okay good afternoon good morning everyone um welcome to this session in the now intelligence area i'm one of your hosts today my name is dale cheeseman as you can tell by my accent i'm not from the shores of america i'm from the uk and with me today is my colleague from the netherlands uh jost van poppel who's also an advisory solution consultant and today we're going to take you through the unlocking of the power of ai and analytics with now intelligence so without further ado i'm going to progress through now the first thing is from a safe harboring perspective please note that everything that you're looking here are forward-looking statements and please take a look at this message and please do not plan your deployments as based on decisions in forecasting forward as these may change a bit of housekeeping before we get going please stay on mute there is a q a uh after the presentation uh please use the q a to ask any questions throughout the session uh feel free to introduce yourself when asking a question we'd love to know who you are where you're from uh the session will be recorded and shared on the servicenow community after this session completes and after the session ends you'll be prompted to fill out a short survey we appreciate your feedback okay so that's the housekeeping stuff done so as i said my name is dale cheeseman i'm the senior advisory uh in the now intelligence group based in the uk in a place called manchester uh and my primary focus here is to provide technical overview deep dives in the now intelligence products with me today is my colleague from the netherlands yost van popple so videos if you want to come after you and introduce yourself yeah thanks dale uh welcome everybody my name is jose von puppel a set based in the netherlands in the south eindhoven well have the same role as dale uh we're we are in the same team uh covering emilia and um later on i will show you a demo of uh the now intelligence story that bill is going to tell about okay thank you so thanks jost so before we get going just a short poll for you here so as of today which of the now intelligence capabilities are you most interested in performance analytics virtual agent predictive intelligence process optimization or hell yes i want all of it so short polls should be appearing any second now just give it a couple of seconds while we answer i'm hoping everybody says all so i can see the results i think we publish them at the end a couple more seconds got a few creeping in okay all right so uh vast majority everybody's interested in all and for everybody who's interested in all good because you're gonna get a whirlwind tour of the now intelligence group and what we're going to show you today is we're going to talk about what we want to introduce in the now intelligence it's very important as of the san diego release is the concept of hyper automation but contained in there is a lot of the now intelligence product sets and yoast is going to take you through some demonstrations of these products but what we want to sort of hear it what we want is to see is the product but don't be put off with how complicated you think this is a lot of these tool sets that we're going to show you today can be deployed relatively really quickly because they're on platform so without further ado i'm going to proceed now a couple of slides i'm going to present for about 10 15 minutes and then you'll get to the good stuff but why i wanted to show you this is the game has changed in the digital business and even before the pandemic you know organizations wanted to automate they wanted to ex you know expedite their processes and these reports here from accenture tech review and mckinsey they all start talking about being future ready now the thing about it is two years ago you know we were all faced with the world shutting down the need for automation increased due to remote distributed working environment most of our businesses found themselves in studies shown that intelligently automating work helps companies uh get more work done faster whether we're in the offices or in the work in a distributed or remote environment intelligent automation leveraging our platform alleviates the burden of repetitive work and boost efficiencies of the entire workforce with these reports accenture found that amid the current economic uncertainty a small core of companies and it was around seven percent have actually achieved nearly twice the efficiency and three times the profitability of their peers these are future ready companies and have doubled down on digital transformation and retooled operating models pivoting from incremental improvements to wholesale reinvention mit tech reviews comments were ran an article showing that productivity is on a huge upward curve largely down to ai investments the past decade has delivered an astonishing cluster of technology breakthroughs the most important ones are in ai the development of machine learning algorithms that have a direct positive impact on human productivity and finally mckinsey's study shows that in the era of the workplace upheaval companies that create tailored authentic experiences strengthen employee purpose ignite energy and evaluate and evolve elevate the organization's wide performance so this is really important so why we're talking about these slide contents is leaders that are transforming their operating model why hyper automation to achieve these outcomes leaders are employing this hyper automation strategy which remains how people and business process actually work most people have felt supported by their organizations through kobe 19 pandemic however there are still many challenges many companies exploring hybrid working models with focusing on enhancing efficiency let's be more productive increasing knowledge and intelligence and most of all nurturing an employee experience that creates a sense of belonging employees are hungry for trust social cohesion and purpose they want to feel that their contributions have been recognized that their team is truly collaborative to drive efficiency and productivity but at the same time automation can be used to replace what otherwise be considered unproductive tasks for example allowing workers to spend time doing what they enjoy more increasing their knowledge and intelligence working on higher value tasks powered by data and information at their fingertips delivered by ai recommendations intelligence increases significantly finally providing a great employee experience requires a profound reorientation away from traditional top down model to one based on fundamental of design thinking this shift allows companies to put this to put its workers first by exploring and responding to how their view their employees journeys then delivering a tailored interventions that focus on critical moments that matter to maximize satisfaction performance and productivity okay so hopefully this sort of makes sense but the thing about this is is that efficiency these these operational outcomes can have direct attributed and clear financial results as well results that improve not just the bottom line top line revenue retention to the most valuable asset people efficiency reduces operating costs seems a simple idea but if a service request takes less time to fulfill or an incident outage is resolved faster then it's going to save time and money similarity and you're going to see in the demonstration with intelligence upskilling workers to do more see more and understand more improves productivity increases satisfaction and reduces operational risk it's directly linked to be to the improved employee experience feeling more valued and ultimately results in a better talent retention and driving revenue so i've gone through all of this because we want to introduce you to the single platform of hyper automation we are very unique in our approach we have a maximum simplicity for automation by anyone and if i can automate anybody can dynamic processes consistent user experience single data model for complete insight the now platform enables you to evolve your automation capabilities and what we're introducing here are the four pillars and in the four pillars you can see discover automate apply ai and optimize now inside of each of those pillars there are phases and applications that are used from automating tasks and processes to apply intelligence with the ai to optimize existing processes and discover new automation opportunities the now platform has it all in one experience and for anybody today joining the call thinking about now intelligence and ai well anybody who's have these products or are looking to take these products you're already going to be on the hyper automation journey because as this slide shows here all of these things that you can see are a part of itsm pro so what we're going to do today is we're going to run through quite a few of these and my slides are going to get very quick now because my colleague yost is going to pick up on some of this content and he's going to take you through the journey now as we said we have we only have a small window here so we have to get through quite a lot so what i plan to do is at the end we have a q a we also have a round table that you can sign up for that will be posted in the chat and the thing about it is is that if you want to see more of deeper dives of these technologies please reach out to your aligned account exec or your aligned sc for more information so the first tool automation discovery this is a cool way of finding out where do you start where do i take a look at automation discovery looking at my data to actually come back to me based off nlu wording structures and prioritize my automation tell me where my conversations my virtual agent chat bots are or it can also highlight where your iton potentials are where your noisy records are so this is the first introduction to the ai piece and machine learning and now intelligence is actually finding out where your automation capabilities are the next thing you're going to see process mining people hear of process mining being able to uncover potential bottlenecks increase your resolution by finding these bottlenecks finding out where you've got gaps or bottlenecks and seeing what you can do to improve those processes we refer to this as process optimization get deeper visibility into your process gather insights on your bottlenecks and track the business value from the process improvements you're also going to then move into the apply a eyepiece and in the apply ai apply a eye area that's a mouthful you're actually going to start to see the virtual agent where can the virtual agent benefit me giving my employees that greater option including the ability to request something 24 hours a day being able to answer those what i refer to as the low hanging fruit high repetition low complexity information but taking the bot further with our automate capabilities in the automation hub being able to actually get the bot to do things for me actually driving changes so being able to provide these helpful services without having to hire and train additional staff so this is giving you this ability and with the virtual agent we expect to see over 50 percent of requests deflected imagine how much work you can get done when the bot is taking place for you contained in that bot is the natural language querying being able to take a phrase and translate it into an utterance for the virtual agent or translate that utterance and put it into a query and actually query the analytics data and come back with the most best answer and best visual for what you've said you're also going to get to see classification framework classification is learning historical information to predict an outcome so based off a short description i could set my assignment group my uh category my service so as this slide is showing here based off a short description this can prioritize the priority and also set the category you're going to get to see similarity framework and for me coming from a support organization similarity is the most easy deployable piece of tech in the now intelligence area because it's not fundamentally going to change anything it's going to add information to your agents being able to be told similar records that's relevant to them being presented with knowledge articles being given major incident detection similarity framework is the easiest fastest way to improve your agents clustering framework being able to uncover opportunities with improvement in unsupervised machine learning the clustering framework is going to do that by clustering your records and grouping them together for where you could see proactive problem management or you can identify virtual agent candidates the regression framework this is the final part of the ai piece being able to use that numerical model of learning from records to be able to predict resolution time the ai search platform feature but enhanced with ai search and our technology giving that greater searchable knowledge being able to answer questions in a format in a conversational way being very smart to understand the user's intent and then provide relevant proactive information for them ai ops is a part of predictive intelligence so as i mentioned you know as i mentioned we all have ai embedded in all of our applications and workflow from hr to itbm i want you to pull this example which is aiops we have a new unique way of utilizing predictive intelligence to actually find those ai operation potentials so then we move out of ai and we move into the final piece which is the optimize area these are when we start to talk about kpi composer a plug-in that you can down that you can install which makes your design of your performance analytics into a digital footprint where you can design and create prototypes you can visually plan how your kpis are going to support your business goals and receive a validated task list for your pa admin to actually build these out also finally performance analytics this is for me the bread and butter of the application for me coming from that analytical background being able to switch from that reactive state reporting and we've all seen it in excel and so on which is all out of date being able to be able to report and delve into on platform data and switch from that reactive state to that more proactive state of play so this is what you're going to see these are the four pillars of hyper automation discover automate apply ai and optimize this is the hyper automation the automation hub is the newest part of the technology with rpa there is going to be other sessions around that but with that i want to hand off to my colleague yoast who's going to provide you a whirlwind demo of some of the now intelligence capabilities so with that i'm going to pass over to yoast thank you yoast okay thanks dale for the great introduction um i will share my screen uh so so to take over from uh the last slide from deal so what we're going to see in the demo uh are parts of the uh hyper automation platform that you see here so uh what we want to show you is the following so a few things we want to show you the virtual agent so how can you deflect instance from your your service desk so how can you um alleviate pressure uh getting those incoming incidents uh away via for example cell service we want to show you how predictive intelligence works so classification similarity what bill was talking about and uh next to that we want to show you performance analytics so how can you optimize uh where to focus on how to excel on your kpis and we also want to show you process optimization so how does process mining uh work within servicenow so looking at your process where are the bottlenecks etc and apply automation on that or see where you can improve your process so actually we will start with with prose optimization so this is the discovery phase of the hyper automation journey and let me go to my right screen here so process optimization it's our process mining capability and uh important to stress out like all of our technologies mentioned is that they are all platforms so really easy to start working with this um if you uh execute a mining activity with process optimization and this is done for the incident process so you see here we run it on the incident process on 38 000 records you will be presented with course output so you will have a summary and inside step showing you some performance analytics kpi dashboards indicators we have some already some improvement opportunities with process optimization we also run automation discovery so those two are since our san diego release are integrated so when you do a mining activity you also will be shown automation opportunities so you will see where are opportunities in my incident base to address uh automation for example where can i apply virtual agent conversation i have for example account unlocks i have uh here see password resets i see the mean time to resolve i see the sort of insights in this cluster and i get already uh triggered that there is something pre-built for this or our out-of-the-box capabilities with our virtual agent available to address this particular cluster of incidents and then the other thing process mining itself so going to the analyst workbench this is the output of the mining so the mining of those 38 000 records shows you eventually how all this uh went through from how the incident went to true uh servicenow so they of course started with new and eventually they got somehow got results closed and process ends but where are my bottlenecks in the process so i can start looking for example for specific use cases and start saying okay i want to see [Music] reassignments so i want to see where in my process are reassignments occurring so i can for example state uh i want to see all the reassignments that uh all the incidents that have one reassignment so i will be shown a process so now i only see from those 38 000 records i see 7.1 000 records um that have been reassigned only once and this have of course has some metrics but i can start comparing this with an other filter so i can say uh here and i have something saved as well i can state here uh show me everything with a reassignment count greater than one so if i do this i will see everything but there's more than one reassignment and now i can get some insight so immediately i'm being proposed with some model statistics and also the difference between my left hand side model and my right hand side model so of course reassigning incidents this uh certainly will not lower your mean time to resolve and that's also what we see here but we also see how much this is so we actually know okay if we resign more than once on average for those 38 000 records it takes one week four days and 21 hours longer so we want to get this number down because if you get this number down you're working more efficient and eventually this will all address to those higher company goals so how do we do that well we we do this with applying automation so our other pillar so for example in this case reassignments we can use the classification from predictive intelligence so predicting the assignment group will certainly help get this number down so lowering the mean time to resolve and those automation opportunities showed password reset we can do this with the virtual agent so this is the first thing i want to show you [Music] in my other screen so imagine for example i don't know my password anymore this is something you don't want uh coming in as an incident because you want to allow for cell servers and we can do this with the virtual agent what i'm going to show you is out of the box so i can i'm not logged in uh at this moment but i can still state um i want to reset my password so when i reset my password uh for when i type in this the nlu kicks in and recognizes what i mean and it says you want to help with password reset that's indeed correct so because i'm not logged in it will start verifying my identity so in this case i have to type in my email address and now there are some uh two factor authentication methods in this case i have a q a so a few questions to verify my identity so i can type in the last digits of my credit card those are hidden at the moment i can also show them to see if i don't have any typos there and this has helped me to through the process um so what you're allowing me to reset my password myself so now i'm verified and i can start typing in a new password i don't want to actually reset it so i do something wrong but that's it so this is an out of the box capability that we have within itsm pro the virtual agent it's an out-of-the-box topic what you're seeing there's uh configuration done to get this up and running next thing i want to show you is when i finally am logged in and i'm going to a sort of different persona at the moment so uh imagine i'm a joe employee in this case i am logged in and um i'm in the service portal so this is the this is our employee portal just make this a bit bigger and here i see what we can do in the portal so here i have my ai search uh capability so i can search and it will actually recognize my intent and get meaningful results back but i also can open up the virtual agent here so i can start a new conversation so for example here's state let me state of it's already started but um if i hear state um [Music] vpn is not working so i have issues with vpn again [Music] the nlu kicks in and it says you would like help with vpn connectivity yeah that's correct and i would start trying to help me getting my getting me to solve my problem myself so what type of machine are you trying to access from it's a mac and it will give me some uh things to try to solve it myself and going through and i say this this didn't work and now i'm being taken through the knowledge base of my instance so i try to match this similar or articles that could resolve my problem in this case i state that i'm done and say that this didn't resolve my issue and now i'm um because i couldn't help myself i can i'm taken to an option to create a ticket for this so i say create a ticket and automatically now a ticket is created but the thing is the only thing i actually stated is that i cannot set up a vpn on a mac so i have problems with vpn on a macbook and but i didn't state where this needed to go so the assignment group and also the category in this case are predicted with classification so this is predictive intelligence kicking in selecting the assignment group based on my historical data set so i don't have to pick the assignment group myself from i think there are 90 assignment groups in this instance so the chances that i have this wrong will are very high and this of course will address this reassignment issue what we just saw in process optimization so anything else i can assist you with nope i'm i'm done uh my experience was very well so this is um well not actually fully helping myself so it was trying to help me and eventually i'm taking to the uh to to be able to create a ticket but it will go directly to the right assignment group i want to show you something else is that the virtual agent can be addressed or can be accessed from anywhere so we also quite often integrate this with microsoft teams so let me just wake it up if it is not already so again here i'm being greeted so it knows who i am also that i have a certain amount of open issues so if i have any questions or i want to escalate any of those issues i can access them in this case i want to say for example account and email so i have issues with my mail and also here um the virtual agent is triggered and it recognizes uh with by the nlu that i have email issues that's indeed correct and now again it wants to tries to take me through some steps to help me solve my problem myself so in this case i can't send email and will be showing me some options okay you can try this it will also show me a link towards a youtube video so we have those integrations as well so let's say i've got it and what i can do now is i can also and i hope this worked because the last time issues so i want to speak to a live agent so i don't want to take me through the topic i want to actually speak to somebody because i want to have this email issue fixed so in this case i say yes that's correct so now i can chat directly with a live agent so i have another window open so here we are with my famous colleague bet england so here we have a chat coming in so if i accept this i will be able to see the chat coming in from joe so i can see the previous conversation that was already taking place and i can start chatting so uh what seems to be the problem and this has been uh let me move this next to each other so this we i can go back and forward right so uh don't send email and we chat back and forth but eventually beth decides to create and let me also make this a bit bigger beth decides to create an instance but is because he can't help him directly so we create an incident and we also say here i'll send email to give it a short description and again here there is an incident created only with the short description but it already knows uh with whom i chatting so it creates an incident for joe and also here we see category and assignment group are already again predicted so joe is being said being said that there is an incident being created and this incident again is being routed to the correct agent so let me refresh this so it is in the in the conversation so if i open up the incident i actually start seeing uh what just has been created purely on the short description consent email so what i've already said category software and um assignment group software are being automatically predicted and this is done because i have 90 assignment groups so picking the right one is is hard from this many many assignment groups the next thing that i see here is that uh similarity is kicking in so i have here my agent assist uh panel or pain and based on the short description so based on only the short description it starts doing things for me so it starts recognizing in this case that there is also a major incident so um there are already there's already a major incident for this so i can link the current incident to this major incident so that we don't are all working on uh similar incidents while there is one major incident for this the other things that are also available within um this agent assist is if i don't have a major instance so let's click this away i have knowledge items so i have articles that are being collected which are maybe addressing this issue of not be able to sending emails so i can start looking in those articles the other thing is that i want to solve this as quickly as possible so if this incident is not the first time coming into my service desk i can start looking for if there are any maybe similar resolved incidents so again based on this short description it starts getting me similar incidents which are already being solved and i can look into those incidents see what happened and if any of those incidents have the same issue and the same and the same solution could be applied to my incident at hand i can simply also copy the incident resolution but more importantly i don't have to figure out what the best solution for this problem is because i can simply look at already solved incidents so this is predictive intelligence helping the agent solve issues faster so we have classification directing it to the right assignment group and we have then if it is at the right assignment group we have similarity helping the agent in uh similar results issues so bringing solutions bringing knowledge articles or if there is a major incident at hand linking it to a major incident or if i quickly after each other there are many of those incidents coming in it will propose uh to maybe create a major incident for this all uh being aimed at helping the agent solve issues faster and optimize eventually the user experience on the other side that's it for for the virtual agents combined with predictive intelligence the other thing that we want to show you as well is when we are in the optimized phase of hyper automation we have also something called performance analytics so let me just refresh this nice sorry about this logged out so let me go to my performance analytics dashboards and i have several dashboards available but one of the out-of-the-box dashboards available are is this one for the it managers so we have several dashboards for particular roles in your organization one of them is it manager and the the relevant kpis that we see as best practice for an it manager are being placed in this dashboard so i have an overview for example for the customer satisfaction score how many breached slas do i have daily planned changes the next seven days and i can also dive into those indicators and drill down deeper i have several tabs available so open activity analysis i have incident channel analysis so here i can see you can see a lot of things and i have those main widgets available on my dashboard but again as mentioned i can also dig in and and dive deeper to see what what is uh the issue maybe with certain indicators so for example here i'm looking at resolved incidents uh by channel so i see several channels where my incidents are coming in phone email self-service walk-in so if i for example here have phone and i start going over this graph i see my related indicators also changing so i see over time how all my indicators are performing and the same thing is for email and what i see here is i have the same indicators but if i stop hovering over this i see that the mean time to resolve is actually colored red so if i dive into this i will see uh so i'm now taking to the analytics hub and what i see here is um a trend over time so coming from july 1st 2021st i'm 2021 i'm coming to uh where we are right now so imagine that today is december 31st uh last year this is where i where i am right now and i also see that there is a trendline available so based on my historical trend it also predicts uh a forecast for the the next coming days of how how this indicator will be for will perform and as mentioned this indicator was called red and that's that's because uh there's already a target set so there's a target set here for uh 0.15 days so this line you see here so if i go into my targets i see that there is a target for 0.15 days so if i would clear this target it will actually remove this red color so that's it so you you based on your historical data we start showing you where you came from where you are right now and also where you will be in the in the near future and you can set targets where you want to be so if you set those targets so let me set a new one let's go a bit back so i'll set the target and i say here for example that i want to be at not 0.15 days but i want to be at 1.5 days so if i would save this then currently i see my target appearing currently i'm in the green so if i go back now i state if i will see that my uh the color of my indicator will change so so based on uh let me take email so based on setting targets within performance analytics you can influence the widgets that are available on the dashboards and within one view you will see if you are performing above or under target and where you need to address issue where you need to maybe optimize your process and if you start digging into these these indicators so let's go back in i actually have breakdowns available and i need to clear my contacts so if i now look at my average resolution time for for all contact types i also have the ability to view breakdowns for this particular indicator so not only for things that coming in via email i'll be for example able to see certain breakdowns for a priority so give me everything that is has a high priority so now i see the trend line for all incidents that have a high priority and also for this so i see my my target which i just said is not available anymore but i kind of set a target here as well so i can set a target here for priority two is high i can set a target for this specific breakdown that i want all those incidents that are have a priority high i want for example those need to be fixed in 0.15 days as of so if i save this i have a new target set and and that's it so per breakdown you can set those uh those those targets in your analytics hub and those will be taken back eventually on the widgets on your main dashboard so that's the the last part of of our demo so we we showed we we tried to show all the all the pillars of of hyper automation so we had process optimization and automation discovery um where am i here so we showed process optimization automation discovery uh the virtual agent combined with predictive intelligence and performance analytics so hopefully the this is a give a clear view of uh how the now platform can help you in your hyper automation journey uh let's go back to our presentation so we have a few other things mentioned let me check if i'm on the right if i'm sharing my right the right screen yeah so um as said we uh what we just saw or this this webinar is part of uh of a series so live or service now all the all the uh things you have seen are uh a part of a series so they're all placed on the community and youtube um we are following up today's session with a meetup uh next week so i suggest you take your learnings from today bringing up your follow-up question to this meeting for deep dive but we also want to open up the floor for any questions right now so i don't know if there are any questions for us we have we've given so much information they're probably processing everything there was one question that i did answer um but just for everybody on the call who may not have seen any of this technology before this may be very overwhelming for you but a lot of the technology that we deal with is very very simple to deploy and the tools that we want you to look at you know really is to sort of say what can the now intelligence itsm pro or enterprise bring me that i can expand and improve my my experience for my customers and my agents so where we started talking about automation discovery clustering analytics finding those automation capabilities looking at the ai technology to improve my customers experience by rooting tickets faster looking at the similarity framework being able to present my agents with information to expedite that process so we've done we've give you quite a lot in that short period of time now obviously the round table that's scheduled for the 19th of april this is where you get your chance to come at us and ask any of your questions now the q a is quiet so what i'll do is i'll i'll give it the once more around the room if anybody has any questions they'd like to ask i hope you've all enjoyed this and i hope to see you all at the round table so any more questions potentials if not then i thank you all for joining today's session and i look forward to the round table on the 19th thank you everybody and have a good evening thank you thank you