hey everybody um welcome my name is Mark Casto I'm product success manager here at service now and today we're here to talk about measuring and optimizing your application portfolio so as we go through today there might be forward-looking statements that are based off our beliefs and and if we make any of these it is our best knowledge at the time but there's no guarantee that it will happen when we're talking about future feature sets and capabilities so we have to present you with this safe Harvard harbard notice um want to advertise ourselves so this webinar and whole series of webinars um that are here to help you learn more about service now get more out of your products um click on the the little QR code or go up to community and look for live on service now um we have all kinds of things going on we try to do these um pretty much once a month so um in fact later this week we got another one today is on APM we got one on Resource Management in SPN uh just two days from now so um we're going to have um the format of this we're going to have quite a bit of time at the end for Q&A and uh if you'll use the Q&A button at the bottom to ask your questions uh some of them we're going to go live we're going to let you talk if you want to or you could just ask it um the rest of my team is on with me Justin and Rob and Shan are here with me to help answer the questions as we go um this thing is going to be recorded and shared so you can always watch it after it and we have a community uh article kind of a landing place for this I I'll talk about that in a moment um and at the end we have a little short survey um ask you to please give us your feedback tell us what you want to see tell us where we can improve we welcome your feedback so um today um I'm Mark Casto I am a principal product success manager for the Strategic portfolio management product Suite um Rob and Justin you guys want to introduce yourselves hey this is Rob Ericson yeah I'm also a product Su success manager and I'll be uh helping Mark out by answering some questions in the q& thanks Rob thanks Mark Justin Catchings product success manager based out of Munich Germany and looking forward [Music] to supporting the session today yeah and just a little context I'm in the Southern United States Rob's up north and Justin is over in EMA so you got coverage for half the world but we're we're here to help you um you know we're our our whole job is to help our customers get value out of our products so I want to start with a little poll just to get a feel so is your company performing some sort of a app rationalization process today and and where are you at looks like most everybody is uh that's answered is doing a you know something that doesn't cover the whole inventory but it's great you're you're doing something and and this is where we're leading all right and so 28% haven't started 53% have something going on um 12% are using service now thank you it's it's wonderful we want to share with everybody today and 7% are doing R apparat outside in an external tool so our next question is if you're doing appar rationalization how mature are your metrics you know what how deep have you gone how expressing Okay so um results is mostly low maturity subjective I'll tell you from my experience in in working with this stuff over the years subjective is a great place to start and actually it is very important it's a facet of your Enterprise systems that you need to understand um it you know the subjective point of view customer satisfaction Biz value all of that those are important attributes as important as some of your objective ones so thank you for that so the objectives today is to understand the current out-of-box metrics and just how they work we want to one of the things we talk about app rationalization a lot with APM um and we've done demonstrations of it and and the abilities in um Enterprise architecture workspace but how do you get there how do you get these metrics today we're focusing on the metrics and in in APM terms it's an indicator want to talk about the indicator also the scoring capabilities because score is really just a special indicator if you will so um we're hoping that we help you understand how to how to define your metrics and build them getting up get them building an in production toine a road map to improve them and continuously improve in the end your rationalization and evaluation capabilities um all of this we've built a community resource it's an article up on service now Community mejor application in inventory with indicator and scoring framework and we we put some content up there um about the current state the indicators that exist Etc but more importantly this will be a place place where we're going to post examples and in fact if you want to ask how can I do this how can I do that we will try to help you this will kind of be a place for the community to gather so I wanted to point that out I'll come back to this and this will be available to you as a PDF so in order to improve you start by measuring um you need to we want to be able to do apples to apples comparisons with your with your your Enterprise systems your business applications metrics can be a mix of subjective and objective as we mentioned earlier the information exists in service now and that could be via an integration so maybe the source of information is really external but in Integrations they're not that hard to get in place and so it could come from an external Source but from that you can derive a metric or an indicator so the indicator framework starts really thinking of it it starts with the service graph all of this information that is in service now and so we we call it the service graph everybody's familiar with csdm and all of this interconnectivity the cmdb core data performance analytics and over here on the side you see we got assessments that's a facility and then we have four key kinds and actually there's a there's another type of indicator which consumes indicators we'll talk about all of these in a moment but um here's our indicators these are a way to pull particular information it could be a query indicator which is just querying off a table custom script you can do about anything you imagine if you have per performance analytics already configured and indicators in that we can we can turn that into a APM indicator and then of course there's the assessment for the subjective that all feeds into a scoring profile it's a weighted sum of each indicator's value so each indicator delivers you a score that score gets a waiting from the scoring profile and you end up with your overall score the design properly imp implemented should give you a consistent score across your business applications and it's it's really there just so that you can speed up your rationalization process this is all tools to get to rationalization evaluation and management of that inventory so there's some key Concepts I'm just going to cover them real quick indicators that's a metric data source what kind of data sources assessments performance analytics custom script query condition then finally indicators which allows you to at least in a a two- layer hierarchy in other words you can have a parent indicator that has child scores it won't do a full hierarchy that's something that we hope to see uh as an improvement as we move forward over the next couple of years frequency how often do you measure this you don't want to measure it every day probably some things might be good to measure monthly maybe such as number of incidents change hours those kinds of things other things make sense to do quarterly and some things you simply do them once a year that's part of your design the direction are what is good is it good the lowest possible value or is it the highest so that direction in the indicators one important aspect 10 is good one is bad so it's always one is bad and always 10 is good so the direction when you take your raw score so if it's something like incidents you want to minimize that so the application with the least number of incidents is going to have a 10 the application with the most across whatever period is going to have a one so all this direction does is IT addresses how the indicator scores calculated targeted Min and Max that's the high low thresholds to throw out outliers that way you don't get skewed by somebody that has just something off the charts and then application scoring profile that's that single score um configuration built out of a set of indicators and it'll give you that single score each indicator gets a weight that it contributes some result key Concepts and we'll go look at the table if we have time today um we should have plenty of time but the application weight that's your raw score so if you're counting number of incidents that's going to be the number of incidents across that period if you're um if you're looking at business value for example that is going to be the score that your application gets for the business value questions APM in indicator in assessments is the equivalent to an assessment metric category so in in assessments we call it a metric type that has categories in it metric categories and the questions are called metrics so a category might have one to 10 20 however many metrics those get summed up their values get summed up in the category gets a value an APM indic indicator when you're using the assess data source is going to look at that category value normalized value so this is that 1 to 10 based on application weight you'll see I I put the calculation in here so normalized values app Weight minus the Min weight over the max weight minus the Min weight it's where you are so you take of your whole application inventory that's being scored by this indicator it takes that it finds it keeps track of the highest and lowest that's your Min and your Max and and that threshold mind you can can affect that so if you had outliers that's where you put the thresholds in there but then you take where's the app in that scale where's the apps raw score in that scale gives me a a a one to 10 value if the direction is set to minimize we just reverse it so that 10 is good one is bad the indicator score so an indicator is tied into a scoring profile so that scoring profile has a weight how much does this indicator's value contribute to the overall score so that is how it's calculated so normalized value times the indicator weight from the profile divided by the total weights of the profile so it just gives you that percentage we actually have capture that for efficiency and Performance Physical period what period are we measuring months quarters years and Target minimum Max we talk about that that's the that's that thresholds to keep the outliers out and then the total weight that's used in calculating um the you know part of the the calculation that's the total weight across all of the apps being uh evalu so what do we do with these things well one of the most this is not the only place but probably the most popular is the bubble chart you can filter by some key things category and business capability are the two that I recommend the very most and of course physical period um and you can use different X by y here we're showing technical risk versus business value effec Gardner time model and we're showing and then the size of the bubble is would be a third metric so we get three dimensions here um in this case tag risk business value and maybe TCO for the for the the bubble size you also have a list view this is configurable this the X by y by Z you can configure using the bubble chart configur figurations um just like you did in the old APM home these are still valid they don't there are some things don't work color coloring is still done by plan disposition in the new one the old one you could specify the color depending upon the quadrant that it fell in um there may be there there has been an ass to allow for different sort of coloring and then also the other thing you can select is what are you what are you measuring in the z-axis are in this bubble size is it TCO is it something else so we want to talk about the types of indicators and run through them real quick so you're familiar with how you can build these different indicators this first one is anatomy of an assessment indicator data source is assessments we're going to maximize it um we pick the metric type that's the assessment we pick the metric category that remember that's where the score is actually coming from and if per application if you have one application it's got three different business value assessments within this the U physical period which is being evaluated say a quarter or a month or whatever it might be um this one the frequency set up for a quarter so if I have multiple business value assessments it's going to average their value vales pretty simple and straightforward the assessment part can get kind of tricky we're going to have information up on the community about that probably a little too deep to cover today but we we'll see if you got questions ask them in the Q&A section uh here in a moment query condition this is maybe one of the simplest ones we simply we specify what table we're going to look at we're talking we're evaluating business applications so most likely you're going to look at the business application table and um you're going to aggregate on some sort of field so in this case usage we're aggregating on active user count we're going to sum it up we're going to group by the CIS ID meaning the ID of the business app and that's what we pass on so group bu always has to be something that's a CIS ID if we picked another table to do this then we would have to pick in this group by we have to pick the column that represents a business application and hat for to to evaluate business applications we have to group our accounts by it so if we had some sort of table Associated to business application maybe TCO might be a way of doing it or something we would have the TCO table but we picked the Biz app table consolidation if we had multiple values again what do we do we average it um if we need to consolidate if a series of records are being Consolidated we might have to spe you know what what's the date field that we're going to use to see whether or not it fits within this physical period so that's a query condition indicator performance analytics if you got performance analytics already configured maybe some of the stuff that's tied in digital portfolio management you could roll it into an indicator so we've got Source performance analytics the performance analytics indicator and what breakdown are we using and should be business application and then if there are multiple records what are we going to do in this case we want to sum it up it's change hours that we're doing and in this case we want to minimize next this is the powerful one you can do anything you imagine um but you know it's you know you just got to write the script so um we'll have some examples um in in a few weeks up on the community site so you can see them there's already a dozen or more examples in the system out of the box so the source custom script consolidation in this case we want to average it could be some what it depends upon what what you want to do the script you see here at the top in this example the script is returning a Json object that has a single array in it called applications the array has an object that has app ID that's got the CIS ID of the application and the number that you calculate super simple that's that's what you return to the system so the way um service now scripting works is it's going to get called there's actually some parameters passed in the start date and end dates so you can use those dates to evaluate your record so if you're out fetching um just say incidents or you're fetching problems you can filter your results that you're going to count or you're going to evaluate by the start and end date um and so that's that gives you the filter you return this Json object and then the system does the rest for you and then finally indicators indicator this as I said it's only one layer we one of the things that we've got a request for is having a hierarchy of of indicators so that you could keep nesting them to to multiple levels but right now we can only do one level so here are the children and we sum up their values average whatever makes sense and you know it it it runs these first then it generates its value so this one in this example is just a overall um operational metric so we're looking at changes problems and incidents in this case bringing it all together into a score this is scoring profile so this is the default one everything is thrown in there are 14 of them um you want to edit this when you go and and start looking at going into production or something use the ones that are working get take the ones out that are not working um there's you know it's there's no use running them and having empty values of that sort of thing so the indicator is it used in the CI score calculation you could actually have it run the indic Ator and not use it in the CI score calculation um seems a little odd but maybe you want to have some of these indicators available for other purposes it's not all just about the score it can be other kinds of measurements other kinds of evaluations so it makes it flexible and the nice thing here you got a nice button to go force all of the indicators to run so you there is a job that will do it for you it runs every every day and it looks to see are we on a boundary month boundary or a quarterly boundary and it'll go run everything that it's valid to run at that moment for the specified physical period um but what if you know if you want to regenerate maybe something then go right or you just introduced a new indicator you can tell it to regenerate you pick what physical period and and off it goes um and then finally the weight so here's that weight that we were talking about when we calculate the U indicator score that's that contribution so this weight is passed down so think about how that is how that's being handled if you have multiple scoring profiles and in indicators in both there's going to be some mix which profile does it does it is it using so when you run this if you run it from directly from the profile we know what the profile is but if we have multiple profiles what can happen is that indicator is going to flip between the um the weightage of the two different profiles um and so have a different weight what I'm saying is that you have to consider the fact that you have multiple profiles using a single indicator and you have to accommodate for that you might have to CL BL that indicator so it's Unique to a particular profile just something to think about and consider as you do your design and that's it for the presentation what we really want to do now is go live we want to talk about some of the questions also we'd love to hear your thoughts on where can we take this what can we do to improve this facility and and you know make your your lives better and and your APM program more powerful so we can open up um Justin Rob if you guys want to ring in shaan if if we have somebody if you want to um you know raise your hand do let us know and we'll unmute you and um we can have a discussion hey Mark there was one question that came in it was kind of a three-part question it's in the answered portion of the Q&A but I'm wondering if you wanted to to weigh in I try I did my best to try to answer it but I think you might have a have a perspective that uh that might be helpful the one from mlo let me see if I can pull that up we have a couple new ones that have come in too all right we use another tool for application portfolio Lan ex any import portfolio with many selected attributes into service now do you know sap lean I I know of lenx yes we are aware of it uh what is the major strength of APM over competitors so this is not really an indicator question but I will tell you we have a builtin facility for measuring and doing rationalization we have direct connectivity allinone platform to not only the Strategic world with strategic portfolio management strategic planning okrs um every form of execution can be managed we also have that connection to the whole rest of the platform think back to that original picture the service graph we're connected we're a part of it so nobody none of our competitors have that to me that's why APM belongs where it is here in service now how to measure the added value of a tool like APM from my your point of view um is the the point is is things are not disconnected we tend to build silos by having a tool for this and a tool for that we actually have a picture I don't have it quickly available but you think of it you've got half a dozen different tools all disconnected and you're trying to stitch them all together to paint a picture of the Enterprise as Enterprise architecture you need everything at your fingertip tips bringing the platform together and being able to understand from the Strategic and the abstract all the way to the physical like plug on the wall on the floor of a data center running a a server having that visibility top to bottom that's not something you're going to get in a disconnected tool so um Andy you said you know subjective metrics there's you've got to take in mind especially for some applications if you're thinking about retiring what about customer set what about business value technical fit so um those kinds of things can be pretty important so is it safe soon there's no benefit of having use the CI score calculation if we're not using multiple scoring profile so Mark the isn't the CI score calculation um almost an aggregate of the different indicator scores of a business application um so that you're bringing several different qualities of a business application a single score so um yeah it is and the reason reason we allow different profiles you might score your your cloudbase or SAS applications differently than you score your data center uh on premise applications that's the reason allowing different profiles so you really can get an apples to Apple's comparison but that single score is valuable so I would say there is a benefit to using it um that's that's the whole point is you have the individual metrics of different context and then you next you also have um you have this view of a single score where you can quickly say this one's good that one's bad it's all about and helping and not it's driving rationalization but really just aiding rationalization well the only way you can waight those different indicators is through that scoring profile so there are certain things that might you might be taking like TCO might be more important than you know customer satisfaction potentially so then you want to wait that particular indicator as more important than the others and the only way to do that is being that scoring profile yeah you must have and let's be clear about this in order for an indicator to work it has to be a member of at least one profile y so question go ahead yeah yeah thanks so so much I think that was basically my my question this is nil um I think that definitely helps me to understand the the real benefit of multiple scoring profile so really appreciate that excellent so we got Eda and shes if you guys I don't know if somebody can take them off mute can we unmute yeah I think they are they have uh talking permitted go ahead um hi um mark this is AA so question along the lines of the multiple profiles so one of the things that I'm shifting everybody towards is kind of capability business capability type um planning and I want to be able to use those indicators but so if we use that kind of as a primary what would be then the reasoning for using the business application profiles like what would be that combination or is it one or the other or it's just going to depend no they're two separate things and and I didn't speak directly on the capability the capability profile only allows a single profile for the capability score so you can't have multiple for capabilities whereas you can on the business application but imagine the business application scores that might be rolled up in your take and made a part part of your technology score and business capability to me that makes good sense the tech you know you evaluate capabilities traditionally on people process and Technology my Tech score could include the supporting business applications contribution to that yes you kind of get the best of both World exactly and you could have you can when you build an assessment I don't know if everybody knows this when you build an assessment you can have metrics in that assessment that come from just calculations from script so it doesn't it isn't just questions so an assessment can literally be a mix of both so you could do your app scores your app scores could be a contributing part of the text technology assessment for your bisc buildings and you line it up by that um you've got a technology rationalization Tech rationalization really in um in APM what we're concerned with is the tech risk there's not a indicator framework you could in theory do some sort of indicator um or a p performance analytics but Asset Management tends to own the individual technology software products Hardware products Hardware models software models all of that there's not really a rationalization over there but it's feasible that you could construct some sort of a technology rationalization I suppose um but the indicators won't let you target specific Technologies I think what you're asking for maybe I'm thinking in terms of trm so we've talked about trm for many of you technology reference model that's your technology standards one of the ask that we've got for enhancements is ability to run indicators against that technology standards and how well are we doing how how you know is is that way you can do rationalization with it so the I you the short answer on Tech ration ization is um other than risk we don't do a lot right now that risk becomes the U sorry Mark that the risk becomes one of the indicators within the apparat world yeah yes there is absolutely it does become one of the risk it is is so we we figure out what you're using we also figure out technical deaths uh where you're using non-standard and then we have evaluate the risk so a application service that has multiple risks as opposed to another application service that only has one risk is going to get a worse score and then it all adds up so it kind of rolls up to the business application got it thanks marike and SES do you still have a question anything else anybody Rob Justin you guys have something you want to share I would you say there is some good guidance out there and I think Mark you've posted this in community as far as if you want to start out with just uh the subjective assessment surveys um you'll probably notice that our outof thebox questions aren't don't gather too much but I think you've got some good examples out on community of some some a good starting point uh for some some uh I guess more valuable questions that could be used within those assessment surveys yeah so this this is that article that I was talking about and you see we talk about what's out of the box some of the old ones that are still out there but when it gets to the assessments I give you some some questions so these you know use them if you like them um if you got some others um absolutely um you know let us know so sh how do you classify the application category and application category group those traditionally the intention is that's the technical function is this a bi tool is this a general ledger application is this a customer relationship application so technical function or technical capability is how we would typically use application category there's a lot of other ways that people are doing it so it's not written Stone so to speak but the thing I want to point out is business capabilities are what your company does having a notion of technical function is also important that way if nothing else I've got 20 bi tools I need to gather them all together how do I do that I put them all under the bi tool category because that's that's really what it does I got Tableau I got cognos I've God powerbi can I get rid of one of these and that's that's kind of the the play on so hopefully that helps um anything else uh and I'll let y'all speak if you have something yeah so um yeah Mark you know I um like your your approach of you know as you start this journey evaluate if assessment based indicators can Surface some initial value and then as you start to think about how you can start adopting data driven indicators whether they query driven or um coming from performance analytics um a good another great starting point would be assess um where you are in your csdm journey because we um in the SPM and APM you know product unit are using S csdm as that model to enable automated indicators to allow you to surface performance metrics of your application landscape into APM and we offer several different services to help assess where you are on your CSD journey to help get you started to help further mature it as well as content within our community and Our Success Center which can help support success in that journey and enable you to adopt um additional uh data driven indicator [Music] capabilities yeah thanks Justin and um Eddie you made a point of um a working example we've got a little bit of time so let's bring up a system and actually look at a working example bear with me one moment while I log in so I'm going to use actually just one of the um out of the box so that we can we can see how it works um to get to the indicators you're going to go to Administration application indicators and here's that list of out of the box indicators in this case I'm going to do business value I'm just going to do an assessment so here's the metric type that's the assessment here's the metric category and we'll go to the category and take a look at it we'll notice that this one's been Rong because we see the successible records um here is why you we give you a list of of questions because out of the box we only have one uh one question so this one was assigned to a system administrator to run it what I'm going to do go back to the indicator and use the indicator facility so what happens is first the assessments have to run another important fact around the assessment ones when I generate these assessments and it launches an assessment instance it has to be complete in fact all of the the assessment the individual assessments that are assigned they have to be complete or cancelled for the assessment instance to be be complete and for this indicator to go actually grab the values so if you run your assessments but you don't finish all of them and you try running the indicator and don't get values first thing you want to check is your assessment instances so here I'm going to generate some assessments um I'm going to able tutor I'm just going to pick his applications so here I've got nine records and I'm going to rather than the user group I'm going to do it by user field I'm going to select the [Music] field It app owner which is the same thing I'm doing up here and I'm going to send out the [Music] assessments so now I impersonate able and actually before I do that I want to go back I'm at the indicator right now if I generate indicator score for 524 and I try to do it for wearing Q2 mine's configured so this is still Q2 for this system if I try to run it I should not get any values if I go to the um the scores and let me open another tab actually so we go look at the indicator scores so um 24 Q2 show matching um this is some bad demo data get rid of that and actually I'm just going to group by the indicator and you see there's no business value out there none of them were generated so these this is broken data this is due to the fact that some old total cost ownership indicators are there so this is this is a data bug in my system so now what I'm going to do is I'm going to impers able Able's got a uh set of Assessments out there so down here in self service business value so I'm going to just go in and rate everything oops miss one so I'm completed it I now go back to Sis admin I go back to my indicator I can now individually I could go to the profile and regenerate everything or I can just individually regenerate it for FY Q2 fy4 Q2 so the U these are done asynchronously it can take a little bit there we go business value and there you see it so whether it's here the data source was a score on those categories if it was total cost ownership or something else that would it it would still look the same we're going to get a normalized value which is based upon the range where where this fall Falls in so I red APM rather high so it got the highest score of Salesforce got the lowest score so any other questions comments I hope that helps Justin Rob Joan y'all have anything no I just want to compliment you on being able to uh give that demo on the Fly very [Music] good know Mark's the man so um I saw a question can we customize that assessment yes that's the whole point is those are you know do you recommend this application it's just silly um the application these assessments need to be configured and in fact what I like to do and we will we don't have time for it today but we will have an example up on the community soon as we can put it together take your assessment questions put them together in a single assessment send that to your it owners or your business owners let them do all of the categories and then you have a separate indicator for each category you pull your values but one assessment not 10 assessments though if you do that they'll hate you so um keep it down to one assessment if you can and we'll we'll have a an example up on community for you guys on how to do that put it together all right um anything else I think m we're all good we don't have any open questions we don't have any hands up so I think yep we're good for today excellent thank you everybody for attending uh today and if you have ideas for metrics that we don't have out of the box and you would love to see it let us know please please you know tell us what you need um we're always looking for ways to improve and thank you have a wonderful uh morning afternoon or evening um depending upon where you are and we look forward to seeing you again next time