okay good afternoon welcome to platform analytics Academy we're going to go ahead and get started we are at the top of the hour so as I said good afternoon or good morning good evening wherever you may uh be in the world thank you for joining today um go through some logistical things and we'll get directly into it so as usual these are always for you we try as best we can to bring fresh ideas and better understanding and practical guidance to you as much as possible these sessions are definitely recorded and they'll be out the link to them will be out in the community later on this afternoon and of course they are out on our YouTube uh channels as well uh q a please ask as many questions you as you would like if you can if you could put those in the Q a uh part of Zoom that way uh some of the panelists that are here are able to answer questions while we're going through this uh demo and presentation please ask any questions you would like if they are unrelated to the topic that we go that we're going over we'll try as hard as we can to get to those when we are done with this uh if you have a question about something that's uh going on in the presentation and demo feel free to ask that as well if we can stop the demo to answer that question we will but like I said some of the panelists may be able to answer that um as well so uh Safe Harbor notice uh I don't know that there is anything in this but just be safe we always like to make sure that we show this to you so please um don't make any buying decisions or anything of that sort based off of the information that we share with you today um my name is Thomas Davis and I am the host and also a presenter and also have isoc with us today isoc is going to be doing some uh talking as well and some demoing of the content that we're discussing today and as usual I have my co-workers uh with me as well Dan Tara and Adam so again they'll be available to answer questions and they may of course come off of mute as well to ask and answer questions as well so um as a title of this today we're going to talk about analytics automation uh so this is something that we've been talking about for a while and it may be familiar to you or may not but uh hopefully after this you'll be able to think of some different things that you can actually do with some of the things that are inside of servicenow that are related to uh PA our agenda today is we're going to talk about the event registry um again maybe that's something that you're aware of or you're not so we're going to show you that and then we're going to actually talk about the uh pa uh events that are there and available and we'll talk about flow designer because we're going to use both the events and flow does designer in our demo and then we'll have some useful links and of course a q a so again if you have questions feel free to ask as we're going through that so as I said the first thing we don't want to talk about today is the event registry so the event registry lists events the system recognizes uh you can use these events to automate activities you can use them in scripts uh script actions or notifications they are there available and here you can see where you actually would go to access those so if you just select all and then event registry um they're under the system section of PA and you can click directly on that so there are some PA ones that are out there and um these are the ones that are there so um I don't know that we necessarily need to read through all of these uh but I did make sure that I put a description in here but this is kind of things that are there once um certain things happen inside of the instance then there's certain events that can be triggered to do um you know certain things so for example some of the ones that uh we're going to use in the use cases today are you know the the PA job DC ended with air so this is when a data collection job ends but there's an error so what can I do with that so obviously you know if you're familiar with data collection jobs you can go and you can look at the job log and you can see when something has ended uh in an error and then you can also see um you know what the error was and the same thing for warning but the question is what else could you do with that could you trigger something else to happen could you create something that would make you know something else happen based off of that particular event um ending and the answer is yes you can do that with flow designer and uh what we're going to do today is we're going to take two or three of these and actually show you where we can throw those into flow designer and create some other things that happen uh won't give it away yet because I don't want to take ezoc's thunder away but it'll actually trigger some other things to happen so again it gives you that automation right by using the tools that we have inside of the platform to do some other things so here are there are 13 here these are the PA ones and these come out of the box they're uh but you can obviously create more events and you can attach business rules to those events and and do some other things within that but these are ones that you know uh Target and threshold uh notification when a Target or threshold has been reached um or you know that the target has been completed or something like that uh and again you have kpi signals in here so if a if a new signal has been created you can do something with that breakdown so if someone or if a breakdown has been deleted what can you do with that action when that actually happens uh or an indicator has been deleted so maybe you want to make sure that you're controlling the governance around your platform so if someone has access to delete an indicator maybe you want to know about that maybe you want the notification when something like that happens uh things all of that sort so again these events are in the plan platform these are out of the box other ones can be created as well but these are the ones that we have uh you know like I said created that you can do some actions on them um so with that real quickly I just want to touch flow designer if you're not familiar with it we're not going to go in great detail of you know how how the flow designer works and things of that so we really just want you to understand that for today's demo and what we're actually going to try to do we're just going to use a flow designer and we're going to create some flows where you know you're actually going to take those same events and use them as triggers for uh other things so obviously there's some servicenow docs that are out there that talk about flow designer and here is a look of what it would look like when you went into it but the other thing is um our own atom did a uh a session a couple years ago which was really awesome called burst reporting leveraging workflows and analytics and and then there he actually did uh some really fancy things with workflows as well not necessarily with the events although he may have done some uh in that but he was just made making some things happen directly with workflows and then you know providing some analytics on that as well so um again there are some very useful things that you can do with both flow designer and also the events and and many other things using flow designer so if if workflows are not something that you have dabbled in yet to create you should definitely uh check that out if you have the access to do that so today real quickly we have three use cases that we're going to talk through and demo and what they are is when a data collection completes with warnings we'll automatically generate a task with a priority of three and assign the appropriate assignment group so again this is one of those things right so if you're someone that is responsible for looking at those data collections maybe in the past the only thing that you've ever done is literally went to the jobs and looked at those and then looked at the job logs and you know that was just something that you had to do well here we're going to show you working you can actually create a task that'll let you know that something one of those things have um had that issue whether warning or an ending and then the second one we'll take one that actually ended with an error and we'll create a task and assign a priority one of that and assign that to the appropriate assignment group as well so and then the last one whenever a new kpi signal was will be created with a PA data collection job completes we're going to actually show you where you can automate and know that that is actually happening so um I'll stop before I pass that over to ezoc although he may answer some of the questions in the demo of some of the questions that are coming coming in but I'll ask my panelists are there anything that we should answer now nope I was just posting a reminder in the chat um I I know not everybody understands we don't really catch the chat as much as we monitor the Q a so if you can remember to post your questions in the Q a that would be fabulous because that's what we kind of focus on so the chat sometimes gets overlooked if there's you know multiple conversations going on so we don't want to miss anything you have to ask absolutely thank you for that Derek uh Tara sorry so let me look at this one from Steven does in with air trigger on node failure currently that will Resort in a record with a status error and Pa job logs but not in PA job log Rose so let us let us get through this demo Steven and then we can kind of come back to that question and see what we can do with that so with that ezoc if you were ready you were uh I give you the floor and we look forward to seeing the great stuff that you're going to show great thank you very much Thomas I'll start sharing my screen good morning good afternoon good evening everyone uh well hi Stephen it's been quite some time that we talked and you are well with the example that we're going to show we do catch the errors on the page of looks we don't have them on the ropes so I'm not sure why you always go to the Rose uh to find out if there was an error because I'm the head of every collection as we will see also during the process the state of the record changes to either resulted in errors or with warnings so I'm going to demonstrate that now as Thomas mentioned earlier we have all these events that are uh created out of the box by us by our developers but you also have the opportunity to create your own events whenever you want with a simple business room I understand the topic it might be a big technical but I think please keep in mind that these can create really a lot of good use cases and problems to be solved even for the business users so please keep in mind that with the events you can sort really a lot of problems because they are cute they do not create if everything is created correctly technical problems or performances on your instance as long as of course you are being careful working through now a bit on the events so I'm going to demonstrate I'm just going to show you where you can find these events and then I will work you through uh uh how things so in the instance so that you understand where each record is so the first part is what Thomas already saw so there is the event registry you need to be an admin to reach to this uh table and these are all the events well there are really a lot of events these are all the events that start with PA so there are really a lot of them and the ones that we are really interested in as uh Thomas mentioned uh are the ones that are related to the data collection ended with warning and with error these are the two ones that we're going to talk to also about the new signal uh being created so this is the place where we just say that this is an event now what happens when that event is created is something that we provide out of the box usually what needs to be done but at the same time you can create your own flaws what do you want to happen navigate so that you can see where I'm going to so if you reach out to Cloud designer in the platform uh yeah so you get the list of all the flows that are being created by uh by us but at the same time you can create or even customize what we created I will walk you through uh one of these uh it's something simple as Thomas mentioned so we are creating just a case task when a job fails and why do we create the kids fast because the collection jobs usually run outside business hours so well you might not be there and the next day you come and you see a dashboard that doesn't have any PA data on it now how how important is that of course all business users will not be able to see their data and that causes a big problem and that's why your admin team that handles this analytics needs to be aware of this and they need to actually jump into it and work on it so here in the flow designer I will walk you through the steps of creating this simple version of it but I just I will take a bit more time with this one and then the next I will demo so the table where we create these events is this event table and well the condition is pretty simple name is and this is the name of the event that we just created so this event is generated every time you did the collection completes and it has an error in it so this is how we know that we found the actual data collection that resulted in error then the next step that we um I created for this use case is that well we want to go and find out the logs and understand which job log is related to this one right and how do we do that and for that I'm going to walk you through to the event log so that you will understand all right um actually put the wrong conditions yeah I think okay so as you can see here there are some errors and warnings so I will just open one [Music] uh so here we see the name of the event and there is the table where that event is linked to that is something that you define in your event itself this is the instance which is the csid of that records on the page or blogs table and then we have the time that is processed because the events although you can create them instantly they are not being executed on the Fly there is a queue and there uh and whenever a worker is free it picks up those uh events and it executes whatever needs to be done by it and this is the the URL of the job and this is something that you will see how we use it later on and there are these two fields which are the parameters where you can when you create an event you can add some valuable information into it and then on the right side we have the user ID the username and this is Adam that was running this job and how long it took to execute this event and this is very important you have to keep all your events executions to be really really low because if something runs more than two or three seconds all the rest of the event will work even so you have to be quite careful and this is the state of the event so if it was processed or not or it can determine error or transfer to a different node so having all this information here going back to our flow here I'm just searching in page all blogs and all I'm doing is I pick up this sysid and I put it in the in the filter and that will result in the records that actually the age of blocks that had their and moving forward then as Thomas mentioned earlier the idea here is to create a high priority task so that people in here in the analytics admin which is the assignment group that we assigned it to will be aware and they'll have to work on it urgently right that's why it's a P1 you have a whole dashboard for example you might have a dashboard that is your CIO or your CEO is using and it doesn't have it didn't get updated with fears in this data and that's why this is a quite high priority of course based on your own use cases and how you want to address this type of situations you can configure anything you want and most important thing that you need to be aware here is that I didn't write a single line of code right there is the opportunity to write but uh I didn't write anything everything was done with the mouse and just typing the names and just to show an example I will move to the table where we are having the tasks and I will just open one task so here you can see it is assigned high priority it's open scientist not there and then we know the job and here you can see for example in the third description we have all this information up to date on the number of Errors number of warnings and execution time uh that it took so this way at least the admin will be aware and then they can take steps to solve this problem now before moving to the next okay okay I think Stephen I'll take that question a bit later on uh I will continue with uh with the demo for now because I see it's quite long um okay uh that is the first part uh the second part uh for the warnings uh let me go back to maybe for the warnings is something similar and I did it on purpose similar here because I would expect someone to ask me a question to be honest so if for warnings the only thing you need to change is actually assign it to a different group here or the priority is not high priority anymore it's moderate because well it's a warning so it's not that important something happened but someone needs to investigate it but as I said this is based on how you want to handle your processes or it might be even related to the job itself you might have jobs that have indicators that are very important some less important particular case and of course now I created two simple flows here but well because flow designer is so simple you could actually do both with one so that is a simple uh if statement as you can see here and everything is created with the mouse so I will walk you through how you can create this flow so I'm going to create a new just demo flow so that we can see how everything it gets created and how simple it is so there is a tool here where you can walk through every field and so that you can understand how the flow is created so uh now the in the triggered section we need to find out when do you want this flow to be executed so what is the trigger and here as we said earlier it is a creation of a record on the CC event table and we need to add filters or the filter is that the name of the event is error or for example each one if we're going to create the the second the use case where both of the uh joke sorry or both of the different types of events are generated and that's it so now we know there are more advanced options here so if you want this to run if it's an interactive or non-interactive I will not go through all the details um there is quite a lot of information in our dirt and within our community how you can handle it but if you have questions please feel free to share so now we know when this event will be triggered and we did it and now the second question is so what do we want to do and here in the flow logic we can say is for example if it was an error and then we can say here so what does it mean it was an error that means that from the event records because we configured it in the trigger that's how this populates uh if the name of the event is an error second paste uh and that is the condition and we are done so now we have the condition and what we need to do is what will happen if it is an error and here you can actually create a record create task delete or do any other type of uh events as you can see there is a really big list of actions that you can do and you can even create your own actions if you want that is a separate the discussion and not for now uh okay so we'll create a record as we said earlier and then we'll just specify which table do you want to create the for example we want to create a new task and then you can just set values just to show you how simple it is you say the third description this button here gives you the opportunity to write scripts if you really love them and I used to love them but since I got this easiness of clicking with my mouse I stopped like the liking scripts we don't scripts are good but I think with this will make our lives much more easier for everyone and yeah then we refer to the event record and we can actually combine the text and you can see that this okay there was this event created for example and then we can pick up from here from the event record the name and then what we can do [Music] we can add in the flow logic and else so if it was an error that is if it is not an error then do something else and continue with your flow so the moment you complete I will not go through all the steps now but the moment you complete with all the steps that you want in your flow you need to save it of course you have the option to activate it instantly so from the moment that you click activate that version of the flow is running but you also have the opportunity to test it so you can see for example if I select the record of the seed error and you select one error you can say run test and you can actually see the result how what would the outcome be such as your condition here evaluated to true because I selected an event that had a error and then actually it says it will execute this for Action which is to create the task this way in general you can create any type of flows that you want for any event that is currently available in the platform but also as I said we can create your own events and create your own process and uh any type of Records either creation or updates or certain notifications or anything else you would like to do for the second part of the demo I'm going to create I'm going to execute the job and this is the job that will generate a signal let me close this and they will have to give it a few more seconds now uh I know there were a lot of sessions earlier about the there were a couple of ships about the kpi signals and what is the signal signal is a statistic analysis that we do in our processes and in our indicators which tells us if the value of the indicator is extracted to statistically expected or not so the the data collection completed um let me check the event log and we got a new signal and this new signal that was created now we can go to kpi details okay and now you can see here in kpi details you can see the notification that there is a signal generated which is this signal here uh but what I did with the flow is not only create this but I also did is this is something that comes from the data collection itself what is being created is actually for this use case it's I created a case and just to show how we can easily create any type of record it's not a user use case I just wanted to demonstrate and as you can see here for example I added in the activity a link to the kpi so when you are here you can actually open it and navigate directly to kpi details you don't need to go anywhere else but the main reason why I did this is also to show that so this case tells us that there is a new signal and here I can collaborate me as a process owner who is Abigail and I can say hey I will discuss this issue with Thomas and with Isaac and I can start with this discussion and say hey guys we have a new signal let's investigate right and then everyone can collaborate in this part here and then they can talk and find the solution and then uh solve this case now as I said the case is not the extremely best example but I wanted to demonstrate what discusses there are the sidebars so that you can actually collaborate with the rest of the colleagues in the organization so that you can troubleshoot and find out what actually happened or if nothing happens uh going back to kpi details when you do get the signal you can just dismiss it right and you can say well nothing happened it was a public holiday for example and that's why uh the values and when you click confirm kpi signal runs for the next checks for the rest of the period and then it found another outlier for this outlier we don't generate an event because it is created by a user action and you can see instantly the outcome so our case will not be created for this new signal that we just saw uh the main reason of course why we came here is just to understand the events so uh and the events in PA we try to add as much as many as we can in many different places so that you can add your own processes in which you can customize what we generate and you can actually uh you send a notification the way you wanted to send it or change or create a record or update someone or trigger anything we want in our platform so you have endless possibilities when you're using our plot designer with payments that are triggered in the platform and that was I hope the quick version of it so are there any questions see there are some comments in the chat and yeah so so the the only question at this point is the same one that that Stephen had which was the the Lincoln one so where he's talking about nodes in um events and all that stuff so I know that you had said that you wanted to um go with that one so maybe we can get that one and then we'll see if any other questions come in I think um I think we might want to take Stephen off mute for this one because it might be a conversation what do you think that might make it a bit easier yeah yeah that's what I was thinking too yeah thank you all right Stephen you should be able to talk if you're there hey folks you able to hear me yeah hello hello so first of all Isaac you're my hero so thank you for thank you for showing up this might um I might have actually this not might not be a question that really hits what you guys are demoing today I I saw that there was some new stuff around logging and got excited because we've been having some problems around logging but basically the issue is that certain types of events that lead to errors uh the most the one that I'm just neck deep in right now is node fail you'll failover you do get the error status on the PA job logs table but you don't get any verbose logging whatsoever that comes from PA job log rows because that's normally where it's like hey why did this fail and what's this this idea of the thing that failed and what's the error limit if I'm gonna go you know do process of elimination so the issue that we've been seen is that well if there's no logs that are helpful maybe it's node failover and where that's a problem is if there's no logs that are helpful maybe it's a security event right because when things start running outside of their scheduled Windows um you know all the admins squinted real hard and say you're not supposed to do that why are you running two hours late and then there's no and there's also no logs of what rank if it's a node failover there's no logs of what ran the event was this a system triggered event was this a individual user triggered event so it just gets very fuzzy very quickly so I was just wondering if the if you know since ezoc's here and he's a much bigger brain on all of this than than almost anybody I know and certainly than myself if he had some clever way of how to disambiguate the lack of verbose logs of oh you're just a node failover that's not great but it's not the end of the world it's just the system doing its job versus oh mysterious things are happening call an admin to start locking down roles right so that was kind of the broader context there and if that's way out of scope for today no problem we'll talk about it another time thanks guys thank you Stephen no it's a very interesting question uh to be honest I have in my mind a couple of ways to capture it but I'm not sure if it's going to capture all the different use cases which is one is the node transfer so for example we fail over to your backup uh side if something if we need to do some maintenance for example or your node for example run out of memory so there are many different use cases why a node will not complete the job and yes yeah that's something I have to look into towards three different use cases and maybe I can reach out separately or even present it in a next Academy session and but it's a very it's a very good point and that's something all the ad means not only PA admins even the admins of the platform would be very interested in finding out because it's not only the collection jobs it's all the jobs in the platform that will run a bit later and yeah if everything is in a tight schedule then it can cause quite a lot of problems it's it is a very good question but I promise I will come back to the student I will not try to No Worries didn't mean to hijack you today this is still good stuff thank you sir thank you thank you for the question very very good question all right thanks zizak and uh Stephen for that um don't seen any of the questions at this point but I think I would like to call on Adam because I feel like Adam this is some stuff that's near and dear to him so I'd love to have him share something uh if he has anything great is a great session I really appreciate it and I I think the thing I want to add as we go back we we don't want you to have to script or have to create flows and do things but you know your business better than we do and so that's why we want to make sure that we're exposing and talking about how you do these things um if it's a really common use case we're going to try and do it out of the box like the kpi signals the notifications and things go on but we want to make sure that we're not limiting limiting you um it'll be great to see some of the things in the community about things that you want to do or things in the idea portal I think would be great as well as I'd like this to happen but there's always going to be use cases where uh your uh further along than we are your your business is maybe a little bit more specific than some of uh some of our other customers whatever the case may be you can do it whether you're creating events listening at the events we drop um using flows and I'm recovering scripting I I I love I love servicenow on what we can do with scripting but I completely agree with esoc flows are so much better I and I'm glad that he showed in flow designer the the debugging information it is so much easier to create to debug I do create my own actions when I want to do other things but then I can easily reuse those by using flow designer and we can also be using decisions and and process automation designer just the kind of the there's no limit to what we can do foreign telligence for example you can avoid having all the assignment groups and the priorities everything can come from machine learning you can actually do really a lot of uh you can enhance your process so much that it can actually solve itself the case of course yeah it's great and so there whether it's a use case that we're not supporting it that's very specific to to what you do in your organization um or there are some other there are some use cases that are common but there's so many variations we can't build it kind of successfully with so many options is working on other things but you can you can come in and go no if there is a PA job that fails I want a task created to go to this person do I want to um when a new indicator is created or a new breakdown is added removed I want to make sure I have a really clean log of that because it's really important or and you're not just doing generic things but the events allow you to look at what happened which indicator happened which kpi signal breached so I notify or take the right action for certain indicators but not other ones just Limitless on what you want to do with with flow designer and decision Builder it's incredibly easy to do you just have to know that it's there and so I thank you for putting the session together for showing us that and as you're doing this I would love to see your questions in the community feel free to tag the izoc you'll be happy to answer your questions maybe me too but izak would love to answer um and and as you're thinking about things you go hey this seems really useful why wouldn't everybody do it make sure you're putting those in the idea portal let us know give us your feedback on how we make it better for you but the goal is that we automate everything I I don't want to see any especially for for those of us who who understand the value of in-platform analytics I don't want to be managing in platform analytics in spreadsheets and PowerPoint we can do this all in the instance and we can take advantage of the full power we have with servicenow absolutely thank you Adam great points definitely so um we do have um 20 minutes or so so if if and I see that there's still quite a few people here so here is an opportunity if if you have a question about this um please add that into the Q a but if you have a question about anything else that's you know platform analytics related here's an opportunity for you to ask away there's some really great gurus that are on the call so um maybe we'll give it a minute or so go ahead so is those as those come in I I forgot and I I don't I don't mean to steal your thunder if it comes in later but we are in 2023 which means that knowledge 2023 is is coming up and as we're talking about um flow designer and workflow Automation and analytics there's gonna be some great stuff uh some great sessions some great Labs uh covering um related things if you haven't taken a step into workflow Automation and flow designer uh there's some great sessions in the platform Academy we have the the link out there for you one of our one of our sister academies there are great session there are great courses in now learning as much as you want to learn the developer program has some great stuff in it as well and knowledge is going to have some amazing labs in there as well so if you haven't looked at flow designer um and you you're still comfortable with scripting you need to look at flow designer and how that works and 2023 is a great year to do that and to remind everyone knowledge is all in one place again this year back to the way it was before covid so we'll all be together in Under One Roof so it'll be a great time to um get together and collaborate and chat definitely and and it wasn't Thunder but you and Tara did did steal my slide that I'm going to talk about later on that's okay that's all right we went ahead and got it done and I think uh you know it is exciting to know that is it is all under uh one roof and and that's exciting of course it's in Vegas so um yeah so um I'm really surprised that that with the amount of people we have there's no no any no PA questions or anything like that here we go I must if I can sure go ahead go ahead he's up yeah uh start tearing my screen a bit uh maybe for Steven hey it's not the complete answer so there is a table in the platform where we have all the nodes that are kind of running in your instance and we have the date that we got updated and their status so if something changes here I would highly recommend you to use a flow to pick up this change and then you'll be able to understand if something happened a person requires a bit more testing than what I just did in a few seconds but I believe from here you can see when a note goes offline and another one comes online then at least you will know that a node restarted and that can explain the delay in your whole process oh my gosh that was a quick hijacked sorry about that no no it's not hijacking man it's just free form so you have an answer we that's definitely a good thing so all right so we had one came in uh VA dashboard how to build a KB breakdown it is blank maybe some context around that unless um yeah I think we could use a little bit more context on that L we talking about like uh the knowledge article being the breakdown choice or I'm not quite sure so if you could um yamping so I don't know if we can maybe bring them off mute here briefly too that's yeah I just allowed him to talk so he can he's willing to talk uh please unmute let us know what's going on oh can you hear me yes yes yeah it's it's in the uh virtuation and uh we uh were in Tokyo even in the Rome hkb breakdown which is from the oob which is black um so I don't know how like others we already uh we apply the virtuation and data is displayed in our quite well the only KV breakdown we don't know how to um how should we do it to because this arbitration dashboard is um like service now protected we even my business wants to just okay let's we do know how to do it it's just like remove it from uh from the dashboard but we won't be able to do it and we don't want to show business this is just like nowadays available so we we want to do something with it but I don't know how well it seems like a KB breakdown is um it's we do not use the um KB uh KB database as a deflection because we do not go there in our level which we just like if people do a search like AI search and then in the served and if they say okay this help this helpful and it will be um like we have a search served and the AI search that's in the deflection pattern uh dashboard yeah it's it's one of the widget which is which looks good but KB breakdown we do not track like okay this KB helped that can be helped we do know you know what KD gonna be bring up so this one it just stays he stays there have a story for that but it just stays here we don't know what to do with it in oob um it's like KB breakdown it's black I don't know if anyone has other thoughts as well but I think we might need to try to find find a virtual agent resource to get in touch with because they uh those different groups are the ones that actually build out their out of the box dashboard content so we don't always have a lot of insight without digging into so we'd have to we'd have to look at your specific scenario to try to I think to try to figure that out but um but but otherwise I think we can probably we could probably try to see if we can find a uh a virtual agent resource that we can get you in touch with well the two options I'd put in one I'd put the question in the community um screenshots um you'll probably get the answer pretty quickly if you do that um and then we have our virtual agent Academy I would ask them um I think uh I think Thomas has a slide of all the other ones but snsn.org academies um you you'll get all of them you'll find them in the community um uh their session would probably be a good one to do that and then I think we'll take a note to go we probably want to do a session on reporting on on Virtual agent might be helpful uh all around to go over what's there and do that joint session with us because I you might run into an issue where the virtual agent team doesn't get the breakdown question and obviously the performance analytics team doesn't understand every day question so we can get together and solve that but to get the quickest answer right now I would uh I posted the community um and then if you if you join one of the virtual agent academies you probably can ask that question there while we put together a better session okay thank you yeah I just searched the community for the uh KB breakdown whites um a few people asking the same question but I don't get really a solution there I see the question over there um yeah I don't know yeah yeah I think like just like Adam said with you ask a question or seeing multiple people ask that question in the community a great thing to be would to jump on one of their Academy calls and and ask that directly in the academy where they can get you an answer um yes thank you okay and Adam you're typing an answer to a question there so yeah the question came in and I'll I will take the answer to come in but also talk about it you know where do we get started um what courses do we recommend and there are no specific courses on using flows and what we're talking about is one of the reasons we do this session there's no specific course on it it's all the same flow designer is the same whether you're you're using it for itom for analytics it's all the same and I'll tell a story since we have a minute I got a question um earlier this week about how do I know who's viewing attachments how do I know who's viewing attachments I want a log of each person who's downloaded an attachment and I don't think that comes out of the box I don't I'm not aware of it of it showing me the report of who's downloaded which attachments and the idea being look if nobody's downloaded this attachment in six months we can we can Purge it um how do you do it well the way I knew how to do it was I went and looked at the event registry Thomas and I were looking at this session last week and I'm like they have a there's I wonder if they're I wonder if there's an event and I went there's an attachment.read event which is extremely similar to report.view which is an event we looked at the PA events but there's some reporting events as well about when reports are viewed um attachment is there it has the event says which attachment's viewed and so the solution I was able to provide was we do the exact same thing that we're doing with the report view uh report view event report view event is where updates report stats to let you know when reports um how many reports reviewed how the performance is if we look at report stats and widget stats it's driven off of a report view event um we have a post and I I don't know if we did an academy session on it about tracking the individual users who view reports where we did a lot of these same things where we looked at the events and then just added a new script listener or this was old old the olden days so we the script listener script listener which is script um instead I would use a flow to say when this event is is detected fire this flow and whether I'm looking at report views I'm looking at kpi signals I'm looking at attachment reads flow designer is the same everything is the same it and so as we're looking for things make sure we're looking at um what do I want to do and does it do that is there an out-of-the-box solution for it and if not is there something else in servicenow that does something similar because I can just copy that I can at least look at how they do it and I can do it the same way um so you don't need to take any special courses for this but make sure you do take the flow designer course um and now sorry now learning there's a ton of courses there's a lot of great um previous knowledge sessions there as well and a lot of just kind of individual things the developer program at developer.servicenow.com does some amazing amazing videos they have a lot of webinars that go on as well that will walk you through things so whether you want to learn flow designer which is kind of daunting or you want to learn how to do something specific I think for me I like to take an intro class you know spend an hour kind of get the basics and and shown around and then I'm just gonna search the developer site search the community for how do I do this you will find an amazing number of resources just how do I do you know this one specific thing the better grab it the easier it's going to be and then leverage the community the developer Community uh is a little is as active or more active than um the analytics one you get questions in in minutes maybe hours but there are lots of people looking at things so when you write a good question with screenshots and describe the problem you're going to get an answer very quickly and get you on your way yeah and with that I'm not going to actually I'm actually going to type all that in because apparently I talked for a long time but make sure you check out what's in now learning which has the old no previous knowledge sessions come to knowledge if you can check the developer program or the developer site lots of great content there and then we have a few um uh posts and blogs and previous sessions we have Thomas talked about it the burst reporting one's a great overview walking you through end to end on what you can do uh and then if you look at the how do I view dashboard stuff uh how do I track dashboard views and Report views similar Concepts as well now those were based on scripting not flow designer and I would definitely do it in flow designer today um but there's lots of great examples out there in the community okay great so we did we did have one come in and Adam I know you have spoke we've spoken to this before but uh we're currently evaluating the tables we use in our instance And archiving those with excess of a one million records the PA scores table is one of these that has a lot of Records obviously is there any best practice or recommendation on archiving as it relates to Performance analytics is there a point where we may account our performance because of the amount of data that we gather based on indicators Etc um one don't archive it using the archiving um the the PA scores table it should be scores tables I'm hoping we're on the PA scores L1 and L2 tables um not the the main the PA score scores table the singular one table that has a plural name PA scores uh is the the older architecture there's a new architecture uh and I say newer it's been out for several years with PA scores uh scores scores L1 and L2 um it it shrinks the size and an amazing amazing amount if you are using scores and you're putting data in scores um take a look and you can search the docs for the migration tool if you um if you want you can put a case in uh but there's an automated migration process um if you have any questions put a case in to help you do that get to the new tables um when we are using L1 and L2 and there's still a lot of Records that's still okay we don't use archiving but we do not use the pa the servicenow archiving tool because archiving is are not archiving but deletion is already built in with the data retention so every indicator you're able to set the retention um there's there are the global retention policy please correct me if I'm making this up by the way if anybody if I have this wrong but they're the global retention policies which are um I think two years of the normal retention policy um but to your for daily scores I think we keep it for two years and five years for for monthly you can change those for your settings and then per indicator or indicator Source I believe you can even override those to make them longer or shorter you know definitely shorter that's how you want to do it uh because that way it keeps it nice and clean uh to come through and having data from seven years ago isn't super actionable we want to be actionable um but don't use I would say don't use archiving The the service network I've been using just the cleaning that comes through because you can't restore like you you would before there's some special things about PA scores that being said space can be an issue at some point performance is not performance is really not an issue every when you the only way you're accessing the these scores is through the PA API whether you know it or not you're using a PA API we're not using live record um as normal and it is highly highly indexed ezoc actually uh who's on here spent a lot of time um and a lot of of thought making sure that the way we get the data is always optimized it's amazingly fast there's no out-of-the-box use case where you are scanning all of those records and if you were scanning 100 million records that takes some time no matter how fast you do it um PA scores are always Prime primary key lookups um do you want to tell me all the things I said wrong no no you're absolutely correct I think one thing I can not only do symptoms question is I I read that you have 81 million records it doesn't matter how many you have the question is how long do you want to see your analytics and your dashboards if you want to see your threads for one year or two years you can configure your indicators accordingly but as Adam said reach out to support to help you in this but be ready to answer that question because support will not can tell you how to delete but you need to know uh the period that you want to keep your scores for Isaac in your opinion you know I'll say if you 80 million scores is fine if you need that data that's not too much right um if you said 81 billion uh we may have a talk but anyway yes we're fine uh but isak if they are still on the scores table and they have not moved to the L1 L2 for that I would definitely suggest to I highly suggest to move to the new course uh because you will save around 40 percent of your space with your literature so I would highly recommend to go through it 81 million uh do we want to go just with the automation or should we open up a case what do you think uh either way it's always highly suggested to test it in a non-production instance so that you can see how long it takes the whole migration process so that you can plan it during of hours so having the case can help answer some questions but if you try it in a non-production instance you will understand the whole process or if you want to be safe or feel more secure please raise the case and our support Engineers will help you throughout the process okay and there and if you search um so I'm going to put it into the I'm going to try and put it in the chat for everyone um the link to them the documentation on migration uh is right there so if you are still using the the level one or the PA scores table and not L1 and L2 take a look at the automated automated migration process this is uh uh a bit new anybody uh what do we have in here um all instances provisioned New York and later are already using it so if you're if your instance is only a couple years old this is not an issue but if your instance is uh five six years old or older then you may want to take a look at this to come through it's all automated and again as easy said if you want to open up a case to help us to have us help you it's fine um and if you want to do it by yourself it's fine starting a sub prod and then go from there the documentation I should explain it but it's it's just better let me know my the new architecture is just better okay great thank you Adam and ezoc we have a couple minutes so I'm going to go through uh the rest of these slides real quick and uh and then we'll get things closed out but uh first thank you ezoc and uh for doing the demo and showing how uh simplistic flows are compared to other ways of doing things and those use cases uh we really appreciate it and of course thank you Adam Dan and Terrace well for your time for being here to answer questions so maybe we'll do there we go so these are the links um that were used for most of this stuff and also added the one that uh Adam did reference uh in the chat for the workflow uh the flow designer and workflow automation tools so that'll be there it'll be in a PDF so you'll be able to get that directly from the community and as we close out make sure that your checking out the community uh there should be a a new look to it uh real soon we're changing up the links there so I'm just waiting for that to go in and of course this information will be out on the platform analytics Academy Community page this link will be out here and all of the previous ones we had in any of the future ones that we have remember that we have the pro tips out there these are real uh little quick hitter videos where we talk about just things around performance analytics uh some really great topics so check those out make sure that you use utilizing now learning uh just search for analytics there's quite a bit of training that's out there so make sure that you're checking that out for um to learn more about analytics and of course you can use it for everything all services now so anything in there you can check for the PA Advanced course is now on demand it wasn't in person but now it is on demand so you can take that and as Adam was talking about before uh these are all of the sessions for our sister session so the virtual agent Academy for um the one that we were talking to earlier you can see that here so make sure that you're checking that out as well and again it'll be in this PDF um and this is like some of you may have went to the old uh Zoom link uh this is the new one that you want you'll actually want to go to so again this will be in the PDF um and as Dan I mean sorry uh Adam and Tara did say uh knowledge23 is coming May 14th through the 18th in Las Vegas so um be on the lookout for that start registering for that and as Adam said there are there will be some really great sessions as a whole but um we in now intelligence will be providing some really great uh sessions as well so in two weeks we'll be talking about itsm on mobile so if you are looking for that information if that's something that you wanted to do or even just curious about it please join us in two weeks and as usual thank you very much for joining we really appreciate that you're spending your lunch or evening whatever time it is with us for this hour and hopefully you're getting great things out of it so we'll see you again in two weeks and thank you for everyone that joined