I'll walk you through both the first example here is of those 21,000 records that we mind there we flag this one here these incidents that are going into the awaiting caller info State at some point in their life cycle and there was 8,100 or 38% % of the incidents that we mined went into the awaiting caller info State and probably the most important number here is this 42 years of total efficien inefficiency there's 42 years worth of time packaged up in just these incidents going in and out of that awaiting caller info State like even if a customer could reclaim 1% of that 42 years that's going to be a win for them and then right from here they'd have this ability to drill into the visualized process map for just those 8100 or 8200 incidents like just jump start them on their anal itics journey and prioritize where to start their digging their further digging into this this is an example of a rule-based one and I I say it's an example of a rule-based one because we specifically configured it to say go find us the situations in which it touches an incident touches the state of awaiting caller info now the next one on the list here is an example of a an automated finding or Improvement opportunity it's telling us that it found rework on the in progress State meaning that an incident went into the in progress State left it and then it some point in its life cycle came back to the impr progress date and in fact it found at least one incident that did that 11 times so there was 2,000 of these and there's 33 years worth of time packaged up in just that rework scenario of going in and out of the in progress State again an opportunity to improve and that I call that one an automated one is because it it's just looking for a process pattern we have these inefficiency detectors or pattern detectors that we've rolled out uh one of them is rework so it looks for situations in the process where process goes a b c back to a let's say and it will flag all those situations we have a a pingpong detector that looks for things that are going ab ab ab very handy when we look at assignment group um scenarios we have an extra step detector where it looks for it finds patterns that go a b c DF and then a variant that goes AB bcde e f and it says hey we noticed this one additional step for this bucket of work slowing you down might want to look at those so there's seven of these detectors now as of Washington and they're just meant for those situations where we don't know we don't know and then we flag them we put them on this list we prioritize them by the total inefficiency time they may be interesting to you they might not just like a dating app you either swipe left or swipe right depending on whether they're important to you or not or you're interested I mean you move forward but the idea is they're really helpful in jump starting you in terms of where you might want to start continuing your conversation with the data I always like to compare these to when I was playing hideand-seek as a kid um when I was playing hide-and-seek as a kid you know like I when I was at home I knew the two or three best hiding spots so when I I had them I was when I was eight I was way more efficient at finding my friends than they were at finding me same is true with a lot of our customers right they they have a pretty good idea of where their inefficiencies are they just haven't a good way to get the data they need to quantify the impact so they can prioritize which ones to act on first right no shortage of opportunities to improve but you need data to help you start um understanding where you can you should focus your efforts first so that's the summary and insights page then of course we get over to the process map I'll just give you a little bit of a harbor tour of some of the things that you can do inside of here same data as before right and I can come in here and just expand this out I've got my 21,000 incidents my 1500 different routes and you can say that doesn't look like 1500 routes and I can make it really messy really fast if I expand this out then your GRC people get really scared and squirly because like all those little lines could be risk um what you're looking at on this example here or in this map here is you're just looking at how things are moving from state to state within the process and the volume and velocity between those transitions now for me the first time I saw a map like this I had no idea what I was looking at but they kind of reminded me of ski trails I and if I think back to when I used to ski anytime I started going down the mountain and then I had to go back up the mountain certainly wasn't the best experience for me and it definitely wasn't the fastest way to get down the mountain so I always start to look for these lines that are going back up the hill because they represent some form of rework in the process um and in this case here we got this one these awaiting Coler info back to the step state of in progress and and I can see that I have about 2600 unique occurrences of that 3,300 total occurrences meaning about 700 incidents took that trip more than once and then I got my one of them that took the trip nine times and then probably the most important number is this 18 years of total duration right packaged up and things just going from a waiting caller info to in progress crazy thing there is that things don't start out and awaiting caller INF they have to get there so what can you do right you can use this bottleneck analysis to come over here and say show me all the situations in which there's a transition that involves the state of awaiting Coler info and here's the one that we were just looking at 18 years here's the other half of that 21 years from in progress to waiting caller info so what you can do is you can actually come in here and then click on this and say hey only show me or narrow it down to only the tickets that touch the state of awaiting caller info at some point in their life so now it narrows it down to those 8200 you might remember or that 817 3 you might remember that from the summary and insights page I could have just jumped right to Here by clicking on that but I wanted to show you a little bit more of the interactivity then of course I can use these breakdowns on the Le hand side of the screen to start slicing and dicing these 8200 incidents that go into the state of awaiting Coler invo at some point in your life and one of the ways we usually look at these things that touch a waiting C before we're going to on hold is the intake channel right because that we can improve the intake experience we may be able to eliminate some of the back and forth and I can see that the majority of these things came in via self-service and portal but check out this email Channel there's a whole additional week packaged up on things that come in via email and go into a waiting caller info so maybe we want to focus in on those kind of make sense right emails a pretty unstructured Channel you expect some back and forth but a whole week like we could totally do better then I can do some other things right now that I've narrowed it down I can use this histogram to look at the situations where that happened it went into waiting call INF for two three four or five times I can start to look at let's say the individual records themselves right benefit of being in the platforms being able to get down to those detailed records maybe we want to start using some clustering to look at the short descriptions the descriptions all that unstructured data in these and we can start to see that hey I've got this cluster of tickets in which people are sending in emails about updating their email address and their profile like that's totally something we should have a catalog item for or some sort of self-service so what we can do next to close the loop now that we found an opportunity is we can come in here and we can add a note and say hey at soand so John take a look at intake or email update attach a snapshot you'll get notified you click the link it brings you into this project you hit the preview button it takes you to the same place in the analysis that I was and then there's also a connection here with continual Improvement management and automation Center I think I saw a former session that you had John on automation Center but the idea is when you start using something like process mining there's going to be like no shortage of opportunities as I've said you can't act on the but you want to make sure that you capture the opportunities so they get tracked prioritized and followed up on this allows you to start either creating new initiatives or linking to existing ones to make sure that we things don't get lost in the shuffle I know that was a lot and I kind of probably went pretty quickly any questions comments thoughts that have come up in the chat there John yeah alen was asking about the uh the duration for like it's 21,000 records we're looking at in this case um up in the upper left there but how long is it like over six months or it doesn't really matter or yeah you you can configure it we'll actually I mean with the time here let me we can create a project real quick but you configure it just like you're configuring a report like you can run this on one record if you wanted to like that's you know what we're were talking about the the create a workflow situation right if you just roll an app out one record through it you could run a project or create a project against it and see it usually we look at like a month or a quarter's worth of time um for for a given process and a subset of the data just to get that gives us a good amount of data and an understanding of how the process is performing um yeah yeah that's I was thinking crawl into it almost at a singular level as part of your postmodem after a P1 event or something like that and look where you lost your lost your yeah you lost your time yeah you actually can do that's a great point one of the things we didn't I didn't show you here but you can actually do this side by-side comparison so what you can do is you can do before and after a process change um to see what the potential impact of it and you can start to look at at like any of the changes um the increase in the average duration the decrease so you can start doing these side by-side comparisons different teams different regions um but very often it it becomes the same process over time like before and after or different version of the process um very handy for looking for adverse effects of a process change um so real quick we in Washington we rolled out this guided setup for creating um a pro you know a project so it's just like I said kind of just like building a report you kind of say okay here's my project what table do you want to mine any table that has auditing turned on is going to be available to us and we just kind of come in here we'll say create project for the incident table and then you just start scoping your data right maybe we only want to look at human generated incidents maybe we only want look at incidents that were part of that that uh P1 scenario you were talking about there so what you do is you come in here and you just add some filter criteria you want to be focusing on d uh proc work that's completed the process right you don't want to be comparing open and closed work together your durations get all wonky you're comparing apples and oranges there but you just keep coming in here like you're building a report maybe a certain time period certain category whatever it may be I don't really have to worry about that I don't have a ton of data in this instance and we hit save then the only other thing you absolutely have to do is you need to set your activity definition right the thing that is changing over time on the record very often we start with the state activity right and look at how things are moving from state to state but you could totally do this with assignment group very popular use case to look at how things are transferring between different teams and the bottlenecks and the transfer process and then of course we're going to want to have our breakdowns right those remember we use channel to look at things coming in via email you simply come in here and you say well what what do you want to break this down by and you can have up to 10 breakdowns per um per project and we'll save this and then we'll choose our Improvement opportunities right so remember those those things on the first page right that they come in the content packs the automated findings and the rule-based ones uh we'll just add a whole bunch of these and a whole bunch of these we'll hit apply to project some of them will apply some of them won't but that's where those those fi initial findings come up come and then we say review in mind and it will package everything up give us kind of a little bit of a summary rate we say mine project we're going to do a sample mine um this is going to just mine those 3600 records that I was mentioning what happens now just so you know based on the criteria that we put in active is false it goes out it harvests the relevant data from the audit logs it then takes that data it passes it off to the machine learning infrastructure to do all the heavy duty number crunching um so there's no performance impact on the instance itself it spits the finished model back into the instance and then we go on our primary way um and use the project just like you saw me do this takes anywhere from 30 seconds to a couple of minutes depending on the volume of the data and our place in line with that shared machine learning infrastructure but that's the level of effort like that's the thing that gets our customers really excited is they see the level of visibility that process mining can provide and then when I walk them through this little wizard here and they see that it's not like rocket science to get going with it um it's a really compelling event like it's not this heavy lift to get started with process mining um I'll while this runs I'll just pause are there any questions that had come in there John that you see yeah Ellen was asking um and I'll expand that a little bit but what what about a process that goes into another system or or the whole process another system yeah so in that scenario as of Vancouver we open this up and I I have um uh just so I don't know if I have a slide on it let's see uh yeah I'll just bring this up so while that m so we we have this ability in Vancouver we open it up there's a little wizard process that walks through importing the process data from an external system basically use import sets under the covers but there's some magic that converts it into native audit log records in the platform so you can in Vancouver there's an option to import external process data in Washington we opened it up to create those uh uh Maps those multi-dimensional maps that combine service now data with external process data into a single map and I run these process Academy sessions on a monthly basis there's one specifically focused on the external data offering um and it walks through the whole demo of importing the data and then the most recent one I just ran last month that's out there on the community site is the walking through the process of linking the internal and external together if you you were interested in learning a little bit more U after we leave here today and that thing you mentioned before about a project is is that that specific project for doing the the mining activity right it's not it's not like a regular project in like yeah exactly exactly it's it's just what we call kind of the the configuration right so like you have all these different types of analysis that we're trying to do like hey okay here's a project in which we were looking at assignment groups instead of State changes uh so it's not like a a PPM or a SPM project it's just the the configuration like we just walk through to say what type of analysis do I want to do oh looking at this dat with this configuration boom boom boom boom boom yeah there somebody commented on it it gets confusing with which which project in service yeah to make it even more confusing this product used to be called process optimization then in Vancouver where we named it process mining that was a really fun exercise to go through yeah and um yeah the process mining Community uh that's that that's on the forums yep so there's a specific product hub for process mining can you can chat that link right there yeah sure I'll people are asking for it so that'd be great yeah we'll put this in the chat um and just while we have this here just so you can the top two things are probably the the the most useful right there's uh the use this thing we call the use case series these are just little five minute recordings on how to use process mining to do different things like multihop analysis SLA breach analysis so that's a pretty handy one um and then there's also a the academy that I was just talking about and this is those 25 recorded sessions of different aspects of the solution like doing that external data one so there's a whole 40 minute presentation on how to do that okay um the other thing that go I'll call out is I just posted this actually for those that are interested as this guide to getting started with service now process mining this is like my own little curriculum of how I would walk somebody through there's like a two-minute explainer the little demo that I just did is out here is a recording on how to create your first projects this is recording on how to use that workspace and then there's the use case Series so this is also a good article I might just put that in the chat too for the team hey Scott I know we're over but did you have a question I I didn't see you had your hand up sorry about that SC Scott you still there do we lose you oh no no question John I think that was I put it up by accident oh okay it was a high five not a question a high five high five yeah great stuff very cool perfect well if anybody has any questions I mean I'm out there I think most people probably know how to navigate the service now email system um or find me on the community or LinkedIn I'm happy to follow up and answer anything that you have and if you send your deck over Dan I post that with the video on our forum and um sweet so this way anybody can get it so appreciate you jumping on that was awesome no I appreciate the opportunity and the audience really great and I hope to see some of you at knowledge I'll see you there thanks guys thanks John