so again we are at the top of the hour so we're gonna go ahead and get uh started i think people will continue to trickle in uh but that's okay so this is definitely recorded so whatever they happen to miss they'll be able to get to it so welcome welcome to the platform uh platform analytics academy um it's been a few weeks about a month since we have um talked and and seen everyone uh if you were at knowledge and in any knowledge and we were able to meet any of any one of you in person uh that is an awesome experience that we really love and we missed that uh over the past couple years when we weren't able to do that so it was great to see anyone that we did meet and hopefully you were able to get a lot out of knowledge whatever location that you went to so with that we'll go ahead and get started here so maybe i will there we go okay as usual these are for you so we do try to bring fresh ideas better understanding and some practical guidance where we can this is being recorded and of course it'll be out on the community later on today and also on the youtube channel uh q a please feel free to ask questions you can ask questions throughout use the q a window if you can rather than the chat i do have some panelists that are on as well that will be monitoring those questions and if they can answer them while we're actually talking then they'll do that via type if there is a question that they feel as though we should stop and answer because it's relevant to whatever the content is at that particular moment we'll do that as well and of course we'll always take the time at the end of whatever time we have left and we'll ask questions there um whatever ones that are related to the content we'll try to do those first and then if there's any other ones that we can get to that are unrelated then we'll try to make sure we dedicate some time to those as well if being recording isn't recorded as an issue for your organization uh you may want to go ahead and drop now but um so because it will be out there in the public all right so let's see we got so again my name is thomas davis if you were watching the welcome video uh you probably saw that i had dan down as a presenter i apologize for that i'm actually going to host it and present it but dan and tara who are my co-workers they're going to be the panelists that are going to be on and again they'll be answering questions um and they'll probably speak at some point as well they are as knowledgeable or if not more knowledgeable than me so we have really good um you know ability to answer the questions today so um uh where there's a question that just came in and i'll uh tara's gonna answer that but you'll find out here in a little bit during the process of it all right so today i'll just post the link and if anyone wants to grab it they can sounds good thank you tara so today we're going to talk about getting started with servicenow uh performance analytics now this may be um elementary to some people that are on the call but we also feel as though it is important to recognize that we may have some first-time visitors or some people that um maybe have been inside the platform for a period of time but have not had to do anything with performance analytics and we find that uh you know there's some value to revisiting some things that maybe some people uh may already know or maybe even know better than us but i think it's a good idea or we think it's a great idea to revisit those things so that's our intent today so uh safe harbor notice um there may or may not be things that are in this that are forward thinking and please don't make any decisions on buying based on the presentation that you have today um this information is probably you have seen it before but we want to make sure that we stick that in there so with that in mind let's kind of get into some things so always one thing that we want to make sure that people do is plan and prepare for when they decide to start using performance analytics within servicenow i am a former customer so i have a different way maybe of looking at things um dan and tara i believe tara both dan and tara also are former customers so uh we have been on the other side as it relates to starting this whole process um but we want to make sure that you think about definitely certain things that you should look at prior to actually getting into the nuts and bolts of things so we've kind of defined a couple here uh there probably are more but these are some high level ones so you always want to make sure that you're able to identify the business process what do you want to measure what do you want to improve is it it shouldn't just be about let me just throw some reports or some form analytic performance analytics out there and hope that meets the needs no find out what the needs are of your organization you may know this from having a prior reporting tool or bi tools so you kind of already have a good idea of some things that you know that you need to report on but just make sure that you're able to you know give them something that they're able to measure the organization on and approve of course the organization is what as well so uh determine the factors and behaviors which will affect the health of the business process again do things show things that will help the organization get better and decide on the metrics to measure so know what those are uh prior to building anything out one thing that uh i always try to tell customers when i talk to them is and this comes from really being a former customer is try to get into the process as early as you possibly can don't let um reporting or performance analytics be an afterthought meaning when you're in the process of launching the platform and doing all of the work up that you need that you need to do to be able to to launch it be a part of that process actually get with the developers and also get with the the stake owners and say what exactly do you need that way you can understand any part of the platform that you need in order to report on that also knowing the table structure so as that development process is happening being aware of the way that tables work and the tables that talk to one another that you can dot walk to that is going to make the process of reporting and pa so much easier to you rather than thinking about it later on quite often in business and this is not uncommon is that during the rollout of something it's all about getting that getting that let's make sure we have a date that we gotta hit and then it's ready to go live and it goes live and then it's like okay now we need some reporting or something some sort of measurements off of it if you're part of that process early on then you can be ready during that launch date as well to make sure that you have all of that pa available and ready for the customer so it's seamless they they've launched it all you know things start getting created whether it be incident requests change many other things but you're ready with that pa already so as it goes so is the pa all right so those are just some really high level things to think about as you start your process in this so i've kind of broken this down into what i would call four different things of the very highest level to be aware of as you're you know starting this process and i'll just kind of high level and go here and then we're going to go into each one of those so the big thing is activate out-of-the-box dashboards listen we we've done this for a while and i can say honestly that a lot of what most customers need i don't want to give a percentage to it but i think a lot of what customers need we have already created and really done that based on what other customers have asked for ideas that have been put into our ideal portal in our community say okay we need this just making our product better we have done all of that within out of the box things that are there whether it be dashboards the data artifacts that make up the dashboards the indicator indicator sources um you know breakdown sources breakdown mappings there's a lot that are out of the box so we want to make sure that you recognize those first rather than trying to recreate the will or recreate something because maybe you don't think that it's already there investigate look see what comes with out of the box information that way you can use that and that of course speed things up for you as well there's there's no process of recreating when you know that it's already there once you've done that then you have to collect the score so everything that comes out of the box it's there for you but there's jobs that have to be ran in order to bring data into those different data artifacts so there's historic jobs and there's daily scores that you you can run as well and we'll talk a little bit more about those as well and then once you've done all of that then obviously you need to check the results you want to look at those widgets inside the dashboards and the dashboards themselves and then also inside of the analytics hub where you can look at indicators directly that would be your way to sort of true up and say okay yeah what i expected the outer box to be there it is it's showing me the data and we'll talk a little bit about that as well um so with that activate so identify your entitlement is probably the first thing that you want to do what does that mean what have you bought right what do you own as part of your entitlement that you have done so you'll need an admin if you're not an admin to go to subscription management and once you're there you'll actually be able to see everything that you have and there are a lot of different performance analytics things uh quite often the very the one that's probably the most highest and what people think about the most is incident but there are a lot of different uh things within performance analytics that you can report on and there's a lot of entitlements that you you have access to based on what you have um actually bought and there's also a subscription management docs page that can help you walk through that process and so you can actually see it so also once you've activated content packs that are out there then the best way to do to look and see what you have is to go to system applications all available applications and all and you'll actually be able to see what you um actually have as a that's a part of content packs that you've actually loaded up so again do this first see what you have available to you before you try to recreate um anything or create anything so um a really good source for you to go to that actually is out there that shows you what is available to you via different subscriptions that you have there's a docs page that we have on available analytics and reporting solutions this is just a snapshot this page is very long in our docs site because there's so many things that you can activate that will actually come with pa when you do it the other thing too is a lot of times there are things that are updated that may be in between actual releases so here we're going to tokyo in the fall right now we're on san diego but maybe there's something that the developers have created that they've decided we don't want to wait till the tokyo release we want to get this out there now because we think it's really good for customers if you go to servicenow store you can find all of that information and these are links that are clickable that you'll be able to to get to with the pdf that we say this is but also there's useful links that i have towards the end of this slide deck that you'll be able to get all of these as well but again go there and visit so you'll know everything that is out there that is available so now that you've actually done and uh you know installed some out of the box things let's get into the nuts and bolts of what you actually have if performance analytics is new to you so the first thing is indicator scores and indicators um basically before you create any new indicators try to you know try to use the built out and dashboard indicators available that again are on that same analytics and reporting solutions page we're going to talk a little bit more about all of these uh here in a second so there's also additional conditions that you can use we'll talk about that breakdown sources and mappings so you create a breakdown source and then how do you map it to the actual tables that you're going to use the breakdowns with we'll talk about that a little bit more and then there's scripts so there are some scripts that come out of the box i will tell you i sort of live by the rule the less scripts that i have the better off that you are uh it's sort of you know that's one less place to look if if something were to go go wrong uh but there are also we recognize that there's times when scripts are needed and need to be used so we'll talk about that a little bit as well but you want to make sure that you review all of these things that come with that out of the box uh information that's been installed for you when you do that so let's talk about um indicators breakdowns and scripts so first let's talk about indicators so an indicator source is a facts table it in in this example here we're using the incident table so it says okay what do you want to get from the incident table now there's a lot of information that's stored inside of the incidence table and i would tell you any way that you can to basically drill down the amount of data that you decide to bring back in an indicator source i would suggest that to you at some point maybe not right out the gate for you but at some point there may be millions of records inside of an incidence table and it's expected to be that but when you create indicator sources can you drill down that data a little bit so you're not pulling back as much or it's not looking for as much incident day but incident data when it looks for that and again there are already a tremendous amount of indicator sources that we have created for you um that are out of the box so make sure that you look for those so then you want to be able to apply conditions to an indicator source so here what we're saying is i know that the incident table is what i want to look at but there's really a subset of data that i need from that in this example maybe you want to look at how many incidents were closed today so by putting conditions into your indicator source you can fine-tune that sort of you know query to say okay i know that there's a lot there but really all that i want is what was closed today and with that you can look at an aggregation of that subset so here how many sis ids you want to bring back which is a number of those there's also columns that are within that incident uh data that have counts in those so maybe you want to sum that up so for this example here we want to say okay i want to see the number of incidents that were closed today and then i'd like to sum on the reassignment count so the total number of reassignments an incident may have that was closed today so incidents can get reopened after they're closed and there's already a column that's inside that incident table that will actually give you that number but when you look at that maybe you decide that you need a formula indicator to actually show you the average number of reassignments of incidents that are closed today this is actually already out there for you it's an out of the box indicator and also a formula indicator that comes but here is just kind of helping you understand what you can do with the incident data and understanding that you want to make sure that your source that you created um is not it gives you the ability to not look at so many the really great thing about indicator sources as well is the ability for you to put governance around the data that people are using to report on so if incidents let's say incidents open today this week this month if there is a particular way that you count incidents that are opened then with an indicator source you're able to actually have the governance around a way that you expect people to report so reusing an indicator source and indicators is something that is common and is that and it's something that we definitely recommend you do not need to use or create an indicator source for the same thing even though you may have some different conditions that are on it so you set some conditions at the um indicator source level but there you are you actually have the ability to also set conditions at the indicator level so when you have an indicator that's pointing to an indicator source the indicator source says i want to look at incidents that are opened today so it's going to pull back all of those incidents that are open today at the indicator level i can say okay not only do i want to look at the incidents that are open today which that indicator source gives me that but i can also set additional conditions at the indicator level to say but i only want to look for incidents that are open today maybe based on this particular assignment group or this list of assignment groups that's where you're able to really define even more the data that you're looking for inside of that indicator now you may have 10 12 i don't know whatever number it is looking using the same indicator source as its base data but all of those 10 15 20 indicators may have different uh conditions that are set that are not only looking at that subset of data from the indicator source but if you you have now applied different conditions at the indicator level that you can now you know fine-tune that better so again it's a really good governance tool to say all of you use this same indicator source because we know this is what we've decided is opened or closed now if there's anything else that you need to do with that data then set those conditions at the indicator of themselves so really good uh ability to put some really good governance around your data so next one of the things that you're going to want to do is you're going to be able to break down that data and look at it differently maybe you want to look at it by priority maybe you want to look at it by the assignment group maybe you want to look at it by the state creating breakdown sources and then mapping that data to the different tables that you want to use on is how you're actually able to look at that so here we can see the same thing the facts table of incident maybe you want to create a breakdown on some data that is inside of the incident table so when you look at it in a widget you can change that data so a breakdown source basically references a table a choice list or a bucket group most of the time i find that i use a choice list this is something that is maybe in the cis choice table and again we have already created so many of these breakdown sources and automated breakdowns for you so that that you hopefully won't have to create any new any new of those but i still want to explain to you the way that it works so here we can say okay we've created a breakdown source on the assignment group and we've also created a breakdown source on pr priority well how do i tie them back to that incident's fat uh facts table well what you want to go what you're going to want to do is create a breakdown mapping inside of the breakdown source it says okay priority i know that i got from my sis choice list but i want to tie that priority over to the incident table pointing to the same data so it knows what it is it knows that there's a priority in the cis choice and knows there's a priority in the incidence table when i map those now i can use that breakdown with the incident table you can have multiple breakdowns on the table you know you can again maybe you want to create a breakdown on assignment group or the state or you know the priority or you know things of that nature you can create all of those and by doing the breakdown mapping it gives you the ability to do that you can take a breakdown source and also map it to a different table so not only have you used a you know priority if your priority is the same for request then you can use that same breakdown source and then create a breakdown mapping on the request table so again governance right it gives you the ability to know that you have created or you're using out of the box something that we've already created for you this is what you're going to tell your organization like if you're going to create a breakdown on priority we know this is the same for all of our tables not all the time but most of the time it's the same this is the breakdown that we want if you need to use another mapping then let us know when we'll make sure that we create that br that mapping for you so again so much is out of the box make sure that you're looking at what is there so next we'll talk about scripts so sometimes we recognize that there are you need to script some data because maybe you're not able to pull the data out of it um without creating a script and quite often i think a script that's used most often i think is when you're trying to look at the duration of something so again you're looking at incidents closed today this is the indicator source and you can actually add a script which will create a virtual column to a record you know set in which to add values based on something that you've created in the script and so for example that we're showing here is maybe you want to look at it before a due date maybe you want to create a yes no column inside of the incident table that says i want if something's based on this i need to flag it as a yes or if it's not it's based on a note and then maybe you want to look at the age of an incident by using a script you're actually able to do that and again you create these virtual columns that are being called from the script or being generated from the script and now you can report on that incident data or other tables whatever you do uh so much more again we have created these scripts for you there you have the ability to look and see and you know use those scripts that are there for you so scripts can be used to calculate the score of an indicator or break down a score to a bucket group which is another way that you can break down information um and there's a really great uh docs that actually shows you that information uh whenever you want to look at it so again you know an incident age gives you the ability to look at the start and the end and actually populate a column inside of that virtual those virtual columns for incident to actually show that and then you can report on that information all right so after you've done all of that well let me let me stop so dan tara how are we doing is there anything we should stop for nope we're good okay i think we're good all right great thank you uh so the next thing is after you've done all this after you activate it out of the box after you've identified those data artifacts that come in that you understand what the indicator indicator sources and indicators are going to do you understand what the breakdown sources and and breakdown mappings and automated breakdowns what all they're going to do the next thing that you need to do is you need to collect that data so when you load something out of the box right it's those jobs are not going to start on their own so you will have to kick those jobs off so let's say for an example that you are doing this a few months after data has actually been collected well you're going to want to run a historical job to go back and grab all of that data and bring it in and then actually report on it historical jobs are one-time things you know you don't want to do you don't want to schedule a historic job to be ran every day that's what daily jobs are for you can you know they're set on manual do not set a schedule if you can on a historical job if you need to go in there and do that if you have no you know no other way of getting the data the way that you want it then go in there manually and run that but do not set it up on a schedule daily scores collections actually give you the ability to collect that data so quite often incident again maybe you need to look at incident data you know you need it to be updated every hour or every three hours or maybe every six hours whatever that thing is well then you create a daily scores job and they have there's ones that have been created for you out of the box and then you just set a schedule up and you and you run that maybe you only need to have it run once a day we'll run that you know set it up set your schedule up go inside that data collection job you know become familiar with it there is a great docs uh page that'll help you become familiar with what you're looking at when you go inside of that but just run those jobs as you need it so again historical is when you're trying to get a big pull a big one-time pool and then daily is when you want you know information is going to change and you need to get that and that information changed as often as it uh actually changes so there are seven properties that are related to data collection uh four that are really important i want to make sure that you you understand those and we've highlighted them here is first is the maximum number of rows that are allowed to be fetched from an indicator source so we kind of set a default at 50 000 uh which we have found to be a really good number um i don't know how often it actually gets to 50 000 but i guess it's possible again if you're doing daily collections now depending on whatever you're pulling on i guess it could get to 50 000 but probably not most likely it's not going to be that so 50 000 works um but if you're doing the daily collection remember it's doing incremental pulls of that it's not doing a big pull like historical but again if you need to change this and you're an admin you can go to this and assist properties and you can change that information if you're not you'll definitely need to get an admin to do that and that's for all of these the max number of records assist ids that are stored during collection so you want to be able to store sys ids the biggest reason why is that if you want to get down to the record level so if you if you're looking at an indicator a single score indicator and you see a number of 100 and you want to be able to click on that so you can actually get to those records then we pull those sys ids in so you can get directly to that we've set that to 5000 again that is something that you can change but we found that that number really really works um the max number of elements for a breakdown so if you've created a breakdown the most that it's going to actually pull back by default is 10 000. i think this number works um dan and tara may have some some examples of when that doesn't work and i think the one that i hear the most office often if someone's created a breakdown on cis you may have thousands or millions of cis um i wouldn't necessarily recommend creating a breakdown on ci's but if you have no choice and you find that 10 000 is not enough you can change that property as well but i think anything really outside of ci's 10 000 is definitely enough um and then lastly the max number of elements produced by combining two breakdowns so the way that this works is that if you say okay i want to look at this pro this priority and these assignment groups so if you have six priorities and you have 20 assignment groups well when you do the math between that that's 120 different ways that the breakdowns can be created and that's just a small example of the way that you can do that but we set that at a million so again any one of these are seven but these i think are the four most important ones that we we get you know questions on or that perhaps are changed uh but again if you're not an admin you'll need to get an admin to to adjust these if you do need to do that yeah i i would actually just reinforce what you're saying there thomas son could you actually go back to that one real quick just to say that it's in general uh we'd say it's fine to increase these values but do it with caution so so yeah those but these values can be increased in but we always want to say is do everything you can to try to bring the indicator source or whatever value you're working with here down to a smaller number like like uh his example of the ci's and that but if you run out of all possible ways to reduce the the size of the indicator source um then it is possible to update those but don't do it unless you really know what you're doing yeah great point dan thank you for that definitely okay so real quickly i kind of want to talk about historic collections again because i want to make sure that you understand you know the way that it works so historic scores collection on an instant it's current values only right so if the the daily collections are actually are able to to take that little quick snapshot of when you've done it and you're able to get that really good training analysis on that so think about it like this so if you have a an incident that you know you have pulled data on and um maybe it's had some different some different priorities that have changed throughout that and you've been able to see that if you were just to decide to run a historic job well it may be the phone of today right and that's just from a value perspective it's just a visual to help you understand but if you do a historic then maybe it would actually take you take you back because it's just going to be of what is today it's sort of like reporting inside of the platform right so reporting is real time it's looking at the table right now the whatever is in a field it's looking at right now pa is a little bit different right pa is only as good as the last time that you've actually ran the data collection so just know that when you do that it's going to change values for the fields that have been changed for an example like last update date if you're looking at that and it closes it's going to change that to whatever the close is a reassignment count these are you know whatever that count is that that's exactly what it is and stayed in the same group so just be real leery when you decide to you know use a historic uh collection that it's going to write over you know anything if you use that if you have that indicator and it's in a historic job then it's going to get whatever that latest is and you know just be real leery on it that's why we recommend it's a one-time or if something changes dramatically that you need to do in a store then just do that manually so um yeah so there are some limitations of running historic collections so it's not possible to retrieve accurate score scores on open records right open records change information within that record you expect to change so an historic job is is you know running something a historic job on open records really doesn't uh make any sense also for indicators that are based on conditions of fields which the values will change in a life cycle of the open record you know last update day uh date reassignment account breach doing this to run a historic job and and and that information being relevant it's just really not doesn't make sense uh for indicator breakdown combinations and have breakdowns on fields for which the value will change during the life cycle of the open record again if you're if you have a breakdown that's pointing to a rather than like a choice table if you've created one that's on a table and you don't expect that data uh to change but you're trying to do it and it's on an open record you're just going to run into confusion and issues so just know there are some limitations for historic uh job collections so do we have anything we need to talk about anything that came in dan tara talk about data collection i think we're good i think we're good yeah okay all right so after you've done all of that right you've you've activated some out of the box you've looked at it you've made yourself aware of it you've decided to run the data collection so you can actually get some data in it so you can see if it's pulling here's where you're going to want to check right you want you're going to want to true it up to make sure that it's giving you what your expectations are so there's some out of the box dashboards that are out there that can show you um information that you know based on whatever about content pack that you've done check those dashboards look at those widgets make sure that they're telling you the information that you expect look at indicators directly in the analytics hub or kpi details the next experience look at those directly in there and make sure that it's giving you the information that you expect and if it's not then you're going to want to maybe go back to the data collection and look in the job log to see was there a warning or was there an error or something like that maybe it didn't retrieve all of the data the biggest thing is just look look at the information that's been provided for you and make sure that it's telling you what your expectation um is you have to check you have to look and that also is going to give you the ability to become familiar with what the widgets that are on a dashboard in the dashboard themselves that way if you get asked questions for them you know what information is being gathered there because you were part of that whole process so make sure that you check it's very important to look at all the information that has been collected for you so that was a really you know a down and dirty and in our mind those are the things that you know if you familiarize yourself with those it's gonna help you get yourself off and running npa does not include everything there's a a wide abundance of things that pa can do for you um but make sure that you kind of start there and familiarize yourself with those we have some really great resources that are out there self-paced learning that can help you become even more familiar with that um i would tell you that you know getting started there's performance analytics introduction it's like a five-minute video this uh these links again will be available for you and then also let me get that off of there all right and also um there's getting started resources and step by step so in our community if you've been to our community which i think most people have if you've been in our community you'll see there's getting started hit the ground running uh there's def there's links within that can help you so much the other thing is we provide some courses i would tell you before you do any course do the performance analytics essentials do that course it will it's not just for pa admins it's for all users it just familiarizes you with pa so i would recommend you do that don't i don't think that you should go into now learning and say the first thing that i want to do is perform analytics performance analytics advanced there's going to be some things that that you're going to miss by going directly to advanced because it's kind of a work up process and i'll let dan and tara speak that more into them in a minute they may have some thoughts on that but i would think that you would you would start with essentials and then you know look at some of the hands-on exercises that are available start with the fundamentals the uh performance analytics advance is now on demand as well so that'll help you out uh but you know start somewhere like that to become uh familiar there's some shorter courses that are out there where you can look directly at a particular thing that's inside of performance analytics kpi composer responsive dashboards widgets you know formerly indicators there are some shorter courses that that you can look at and then again like i said you know the the community that we have i think is very powerful and the other thing i think not to forget about is these right our platform analytics academies that we have we um we try to do some things that are the latest greatest but we also get questions inside of the community that are more in depth and we said ourselves you know what that would be a really great academy so we have those as well and you can get to those inside of the community and then you can look for previous ones what's coming up in the future there's a lot that you can utilize if you have some time prior to you even going to performance analytics and service now i think it would behoove you to take advantage of all this stuff that's out there so that way when you're ready to go and your organization's like okay we need to start reporting on stuff you're well advanced and you you're ready to go on all the things that are uh out there so we have some useful links uh all of these will be this will be a part of the community event as well that you can go directly and these are all referencing things that i've talked about inside of this uh presentation and you know they'll be there uh available for you so with that uh let me first ask refer to tara and dan because i by no means know it all so if there's anything that they'd like to add from a getting started perspective i'll i'll turn it over to both of them because i'm sure they have something or if not that's okay we always have something to say right yeah no this was really great and this is a really good way to kind of get the ball rolling as far as getting started goes um in addition to taking some of those training sessions and it is helpful to make sure that you have more than one individual that understands how your data is structured what your business processes are before you start diving into how to use the tool is it because it can be it can be um hard to stop yourself from wanting to just jump in and start building things without understanding what you need to build and why so it's good to have a group folks in the room who understand what your business processes are how your data is structured what your goals are and what you're trying to achieve based off of um uh building out any and before you build out any dashboards kind of have those conversations first kind of look at the out-of-the-box content see if some of that stuff you can reverse engineer and play with that kind of really helps you um after you've taken the training to say that first because it's not um it is a tool that requires you understanding how your data is structured a little bit understanding of how the platform works how to navigate throughout the platform that sort of stuff so it is important to take the training first and foremost and then next definitely um make sure you have the right people digging through and looking at this stuff because if you don't have an understanding of your workflows and what business processes you need to improve um what areas you need what uh where you're inefficient and how you how to surface up those inefficiencies if you don't understand that um you might end up with a dashboard with a bunch of numbers on there that are not useful and won't get you moving in the right direction so before you start going out there and building a bunch of stuff make sure you have those folks in the room and are and they're talking to each other great thank you tara dan yeah i would yeah i don't have a whole lot to i mainly would just reinforce the whole idea that the pa essentials um the training that's short it's it's very easy to follow it is probably the most critical piece and that's not just for people who are going to be building within performance analytics but even for like let's say business analysts who are there who who their main purpose is to kind of help determine what the kpis should be on a pa dashboard or things like so not just the hands-on people but the people who also are that feeding into the data that's going to go on to you know they have that have those insights i would recommend i recommend pa essentials for any of those people awesome great i do have a question i want to answer live especially because i have uh you all here anna contreras asked um uh as far as i'm aware of off for regular reporting the data is not stored just the last login would that be would that historical data be available somehow with this module so how that works anna and and anyone feel free to chime in is we don't we do go back and look at what what data met the criteria of your inde of your query so your indicator source like thomas was explaining earlier we do go back and look at the time three months ago two months ago 30 days ago whatever what data met that criteria and we do pull the cis id of every record that met that criteria now it's not historical data in the sense that it won't show you what it was at that point in time today but it tells you that it met that criteria at that time so for example if you're looking for something that was in an assignment group called um you know first first look or something like that and that assignment group you want to know um how many incidents that assignment group had 30 days ago we will tell you what incidents met that criteria 30 days ago from a historical perspective but it will pull the sis ids of the records as they are today so it might be a little bit jarring when you look at the actual records because it'll show you what they are today so half of those could be closed you would hope if they were open 30 days ago um or something like that um i don't know if that makes sense to you but it is it is a way of pulling historical data in the aspect of you do get a get to look at how and what met that criteria at that point in time so that's where it becomes pretty powerful because what you're really interested in is that trend and whether or not um you can affect that trend if it's positive or negative and um where where you can jump in and and make some fixes to your workflow or or anything um you know is is broken within your process you can jump in and kind of pinpoint where where that process is broken and that's where the power of that comes in is is that we have that ability to do the trend based off of what our records look like historically um i will say as a kind of pitch towards the future and we are changing the way our um underlying data is structured and it's not scheduled to be out for a couple releases still but that will have a more accurate look at what the re records looked at historic what they look like historically at that point in time so it won't be that that um confusing but that like i said that's not for a couple releases so it is important that you understand when you're pulling historical data that you look that you understand the the difference between what was pulled and what you're looking at today i hope i hope that answers your question great thanks tara there was a few things that i i've thought about just listening to yourself and dan that um i probably should have put in this presentation but i think we can talk about it and and then there was also a question i'm going to answer live as well so the first thing is i think um i think people need to be very aware of their scheduled jobs their scheduled data collections and uh being a former customer i i i remember this being something that sort of um we thought about after the fact and i think if you can do it early on i think it will help you tremendously and and that's just be aware of when things are scheduled to run scheduled to kick off so for an example if you have you know let's just take that you all the out of the box data collection jobs that come out of the box when you activate something and let's just say that there were 20 of those right and you go in there and you set schedules for all of those and you decide yeah i'm good with you know um i don't know 9 a.m you know and you said all of them to run at 9am um is there going to be an impact to the platform probably not right but i think it's still important for you to to offset those jobs right because you know it's all about how many things how many things are happening on a particular node at a particular time because data collection jobs as it relates to pa are the not not the only things that um you know jobs that are that are running inside of your platform so i think if you can help and and really all you have to do is offset it by a minute because one will start and then another one will start a minute later and generally will attach itself to a different note so just be aware of when you um you start scheduling these things to make sure that they're not just being scheduled on top and if other people are allowed to create data collection jobs um make sure that you put some governance around that so they're you're aware that somebody else created one and maybe you give them a time of when you when they can actually schedule that to be ran just something i think that's um important and it just keeps everything running smoothly the other thing is too is and there's actually a blog that i created wow a couple years ago but when you start creating things if you have to go beyond what's out of the box pay attention to the naming of something right uh put a description in something where you can because if you're able to create something that someone else can use as well rather than also creating something you know by having a really good name and a description helps to alleviate just recreating the wheels something that's there's no sense in there being two jobs that are running the same thing there's no sense in there being two indicators that are doing the same thing two indicator sources that are doing the same two breakdown it just doesn't make any sense for it to be doing the same exact thing so just be aware of what you name and put descriptions where you can um one of the questions came in says will there be a demo of what pa looks like not today uh the intended today was not to really look at any any type of demos however if you go to um the performance analytics academies there's probably a previous uh or two or three actually of showing actually what uh performance analytics looks like i will make sure that i put uh some links inside of the community event for you to actually you know see some things that are out there if you if you've never seen it or whatnot but uh i'll make sure i put some stuff there so um i believe that was the two that i was thinking of yeah dantera anything else no i was looking at the i was looking at the questions too and i was also reading josh's comments i think i think we got everyone answered does anyone else have anything else use the community as much as you possibly can listen we know a lot but there are some customers that know a great deal as well you can and and they can be there for you so quite often questions that are in the community are actually answered by other customers so if you have a question that you need to put in the community i would tell you to search first right search and see and absolutely make sure that it's not already answered there i was just going to reinforce what tara said earlier where one of the key things of performance analytics when you're putting this out there for customers is to make sure like she said it's super important that you figure out what you're monitoring first um that's one of the things that makes a it can be challenging to do a successful pa deployment because it's this weird situation where you can't really build it until after the processes are up and this is in the systems live so that you have the things that you can report on but you also have to think ahead of time about those you know those levers right you know those those things that you want to break down on because in order to break down in a way that's useful to your process you may end up having to add fields which is then difficult to do after the fact so it can be tough to plan it because you've got to think about what do i want to measure but i still can't set it up until after everything is live um yeah one of one of the things that um that john ashby used to say he had that concept called a good old boy metric you know you can put scores up there it's not hard to get a score but if it's not telling you anything then it's just like bragging rights it's it's like a wow did you see how many incidents we closed not like wow that number you know if it's not literally driving a business decision then it's not useful to like maintain and measure right great thank you i apologize again john not a problem hey yeah josh is amazing as well so but there are customers out there that are amazing as well so um do we have any other questions sorry i muted myself now um i know there's one i threw out there i don't know if raja if you're going to want to ask your question or not i had mentioned that okay i had mentioned that if time allowed we would let him ask his question but it looks like he drops but anna juan luke sounds like you guys are kind of relatively new to this so um we'll kind of oh you do yeah i don't so okay well okay don't worry let's go ahead just go ahead sorry um no no i was just gonna say maybe we talk we talk about doing like a next phase of this and maybe showing a little bit of demos and talk a little bit about about those best practices that josh and i talked about making sure you have a responsive business goal knowing what data you're going to put together that sort of thing that might be uh kind of fun awesome all right well thanks for everything we'll go ahead and close this up we'll probably get be able to get back a couple minutes here so as usual just kind of touch base on these things we we talk about these a lot make sure you're using the community there's great information that is out there so make sure that you're you're using that again these will be out uh the recording of this will be out a little bit later in the same event and also to be out on our youtube page so look for that later on today and then also look for anything that's out there that you want to check out um there's a lot of information that we have a lot of sessions that we have done uh some that are uh very advanced in some that are not so uh hopefully there's something out there for um all of you so i can see raj yeah yeah i just said enough that raja if you want to if you are you can unmute yourself if you're still out there and if you want to ask a question if not that's fine too but i just wanted to make sure you had the opportunity yeah i think if you're referring to the question about breakdowns i was just curious what we would do if you're more than 10 000 ci's and if there's a way to like for example when you do a group buying reports it would show the top categories and then look at the remaining into others right that was the question i was asking i think the first question you want to ask yourself is how valuable is it to have 10 000 breakdowns um right um there might be a way for you to um i saw the way i saw how dan answered your question and he isn't wrong like if it you know you can have them fall into an unmatched category or whatever but um if you're breaking down like that you could have a performance impact as well as as trying to understand and what value you're going to get from that if you're looking for the top end of something um you might want to just filter your criteria a bit more so that you don't have such a humongous list of breakdowns so if we want to do this breakdowns by so one of the challenge we have is we have multiple services right and those multiple services have multiple ci's so so while you're right while all the 10 000 ci's may not mean much when you look at holistically but there could be a particular service who could be interested in 10 of those 10 000 so when we design such things we're curious what to do could you break down by service rather than by ci first yeah yeah yeah so we'll break down by service and then within each service you have ten tens of thousands ten thousand or so cis let's say they're like 500 of importance right mm-hmm uh but sometimes that list needs to be dynamic what those are of important right because yeah you want the top the the top categories of those ci's not necessarily all 500 so we're just curious how we would design such jobs i'm going to ask raja really quick my team um could could you use spotlight to meet that joshua or dan or thomas what do you think about that possibly i'm not really sure um because i know spotlight have you played with spotlight russia not really i mean for incident we saw the demo but we didn't implement it ourselves so yeah so this gives you a way to like stack like weight uh what's important um in a list of records so that everything that's more important floats to the top and that can be run dynamically so you put in all your different criteria what's important like if it's a you know customer or um you know a percentage of x whatever your priorities are um and then it it allows those that are the most important to float to the top and that's just kind of using reporting um you do have to have a pa license to to initialize spotlight but that might help um that's kind of a neat idea my only concern would be that like depending on the customer cis can get into the millions yeah i know that's right that's why i was trying to think how do i break how do you break that down list so that it's in consumable chunks no exactly it's you'll want to be very careful to make sure you're limiting it to like just what you're interested in yeah not not all cis but like just servers just windows servers yeah something like servers in this data center that like anything to get that number into a more you know into a smaller chunk because it's going to do a calculation and it does calculations super fast you know but like you you know a hundred thousand maybe even a million okay probably but if you know i've seen customers that get up into like 12 20 million and it's like you would not want to turn spotlight loose on that probably yeah well i was trying to figure out like in in when it comes to ci's um and the ability to only break down on two levels at this point in time um how do you work around that so that you so one of the things because it actually was a uh it goes along with what we were talking about um a bit earlier about like being able to sub like if you start to get too much data for pa um you can split those things out and that actually gives you a third breakdown because you're you're basically as long as the split is the breakdown like um kind of like with incidents if you had too many incidents then you could say okay well instead of giving it a number of open incidents indicator our number new incidence indicator show me number of new incidents that are high priority number that are critical and then you add it together afterwards so that basically turns one of the breakdowns into the indicator right so you're basically adding that third layer of breakdown already so if you did that like what that might look like with cis is if you broke it down like already ci's is like basically kind of lumped together so if you had an indicator that was the number of computer ci's number of server ci's you know where you're differentiating like like servicenow does where it's like a desktop versus a server then that gives you that other level like the class breakdown in addition to the other two does that kind of like include the breakdown yeah is that does that help you russia at all yeah yeah yep yep so i think today what's happening is the group is taking too long yeah yeah we're just trying to see oh yeah they sure now when we do the group buys if you could do a report but do it where it's more like but instead of doing like a list view that you have a group by instead do it like a um like a a column chart or even a pie chart where you get the the yeah yeah yeah and you you can get it those will run a lot faster and then you can drill down into them but yeah okay so yeah so you know what i'm talking about then yeah yeah the other thing i was going to mention is um i don't know if it fits your exact scenario and i don't because i don't know enough about the specifics but um a newer product digital portfolio management is something that's part of our itsm product that actually gives you some ability to to like do a lot of the um reporting or you know analyzing your services by being able to get from services to the ci's within the services to the uh you know to to a you can really get some you can really get some some fairly detailed breakdown so i would also maybe suggest you uh you take a look at that as well digital portfolio management okay good suggestion i do see before we run out of time jackie has a question hi can you go here thanks okay all right thanks thanks um so uh i might have i might already have been using pa i just don't know what it looks like um like those widgets and dashboards is that pa or i'm thinking of something else well widgets are just visual ways to visualize data so you can have reporting widgets on a dashboard canvas um if you're using a pa you would had to have created an indicator source uh like thomas went through you would have to have created an indicator that sits on top of that breakdown i mean unless someone else has created it and you just inherit it which is you know a possibility okay i i probably didn't i don't have pa then so just one thing i'm trying to achieve um i work in the infrastructure space at my company so we track ci server ci's um hardware uh server racks all those network gear all those type of things right so we recently turned on servicenow discovery and so we went from 160 000 ci's in the world into the million and it might just exponentially grow which is a good thing we're discovering more stuff it can get out of control so what i'm trying to do is i'm trying to track a subset of ci's in my space and see month-to-month how where that count is going so i'm just do a count by cis id or count by name and see if we're trending up or down and see what the deltas are between the months now can pa do that yes something like that cool because the way i'm doing it right now but to what josh was talking about earlier um you want to really make sure that you structure your queries so that you don't have performance issues chugging through all that data so make sure you're like you're specific on what you're looking for okay so there's a lot to learn so i'll tell you how i'm doing it currently i'm manually extracting from a report every first of every month from five different tables and i'm just combining them and then doing power bi that's how i'm doing it today okay so yeah you could do that for me directly from servicenow so i could skip all the manual data etl part yeah it definitely can help you with that like like i said just make sure you structure properly i would highly recommend that you take those training classes that they pointed out earlier today so that you understand um the way it's structured because there is it can be a little bit daunting for some folks especially if you're used to using power bi or tableau it's not meant to do that at all um so the learning curve there's a little bit of a learning curve once you get it though you get it and you're off and running um so i definitely recommend that and then um in this situation you aren't the first person to cross this bridge so i would just put it out in the community i guarantee you that there's uh about 50 answers to that question alone yep and like like tara said there are things there you know there are going to be things that power bi can do that that we don't do in performance analytics that we don't want to do in performance analytics but that being said for like 98 because you fall into the trap of when you're a carpenter everything looks like it needs a hammer right you know it's like i know how to do this in tableau i know how to do this in power bi so it's easy for me to do that rather than learn a new tool but 98 of what you typically or what customers typically try to bring out into power bi and tableau from servicenow is stuff that pa can do for you you know essentially and the nice part is pa automates a lot of that it's it's up to the moment it's real-time data it's stick i set it up once and then i can come back and check and i'm not having to like do those data exports daily weekly monthly to get those numbers out and then when i do find the numbers i can click into them and go straight to the records so it's like there's yeah if you're doing like real deep data analysis then you're still going to want to bring that out probably into power bi but for like just general you know for most of the general reporting needs pa will do it and it's going to make your life easier one of the things i'll add for the folks that are still on here as far as the you know we're touching on pa versus power bi tableau all that stuff out there uh pa was created to be a performance management solution so you bought servicenow you're automating workflows you're looking at process automation uh where where you can make improvements in those workflows um whether you're you know making things 100 automated 50 automated there's still automation in play um servicenow uh sorry performance analytics is meant to help make that more successful so we definitely aren't in the market to compete with power bis and and tableau like josh said there are going to be use cases for that but this is actually to surface up how am i trending how am i performing where am i inefficient and how can i improve on or are there areas for further automation where i may have missed in the past because i didn't have this data in front of me and so that's really what you want to think about from its analytics as is is improving the investment that you made with servicenow and making sure that um you you figure out like you know if you have teams or groups that are not performing um you can actually highlight that with uh performance analytics and bring that to the surface so you know exactly where you need to go and and make those uh enhancements or changes okay great great so um yeah so real quick we're kind of actually over the hour so i'll just kind of hit some things quickly um maybe if that's the work okay so we talked about that we talked about that uh make sure that you're utilizing now learning for a lot of the stuff that we talked about today and even more just you can go in there and search for analytics and it'll tell you some great things um like i said before the pa advanced is now on demand so check that out if you need that uh some other academy sessions you might want to check out is the creator workflows and platform academy mobile academy and virtual agent that's just a few of them there are a lot of them out there so just check them out you can find them all in the community and in two weeks we're actually going to get the why behind the kpi with in platform process mining so prop process optimization is something that was new in san diego and um a great speaker speaker's gonna come on and do that dan grady he's really awesome so that'll be in two weeks and with that i think we're good thank you a very interactive session today uh happy to have dan tara and josh great to talk to all of you and we'll see you see everyone in two weeks thank you bye everyone have a great day