okay let's get started uh once again welcome everyone uh this is a webinar for getting started with performance analytics and the focus is how to use the out of the box dashboards as i mentioned before using the out-of-the-box dashboard really gives you kickstart those out of the box dashboards the content the content packs they are developed by the application team that also developed the application so let's think about knowledge management there's an application team for that they also develop a dashboard for knowledge management so that is really tailored to well the way knowledge management comes out of the box so the dashboards are really fit for that um so how does that work i will tell you and show you uh also in an instance how that works and the things that you have to do but before we get there let's first have a set of housekeeping rules here uh please say i'm mute i'm actually not sure if you can unmute yourself but there will be some q a after presentation if you have questions at any time just put them in the qa panel or in the chat panel and i'll try to address them later this session is recorded as you probably know and that will be shared afterwards on the community um and after the session ends there will be a short survey uh i would really appreciate that if you would fill it out and to get your feedback to make these types of webinar better next time so just a short introduction of myself my name is peter horos i'm a product manager for it service management and devops i think since end of last year the devops moved into or under the umbrella of i.t service management so for you who have a ibm pro license devops is now part of that you're entitled to it but that's not the topic for today the topic for today is performance analytics um if you feel that you want to contact me after this meeting you can also send me an email peter.goris servicenow.com and although there is that i in there you still pronounce it in dutch as peter so yes i revealed that i'm dutch i'm based in amsterdam so we make it a 7 p.m right now but i'm happy to explain to you in an hour how do you get started with performance analytics um not joining today is paul denis roy my colleague he's also a product manager for itsm he will be joining though next week when we have a roundtable about how to get started with performance analytics i hope you will join there as well this is the webinar and that means it's a lot of talk and presentation and showing from my site um and i will answer some questions but next week is really the round table where we hopefully have an open discussion and where you can also discuss and uh with your peers on the struggles that you might have or or the value that you found in performance analytics okay let's get started here first we have three poll questions uh the first one is have you already implemented pa and so um there are three answers possible here i have not started yet but i'm interested in how best to get started i'm in the process of rolling out pa within my organization and the third is pa has been rolled out and it's actively used within my organization now i hope that jamal you can kick off this poll because i'm not sure i can oh here we go thank you and just give it a bit of time uh see that we have like 34 attendees and about half of you have answered just give it a couple more seconds that's nice i i can see some preliminary results and yeah so let's end the poll and maybe you can share the results as well so now actually i can share the results as well so here we go uh so 26 of you gave an answer um most of you have of you uh are have not started yet with performance analytics um but 30 is in the process of ruling it out and 23 you have already rolled out pa and you're actively using it already in your organization that helps me also to pitch my story a bit thank you for your feedback um let me go to um the deck getting started with performance analytics and especially using the out of the box dashboards the content packs is actually four steps that you have to do uh first of all is to figure out what you're actually entitled to and activate those dashboards that along with that entitlement and what you want to do is review the out of the box data artifacts that come with the content pack and and that's really the building blocks for your indicators and and the scores that you're going to collect in your dashboards that you're going to share with your constituents and they should be like 90 95 100 right because once again they are developed by the application teams uh that actually built the application for which that content pack is is created but it could be that you have done some slide [Music] configuration on the out of the box application and if that's so then you probably also have to do a bit of modification on the data artifacts once you have done that review you're going to collect scores and there are actually two actions that you have to take there uh one is the historic score collection you only run once and the other one is the the daily score collection uh in the out of the box content package primarily daily score collections and those are the collections that you're going to do moving forward so you just schedule the job that is already available you simply have to activate it and then finally is check results that show up in the dashboards in the widgets in the analytics hub arduino is okay and you probably want to do that check together with the subject matter expert that knows about the application for which you activated the uh the dashboards and content pack okay let's go through the especially uh step uh one two three and four you probably won't have time for that but let's let's get started so if you think about activating out of the box dashboards first step that you do is try to identify your entitlement so as an admin you have to be an admin you can navigate to subscription management in your instance and it will show not only performance analytics but all the subscriptions that you're entitled to and with that you will see also plugin names and the plugins are actually the content packs so those are the ones that you want to activate now for a given entitlement like apm you could have multiple plugins you don't have to install all of them multiple content packs you can pick and choose which ones you want to activate and and that's the next step actually uh so you know um your entitlement you know the plugin names uh and then actually activate those plugins uh most of them are plugins some are store apps but let me share an instance here with you and move to that instance and it is this one so we also have this uh page here about the available analytics [Music] and reporting solutions it's on the docs and it's actually over here it's available for all the uh supported family releases i'm looking at the ones for san diego and here you can see all the different content packs so we have some generic ones that includes something like configuration management for knowledge management we have solutions specific for apm and as you see there are within apm application portfolio management there are a couple of plugins and content packs available it's not only apm like what you have with customer service solutions it could be about advanced work assignment or change management or something about communities and so on and so on and we have that here listed for all these different entitlements um and if you it depends a bit on who has created the content pack but here is an example um if you would go and check out the knowledgement one if you click on it it takes you to this specific page for knowledge management in this case it shows you some generic information about analytics and reporting solutions in general something about domain separation and so on but also it has here a link to the actual dashboard itself and here we see in the docs um just a couple of screenshots about this dashboard there are three tabs one is about content usage this is what it's going to look like so you see some time series you see some single scores and so on then the other two tabs and then below that you can actually see what was the reason the rationale to build this dashboard for whom was it built what is the goal what is it that you want to achieve do you need a specific role to to access this dashboard or not and then they listed out all the indicators that are used directly or indirectly in this dashboard so this is what you get with when you activate a content pack you get a set of indicators you get a dashboard and it's all ready configured including some roles that are needed or not now so what is it that you're actually going to do so this is gives you the full information once again it's on the dark side the visible analytics and reporting solutions and if you go to your instance and you would go to a subscription i'll show it just for a second but it doesn't show up anything here for me this is where you want to go go to all your subscriptions this is a demo instance it's not a production instance so it doesn't show anything but this is where you normally would see all your entitlements and then the next step would be if you know your plugins and the plugins that you want to activate if you go to plugins it's still this is still the old way to do it but it works pretty good actually then it will open up what is called the all applications that always takes a bit of time so you have to be a bit patient here and it shows you a list of all the plugins uh the ones that are installed the ones that are not installed the ones that you have subscribed to or not subscribed to and so on and let me be a bit patient here and then we'll get the list of as you can see uh next week is that round table shamal put in the link in the chat box so please sign up i'm really interested to hear from you and that's what that roundtable is all for that we can talk actually have a discussion with with each other um so here we are this is the plugin list and now it it moves you to all applications but if i go now and search for performance analytics for instance it gives me all the performance analytics plugins 110 of them some of them are already installed some are basic ones that you want to have anyways but if you scroll down a bit more we can see some like um the content engagement analytics this is a content pack it's an out-of-the-box dashboard for this specific application as you can see this is actually a store app so if you click install it will take you to the store and then from there you do your normal installation as you probably know a store app technically is still a plugin and here we have a content pack for advanced work assignment this is a traditional plugin so you can install it in this case with my i'm a server now employee so for me this is all available um but once again check your entitlement and understand which plugins you can activate um usually what the content pack does it will not only give you that access to dashboards indicators and so on but also it is limited to a set of tables that belong to the application you're not supposed to once you have installed one plug-in for for knowledge management you're not supposed to create indicators for um item if you have not bought that subscription okay if you have any questions once again please put them in the chat or in the q a panel i'll try to address them later in the webinar let me switch back to the deck uh shortly and that is this one so that first step activate you need and serve that admin to help you out at least the person with that role of server not happening if you don't have that rule yourself uh but now you know uh these are the plugins um and then next step is would be to review your out of the box data artifacts and what i mean with data artifacts is really the building blocks that eventually will allow you to create your dashboards and the widgets on top of the applications and the data in servicenow and when i say data artifacts we're talking about the indicator sources and indicators the breakdown source and breakdown mappings and maybe some performance analytics scripts that are part of that content pack i listed out uh separately here also additional conditions they are part of the indicators and that is something that you also want to check uh if the conditions because that's what this actually is all about these are all based on getting the queries creating to queries running the queries against your database and so you want to make sure that you have put in the right conditions for those queries so what does that look like so let's go back to the instance [Music] this one right so um if we go to performance analytics what i usually do is i just type in yt and then i'll scroll down here to performance analytics in this case i'm interested for now as a first step in the sources so i'm going to skip indicators and breakdowns and i go to sources and take a look at the indicator sources in this environment in this instance it's it's a limited set it's uh only 33 in indicator sources the more content packs you install the more you records you will get as indicator sources so it can become a bit confusing um but the ones that i want to take a look at are the ones for incident management and and you can as you can see here there are only four of them um just this little side step here you see all kinds of indicator source for benchmarks as well so if you sign up for benchmarks we will collect scores and then give that back to you and compare them how you do against your peers and that's all based on performance analytics so but let's go back to what we want to do indicator source an indicator source is from a technical point of view it's pretty straightforward it's just a set of conditions on a table as you can see here the facts table here for these four are incident now if we want to create an indicator we're not interested in all the records of in the incident table that could be a lot right but we're only interested for our indicators at least for only maybe the new instance or the ones that are currently open or the ones that are results these are the three basic ones um that we use there's also a closed one but this is also the three basic ones that you would have in general for anything that's workflow or process right it's all input throughput output and so your basic interest would always be in these subset of records in this case from the fax labeled incident so what does that look like so if i go to incident new year then i indicate source has a name and then it's based on a fact stable incident and then it has a condition opened on today and that's it and that gives me a subset of records now in this instance i can actually take a look at them but in this instance i know that i don't have anything because i only have records up till yesterday in this demo instance so let me change this but this is how it comes out of box and this is how you want it to be but let me change this to yesterday and now do a preview and then it says oh it will retrieve 160 records so you can check them if this is really straightforward just one condition not too difficult and then if you click on that link it will actually show those 160 incidents that haven't opened the date on yesterday so this is how you can check if you have set your right conditions um well in this case this is once again this is pretty straightforward now this is a indicator source so just we have the incident table apply condition get a subset of records and from this on this what we're going to do is we're going to um create indicators on top of this indicator source on this subset of records and creating an indicator means primarily put an aggregation on one of the columns so it could be a count so you're counting your cis ids for instance but it could also be um an average duration on the business duration of those incidents that were opened today the new incidents of today that's the indicator part so if you think about it in general an indicator source you set a condition it's workflow so it will very often be some um a daytime based condition this case it is yesterday so if we're going to collect moving forward from this one okay it will always collect let's get fetch that set of records of incidents opened on yesterday and for that reason it makes it fit for a daily collection if i would have and this is a bit not just a side step here if i would have something like a condition like opened on last week okay what would happen if i would collect this going forward i would see the same set of records for seven days so if i would do a count of those open incidents on last week for seven days i would get the same number and that's correct but you could ask okay how valid is this how or how useful is this to collect this on a daily basis that's where you can switch that to a weekly frequency for instance okay out of the box all those uh content packs work off the standard calendar the building java calendar simple server calendar but if you have uh created your own business calendars you can select a different business calendar and those calendars have their own frequencies for instance um but like i said out of the box standard calendar works perfectly for all your applications and all of your out-of-the-box dashboards and indicators okay if you have questions please put them in the chat or a q a panel and i'll try to address them later on like i said indicate a source a set of conditions let's take a look at another indicator source and here we have uh this is always a confusing one this is the indicator source instance open so we're interested in instance i would see a subset of records of incidents that are not resolved or closed yet so normally what you would do in if you do your report you would say something like right active is true um in this case we did it a slightly differently we got this set of daytime conditions in here and the reason why we do that is because we do this to make this fit for historic collection purposes as well if you do active is true and you do a historical action you will not get an accurate number going back like today is april 14 if i would collect that for april 1st i would not get an accurate number based on active is true now with this set of conditions i would get an accurate number so that's why we put that in and it really says like for anything that's open right now that means it has to be opened on today so it has to exist or open before today right makes sense and then it's not resolved or resolved is after today and that's the weird part but that has to do with the historic score collection that i'll explain a bit later remember that we have this uh argument in here called today it's just a placeholder for an actual date and time at the time of the score collection this got translated into real dates and and day times i'll show you when we get there so this is out of the box indicate uh sources for incident management new open close like i said on top of this indicator source we're going to create indicators and for instance here we could look at the number of open incidents so if i have an indicator or seven indicators created on top of this indicated source now we take a look at the indicator itself number of open incidents and this is a pretty straightforward one so what it does once again it has a name and a description and it takes this indicator source okay and it does it can i can i can also create a new one and say no i'm not interested in account i want to do a uh um an average of the duration of the business duration or or something like that right so we have average of duration and that would be a new indicator i can uh say insert and stay and it will create a new new record as a new indicator but what is nice about this on the same subset of records so i don't have to every time when i create an indicator on uh based on open instance i don't have to recreate that subset of records every time i don't have to create that query with that subset of any conditions because i have done it once and i'm reusing that over and over again so we know that that subset of records once we have validated that that's the correct set of records then all the indicators based on that we don't have to check if the number if the right set of records are in there because they are um and that's that's the beauty of that concept of an indicator source and what is also very good to know about an indicator in general this is will produce a number let's let me go back to the can and as you can see i don't have to specify a column when i do a count and then it's straightforward it just counts the records in in that subset of uh records that we have defined over here one of the things that that number that especially about something like number of openings that is interesting but you want to put that in a context and the first context that you can put it in is give that indicator the direction and in this case we gave it the direction minimized it means we want this number number of open incidents we want that to be as low as possible and when the lower the number gets the more happy we are in other words when that number goes down what we not only show you is a score but also the change and a change percentage based on this direction the change in a change percentage they will color green when that number goes down because that's what we wanted to do if the number goes up number of open incidents go up the the change in change percentage uh will turn red now this is one step in setting context about the score that we collect another one is about targets uh i'm not going into it but yeah it's what you expect it to be you set a target of number of open incidents i don't want it to go over a certain value um as soon as it goes over over that valley and you have a minimized direction the score itself so not to change and change percent but also the score will turn in this case will wrap if it below that target and you have a minimized direction then the score itself that number will turn green of course you can change that direction to maximize and if you are indifferent or you simply don't know you can give an indicator a direction none and of course then you won't have any color coding on scores and changing change percentage so that's that's about indicators um what i said also is what is important is and i'm not sure if it's on the next one when you check your out of the box artifacts you also want to check additional conditions and i should have shown you in the previous one indicate there is no additional condition but in this one there is as you can see the indicator source is still incident open so it's retrieving all the incidents that are open right now but i'm for this indicator i'm not interested in all of them i only want the ones that i have not been updated in the last 30 days so i have an additional condition here and says updated date relative on or before 30 days ago and that's those additional conditions that you also want to check for the out of the box stuff once again as i said in the beginning i would expect that 95 100 is okay but if you have made some adjustments in the configuration of the application as it comes out of box then it's good to check this okay indicator i said indicator source indicator additional conditions next step check your breakdown sources and breakdowns so let me go now to the breakdown source once again it's under sources so written on sources like what i said we have something like an indicator for incident management number of open instance gives you a great number but now i want to see a bit more details i want to add a dimension to it i want to see it not only that overall number but i also want to see it per category per per assignment group per service you name it those types of dimensions or maybe filters or what we call breakdowns first what we need is that definition of that breakdown itself the breakdown source and as the name suggests and what you already know from indicator source it's just a set of conditions on top of the table so if we take a look at for instance the uh the groups here the assignment groups it goes to the facts table assist user group that's where all your assignment groups are stored and what we say is the condition is active equals true now maybe you have some additional conditions that you want to add to it because i could imagine that in your group table there is more than just the assignment groups that you use for incident management or service management there could be other groups as well maybe you want to exclude them for that indicator of number of open incidents if so add that condition to it okay so we have here the breakdown source once again straightforward it goes to a fact table puts a condition to it so it limits the set of records that you get from this facts table let me go back here a second um what is oh tends to be a bit confusing as well is this one the um the breakdown sources on choice lists but you have to know and it looks complicated again but the point is the choice list the choices like impact priority urgency category all the there are all stored in one big table called cis choice but you're not once again you're not interested in all of those records you only want to have a subset of records in this case just for the impact and so you have to know how that it works so this is a good example if there is a breakdown in there that is not out of the box but you still want to use it take this one adjust it make this not impact but something else that element setting and you have your breakdown source defined and do an in certain state and you have an additional new breakdown source like with indicator sources the breakdown source is only the first tab with any case what we do is now we're going to determine which aggregation we put on that subset of records for breakdown source what we want to do with that is a bit different what we want to do is have okay we have a number number of open instance incidents and we want to break it down view that by or group by by impact or assignment group so we have incident table which we have our indicator we have this is choice table which we have defined our breakdown and now what we have to do is make sure that we relate the two tables to each other right so from the breakdown source we first go to the breakdown let me go here we give that a name okay and we say it's based on the incident impact and now we're gonna create a mapping so we're gonna relate that choiceless table or that subset of records from the choice list table to that incident table and in an incident you will see that the instant table has a lot of columns and one of those columns is impact so this is where you make that relation and it looks pretty straightforward um so you have your instant table you have your field you just select it from this uh box if for whatever reason you do use impact but you created a user-defined field for it okay then the out-of-the-box will not work for you so you have to then go scroll down and then here you will probably find your user defined field and relate that or change this breakdown mapping to that field to that column that you are using in your instance but once again in general if you did not make those types of changes in your applications the out out-of-the-box content packs for okay that's breakdowns sources breakdowns and finally briefly touch upon scripts we don't use a lot of scripts within performance analytics certainly not within the out of the box content packs because uh scripts uh are very expensive on the execution on the collection during collection but i just want to show you because people tend to be a bit scared about scripts you don't have to be the ones that come out of the box i'm i'm not technical at all but i can read this it's it's not that difficult here we have something like instant resolve time in hours and this is really where you want suppose you have an indicator number of uh closed incidents or results instance and you want to break them down by some buckets of time that it took to close them so less than one day or between one or three days and more than three days and what you need to do then is okay how long have they been open for um and then that is would be what is the difference between the result date and the open date so you just want to subtract the two dates uh from each other now in servers now what happens is date times uh our date fields are stored in the database as an offset of january 1st 1970 i think and they are stored as milliseconds so if you would go to the database uh directly without a ui or whatever and and look up this open you will find a big number and it's just the number of milliseconds since january 1st 1970. so what we're going to do is we're going to distract these two numbers okay and what is the result of the destruction is yeah a a number of milliseconds again but i'm not interested in milliseconds i want to see hours so what we're going to do is that number that we get back we're going to divide it first by a thousand to get seconds then by 60 to get minutes then by 60 again to get hours and that's what we did it's based in this case two date fields opened and resolved if for whatever reason for instance you don't use open or you don't use results but you use closed app okay adjust the script like this okay when you once you do that make sure because that's a platform thing that you also change this one take out the results okay and put in closed where are you and that's it that's how you adjust the script you don't have to be able to create your own script as long as you can read it uh you can change that accordingly if you and of course if you're not feel secure get somebody in who understands a script better than you do um that's what i do usually anyways so that's the final check that you want to do on your data artifacts once again scripts are not used very often uh in the out of the box stuff i if you can avoid using scripts that would be uh really good um but sometimes there is no option and you have to use scripts okay let me go back to the tech and here we are so that is that second step review um and then first that's reiterated short uh some of the concepts that we just talked about so you have indicator indicator sources first of all you have effects table called incidence for instance what you do you apply conditions to it so you get a subset of records and that's where you put your aggregations on and an aggregation on the subset that's your indicator that is your automated indicator that is the indicator for which we collect the values okay now so we have here as an example number of incidents closely and another one is total number of reassignments of incidents closed today right it's the same set of records and those are for these two indicators we collect the values and collect the scores and then you can use them in a formal indicator something like average number of reassignments of instance close to date and see that in a trend well that is interesting information okay and this would be a formula indicator and for a formula indicator we don't collect values we calculate them we calculate them on the fly as soon as you open up a dashboard an analytics hub wherever you use it it will calculate the score and the whole trend okay now we talked about breakdown sources and breakdown mapping so we have that incident table we maybe have an indicator source uh we have something like a number of open incidents and we want to break it down by assignment group by priority see the dimensions the filtering and so on so what we do is we first have to identify where is that information stored of the breakdown so that could be in the groups table it could be in a choices table we put a set of conditions on it so we get only the subset of records that we really interested in and then we have to make that mapping so where is this assignment group table where is that reference in the incident table oh it's in the column group so if you would look in a database of course in a database it has that sys id okay if you go to a record in the ui it will still show the display value but of course the collection does not care about display value the collection cares about cis id okay so that's a breakdown mapping and then scripts the way i think about scripts is uh what we do is we add some virtual columns to our record set so we have incidents closing today but we also want to know if they are before a certain date if there is a yes no flag that we want to set um what the incident h is because that is dynamic of course that's only the h at the time of collection it could change uh um well for instance close on today it's different but uh but if it would open incidents it could change every time and it's during collection that we create these virtual columns and then that script can be used to do an aggregation so we do a can or or something like that or we can also break it down to a certain bucket group so that incident age per record that shows you a number in something like three days two hours and four minutes uh but you don't want to break it down on three days and uh four hours and two minutes and do that for each record right you want to buck it into something like less than one day between one and three days uh more than three days and that's where the bucket group comes in it goes a bit too much detail to explain that but okay just know that it's there let me go quickly through this so before we go to the next step uh first and next poll question so what value do we expect or have you experienced when starting with performance analytics is it to automate reports is it to for cost savings due to increased process efficiency or is it to find and avoid bottlenecks in workflows um right yeah i see uh i see your question it's quite a detailed question uh i'll pick it up but i'll do so offline if that's okay with you and looking at the poll questions um thank you let's close this poll we have 24 answers in here yeah let me share the results it's it's kind of mixed um automate reporting that's usually how people get started with performance analytics and it's a very good use case and it will save you a lot of time and money and and of course because the reports are in your platform always available live whenever you need them we do that automatic score collection so you have always up to date indicators and numbers there so it is a very good use case but it's not why we design performance analytics performance analytics really about improving your performance of the processes and workflows that you run in servers now so hopefully for the people that start with to automate report you can also take the next steps and and and this is where we really uh design performance analytics for it's to find and avoid bottlenecks in your workflows and get the constraints out and make life for yourself and for your customers a lot better [Music] so let's move on and of course i'm running out of time again somehow i always seem to manage this okay next step would be collect scores two collections that you want to do um when you get started with an out of the box uh content pack you want to do in historic score collection so go back in time and collect scores for the last 60 days or three months so you don't have to wait for the trend to build up in the next six weeks no you already have a trend once you get started and then you have your daily score collection so moving forward going forward you just want to schedule a job that will do collect those scores for those indicators on a regular basis usually for the out of box content pack that would be a daily score collection so what does that look like in the application so going back to the instance if we go to jobs here as part of data collector then this is a very limited set of collections just not to confuse you too much but like i said usually you have something like get the right one pa incident historic data collection and pa instant daily data collection let's take a look at the historic data collection and this is what it looks like out of box for this content pack incident management it goes back 60 days and it ends on yesterday that means what this will do if i would run it it's an historic score collection it's based off the daily score collection it will run 60 consecutive daily collections so you can change this you can even make this a fixed uh period sometimes that is needed if you want for whatever reason want to recollect something when you recollect it will overwrite any existing scores but let's go back to the right after one you can change this you can make this uh something like three months for instance and go back sometimes people want to do a recollection for the last year always ask yourself how useful that is what are you going to do with all that information from last year there's not much you can change about it especially on daily collections i would three months that that sounds like a pretty good set of uh date range to collect for a bit it's up to you you run it as a certain user that could be interesting if you have a domain separated environment uh look up in the docs what that what the implications are you want to run it in a time zone it's usually the system time zone but you can change from that i know of customers who do all the reporting for instance in utc uh but yeah that once again that's up to you and in this case this is a job that is on demand that means it's by default it's not active even if i would check this nothing happens i really have to manually check this or click this button execute now otherwise nothing happens for daily uh ones if this one is active if i would save this what would happen it would run the collection every day collecting for the last three months that's not what you want because you're collecting the same values over and over again that's not very useful so let me switch back to on demand and so on so we don't make a mistake here and now what it will do is it will collect here for these 11 indicators and as you can see it looks like for most breakdowns but for some the breakdowns have been excluded especially for the ones that are based on the indicator source instant open there are explosions uh and that's the wrong button i have to click this one oh my whatever so here we have uh the breakdown assignment group and state are excluded for this number of open instances why that is i'll explain in a second it has to do with historic score collection now um all good let's execute yeah so the question here from bruce uh yes you can run it multiple times per day but in this case what will happen uh and that's how most of these um collection jobs are set up it will end one day ago so it will collect up till yesterday so on a daily uh score collection it will collect all the values for yesterday so there's not much point on recollecting what happened yesterday multiple times per day but just be aware of that um going to the job box and i did some tests before so this is the new one actually so forget about this one but let's take a look what does it look like um scroll down a bit and here we have something like this one so here it's collecting uh don't worry about this too much but i wanted to show you is these two date time fields so it says greater than uh was april 13 7 a.m and smaller than april 14 7 a.m more or less 7 a.m and and i told you while it's collecting for yesterday and that's because the dates are stored in utc time in a database so it makes that translation because we want it remember in the settings in the uh job we said oh collect in the la time zone and so there's seven hour difference between utc and uh la times on the pacific so that's why it has this seven o'clock also remember when i showed you that indicator source we have something like the conditions like today those are really placeholders for real day during collection it will get uh replaced with actual dates and date times and let me go back and actually let's take a look here still collecting almost done it should generate the same number uh here let me also switch now to the other job i'm really running out of time here i'm sorry about that i hope you can stay on a bit longer um almost done i think another 10 minutes or so here's that out of the box daily data collection it's uh like i said uh bruce it's going back one day so it's collecting for yesterday only you can change it you can make it like zero uh but i'm not sure if you really want to do it you can but be very careful about it apart from doing uh multiple intraday collections there's also an option for indicators to turn on the real time i did not point it down but that means you go to an indicator and it shows on the indicator hub or or on a dashboard in a specific widget not the score as it was collected over yesterday but the current number of open incidents like it is right now um [Music] let me collect this one okay and let me also go back here for a second to the slide deck because i want to also talk a bit about that historic score collection um i'm going to skip this i'm going to skip this as well and i want to explain about some of the properties so we have score collections there are seven properties with that but the four most important ones are these ones there is a maximum number of roads that you or that the job will fetch from the database based on an indicator source and that is 50 000 and these values are much more like a guard reel like like uh alcue uh maybe you made a mistake think about it so i have a number of new incidents or number of close incidents or even number of open incidents and that subset of records number of openings that would be more than 50 000. that is quite a lot uh it could be but it could also be that you maybe made a mistake in your indicator source setting you've got a condition so it's retrieving much more records than you actually want so that's what this is all about um also you can store the says ids that make up that number and that is also by default limited to 5000 then we have a default value for the maximum number of elements for a breakdown like 10 000 records think of it we're talking about assignment group more than ten thousand assignment groups more than ten thousand services could be more than ten thousand cis i'm sure you have uh but the question is also once again yes we could collect for all of these but who is actually ever gonna look at those numbers 10 000 on a daily basis i don't see it happening but it's it's there um you can adjust it uh to your needs and then i did not discuss it but maybe you're aware so what we have is second level of breakdowns but does it mean you have number of open incidents you break it down by your signing group so you see per assignment group and then you can do another breakdown in inside per assignment group you can break it down by category or something or by service and there is a a limit to it says here maximum number of elements produced by combining two breakdowns that's set to 1 million if that's not good enough for you you can raise that number but be careful it also takes away performance and then about historic score collections uh what we do is we go back in time to find the number that score on uh period minus six minus seven and so on but there are some limitations because what we do is we collect from the incident table and the incident table only holds the current values so the incident tables showed me the values as they are on april 14. it does not show me the values as they were on april 1. that is an audit table but we don't query audit table right that audit table that's a big one so we have that limitation that we can only collect the values as they are right now in the incident table and that has some impact especially on fields like the last update date or the reassignment count because that reassignment count for a given record if i look at it right now it could be three but going back like to april first maybe that number was one or two but i don't know but i can only see in the incident table the current value so if you have an indicator based on something and condition or whatever on reassignment can yes you can collect that moving forward but not historically and that's what this is all about about the reassignment state assignment group usually those values those fields that change during the life cycle of a ticket of a record for that you cannot do historic score collections so that's that's now also stated here in in uh in these bullets and then the final step that you want to do is just make sure that the out of the box works as you uh hope it will work you have to check the results and what you do is just simply go to the dashboards or uh and the widgets on top of it or you can go to something like the analytics hub and this is another way of seeing that we have collected scores for these indicators you can see trends for some you don't see a trend right because we could only collect the latest score we could not go back in time to find because that's uh that update date so that does not work but you can already see some interesting things here like the average age of open instances going up while at the same time a number of open incidents is going down so we were apparently closing um not closing the older tickets that's what happened here if we take a look at number of open incidents really quick um i'm gonna sorry we've got five minutes over uh but i hope it's worth the uh the effort what you can see here is those scores that we have collected and as you can see right now i i'm based in amsterdam so it shows me now uh 8 p.m but right now that value is 274. this is a real time value the actual collected value for yesterday is 711. okay so oh wait wait a second for yesterday's actually 499. so in this example we close over 200 and what is it 25 tickets today um which is good because we want that number to be as low as possible so this is called green right what you can also see is go to the records it also collects those records it collects the cis ids that is so um if i go back in time and go here then we see our thousand open incidents on april 1st these are the incidents but you can see a lot of them have already been closed because they were closed april 4 but on april 1 they were an open incident so going back to or going to the last one this is a way if you see a couple of these records that are still open the ones that have an empty close date this is what you like to do you go in you check your numbers probably with the subject matter expert you also do that for the different breakdowns see if these numbers make sense if they are correct and so on and if you feel good about it then it's the time to share your dashboards uh with your um users so that's that part final slides here um so that's you check what you're gonna do now i mentioned that we have a round table next week set up um and we want to hear from you and this is really like a first step in this round table um when you are are there or were there any impediments that blocked you or your organization from starting with performance analytics uh was it because you were not sure how to get started uh is it too much overlap with reporting or is pa too complicated to set up um and i hope i i i see some of you answer it's too complicated i hope i demystified some parts of that in this webinar um it's just in just an hour took me to go through all those basic data artifacts i know there's a lot more detail to it uh but the basic artifacts you know i collected the scores you see results directly in your analytics of via dashboards and so on so to get started it's it's usually not that complicated um because of time let me close this poll and share the result uh so a few of you said not sure how to get started like i said i hope i demystified that a bit what is even more important and this is the best tip of the whole webinar i'm sorry you have to wait a little hour for it but if you want to know how to get started um what you need to do is this one let me share my browser again what you want to do is you want to go to the community and go to the forum and this is an easy one it's the top one analytics intelligence and reporting okay and then from here you have that first tile over here it says getting started and this really helps you walk you through all the steps that i just been talking about uh it starts off with quite a lot it's much more information in there but the basic steps are all here it has some intro videos and so on uh some are five minutes that tells you to activate to configure to analyze and then also what we have here is a whole bunch of resources you can go and click and take a look at pages and you go step by step all the things that i just showed you is explained here as well but what's really cool is this whole list of um all kinds of material and collateral available to you a lot of them are self-paced uh some are particular like an hour some are even shorter um you can have do a lot of hands-on uh labs as well there's actually much more than is listed out here if you really want to know you go to the learning portal and now learning and what you can do over here is yeah look for performance analytics and you get all kinds of collateral a lot of them are what we call on-demand there are self-paced um and really this one this is actually a very good one the essentials it's not only for your admins to pa admins but also for users so they get a good start understanding what performance analytics does and cannot do and and so on and a lot of on-demand uh courses short ones uh like for instance hey how does that work with widgets and so on very very good set of resources here so final slide here um and here we have the also that list again um i'm sorry i don't have time anymore i hope that you will join us next week for the roundtable and then we can have a more open discussion uh amongst ourselves and uh try to help each other uh moving forward with uh performance analytics i want to thank you all for attending uh once again apologies for going over the time but i hope it was still a very useful team thank you all bye-bye enjoy your easter weekend by the way