one of the biggest advantages to taking an in-platform approach to process mining is the speed at which she'll be able to start getting visibility into the reality behind your workflows in a matter of minutes you can have insights into your process and efficiencies non-conforming activities and Improvement opportunities now as an in-platform solution process optimization helps our customers avoid many of the data access and availability challenges and data manipulation work that is typically part of a process mining project in some cases that data preparation work alone can take weeks another Advantage our customers have is they already have the skills to configure their first process model I like to say if you can author a report on the servicenow platform you can create a visualized process map we're going to use the next couple of minutes to walk through the process of creating our first visualized process map with process optimization so what you're going to do is you're going to come over here to your left hand navigation and you're going to search for process optimization once you get to that application you'll click on the analyst workbench once you land on the analyst workbench you're going to be presented with all the projects that you've created or have been shared with you and we're going to use the create new button here to create our first project we're going to be mining incident data here today so let's come in here and we'll say incident or actually you know what let's call this my first process mining project and we'll hit save now in the Tokyo release we we added the ability for you to create process models that involved interconnected processes or multiple tables so what we're going to do today we're simply going to be mining the incident table but we're going to come down here to the table configuration section and say new to add our incident table to this model we'll say incident as the name and then we need to choose our table in this case we're going to choose the incident table now any table that has audit records is available to be mined in this list so we're going to choose incident because that does have audit records and then what we need to do is we need to start narrowing down the scope of the data that we want to mine to help us answer the questions that we want to answer as part of this mining exercise now best practice dictates that we mine close records in this case it's going to be closed incidents that's going to give us an accurate representation and make sure that we're comparing Apples to Apples and then you might want to add some additional conditions one of the things that we typically like to do is separate machine generated incidents versus human generated instance so we may choose to come in here and say hey you know what from a channel perspective I don't really want to include things that are generated by alerts in my analysis here today and then you might also want to start bucketing it from a Time range perspective right you don't really necessarily need to mine two years worth of incident data I just want to mine a relevant set of data for your use case sort of again for the questions that you're trying to answer with this mining exercise I'm going to come in here and I know that my demo data um all was closed before this year so I'm just gonna come in here and narrow it down to incidents that were closed before this year I can use this preview button to come in here and give me a sense of the records that I'm reminding if you were to click on this it would show you those records and you can just make sure that they match the criteria that you're interested in once I've narrowed or scoped the data that I want to mine you can click the save button and we really only have one more step that we need to take before we could actually mine this data and that's what we we need to add an activity definition now the activities those are the things the Hops or the changes in the records that we want to track over as the workflow moves towards closure now typically what we start off with is we want to start tracking the state changes within a workflow that's usually the the first activity that we start and for our first mining project so we're gonna we're gonna add state but you could also add things like assignment group if you want to start looking at assignment group hops inside of there or maybe you want to start looking at when a problem was created or attached to the incident you could add that as an activity definition we're just going to keep it as state and I recommend for your first project you keep state or you state as your your first activity now that's it we could be done we could mine the data right now but to help with our analysis we might also want to add breakdowns breakdowns help us slice and dice the data that we're going to see on our visualized process map so typically you're going to use things like hey I want to break these the map down or the data down by categories or you know a very popular type of analysis is channel analysis so let's add our Channel you can add up to 10 breakdowns per project so we'll come in here and we'll head Channel and we'll hit submit and that's it we're done we've configured our first model so now what we'll do is we'll go to Project definition and then you've got two options we can generate full model we'll run this mining exercise against the full set of data that we scoped with our filter criteria or you can generate sample model and this is just going to generate it against a subset of data I'm going to use generate sample model here I recommend for your first try use sample model as well and what this is doing now is it's harvesting the relevant data from the audit logs to match the filter criteria as well as our activities once that data is harvested it's going to pass that data off to the machine learning infrastructure to do the heavy duty number crunching one of the questions we typically get asked is hey is process mining or process optimization going to have any impact on our instance and the answer to that question is no because all the heavy duty work is done by the machine learning infrastructure once that data is picked up by the machine learning infrastructure it crunches the numbers and it's going to spit the finished model back into your instance and then we'll have a visualize process map this can take anywhere from a few seconds to a couple of minutes depending on the amount of data that you're mining you can also schedule these so they run off hours or be notified when it's finished in this case it ran very quickly and now I can immediately go to my analyst workbench we're done that was it that was a matter of two or three minutes I could check the clock you authored and created your first visualized process map now what you get by using the out of the box content packs that we provide is you get this summary and insights page which gives us visibility into some of the key kpis as it pertains to the process that we're focused on in this case our incident process so we're looking at things like our mgtr as well as our first call resolution rate it's all about the speed and productivity when it comes to this process you also get pre-configured insight cards to help us identify common inefficiencies and problems with our workflows or non-conformant activities so things like hey I've got records that take longer than seven days to go from closed to resolve we can focus in on these or let's say here's a quality issue where incidence resolution has been rejected by the customer one more times I can quickly click on this and take me to a map of just that subset of incidents that at that criteria we also get variation analysis which shows us the common routes that the tickets are taking with the long running routes if we want to focus on that type of analysis as well as bottleneck analysis which is going to show us the individual hops that the incidents are taking in this case that's just the summary and insights page then of course you have your visualized process map which shows us the paths that all the incidents are taking to get to the closure or just looking at the top 20 percent here we can expand this to look at more incidents we can focus on focus in on individual hops within that to see the volume and velocity in which incidents are moving through those certain Hops and then of course we have those breakdowns that we configured in our model on the left hand side of the screen to help us start analyzing different aspects of the incidents that we've mined in this case all of that power at your fingertips in a matter of minutes huge huge uh Advantage by taking an in-platform approach to process mining to get started what you need to do is activate a few content packs the process optimization content pack as well plug-in sorry as well as the content pack for the associated workflow and there's details around that in the related blog post in which you found this video