ServiceNow, Inc. (NOW) Earnings Call Transcript & Summary

Hello, everyone, and welcome to today's webinar, Using AI to Enhance Field Service Technicians Productivity and Efficiency brought to you by Technology and Services Industry Association and sponsored by ServiceNow. My name is Vanessa [indiscernible], and I'll be your moderator for today. Before we get started, I'd like to go over a few housekeeping items. Today's webinar will be recorded. Owing to the recording of today's presentation will be sent to you within 24 hours via e-mail. Audio will be delivered via streaming. [Operator Instructions] I would now like to introduce our presenters today. John Ragsdale, distinguished researcher and VP Technology Ecosystem for TSIA, Bulent Cinarkaya, VP and GM Field Service Management for ServiceNow and Brian Philbin, Senior Advisory Solutions Consultant, Field Service Management, also with ServiceNow. As with all of our TSIA webinars, we have a lot of exciting content to cover in the next 45 minutes. So let's jump right in and get started. John, over to you.

Well, thank you, Vanessa. Hello, everyone, and welcome to today's webinar. We all know that AI is one of the hottest topics out there today and the potential for field service is huge. And we're going to be talking today about some of the use cases for leveraging AI and even Gen AI with your field service organization, the business impacts. And I think we're also going to touch on ultimately how this can affect the customer experience and customer effort as well. So in the abstract for this webinar, we teased up this idea of reactive, predictive and proactive. And I think we all understand this. Unfortunately, many field service teams are still in a very reactive mode. We have seen a big push towards predictive, which is seeing certain error codes that may indicate something but with AI, it's definitely going to allow us to be much more granular and pattern matching, predicting of what's going to happen and getting into a really more of a proactive mode that things are much more automated, that we can detect things much earlier and ideally prevent downtime entirely, have parts shipped before a part even fails, et cetera. So we're going to be talking a lot today about predictive and proactive. I wanted to share a little data just to kind of set the stage. And I understand that service organizations aren't always the first in line for budget when it comes to things like AI, the first people in line to get those data science resources but luckily, as you're going to hear from our guest speakers today, there's increasingly more capabilities available off the shelf. So this is no longer something that you have to build yourself. So a few data points, just about 1/4 of support and field service organizations are doing any experimenting with Gen AI so definitely some big opportunities there. For many years, we talked about wouldn't it be great if we could build self-healing properties into technology, which, again, would take a pretty healthy dose of artificial intelligence. Currently, only 0.3 incidents are resolved through self-help. And I think one of the issues there is development priorities are often set by the sales organization or maybe customers clamouring about bugs and enhancements and not always taking prioritized input from the service organizations on what the impact to some extra development dollars could be for them. And finally, just under 6% of incidents are resolved by diagnostics embedded in the product. So we know that there's a lot more abilities today for accessing and monitoring remote equipment. So again, with more AI capabilities to analyse that, hopefully, that number will continue to go up. For the pacesetters, this is from a recent survey we did on IoT and connectivity and we see that for the companies that have adopted some AI to look at remote logs and monitor remote systems some really big business impacts, 26% improvement in resolution time and a reduction of 76% in overall resolution time. So some really big business impacts that are available, but you've got to pull the trigger. So the majority of companies say that the top use case for remote telemetry is analytics. And we're going to, again, be talking about the role of AI today. It's one thing to access that customer equipment to do some reactive problem solving. But the more data we can access and the more sophisticated the analytics become, the smarter we're going to be about moving into that proactive and predictive state. And the final piece of data I wanted to share, we just completed our 2023 technology stack surveys. This is the first of this data that's been viewed publicly. I just sent the results to the research organization yesterday, so I'll be publishing a report a little bit later this year that summarizes some of the findings. But this is looking at the planned spending around some of these more sophisticated capabilities for field service, AI enabled troubleshooting, 2/3 of companies had budget for that this year, 43% next year, installed base asset device management, which is not only intelligently understanding what's on site with each customer, but can also be getting into some adoption consumption information. About half of companies were budgeting for tools this year. Remote assistance technology, we used to think of that as remote control and screen sharing, but it's taking on a lot more meaning today with intelligence, monitoring what's going on and remote system, 68% of companies had budget for that this year. And finally, work assistance technology, and this is a category that TSI uses that includes Gen AI and productivity enhancement capabilities such as Gen AI for field techs. And 40% of companies had budget for that this year, 45% next year. So hopefully, we'll see that 27% who are currently investigating start to go up a little bit more.

So enough for me, I want to bring our guest speakers and to the conversation. And I want to start with Bulent, who is Vice President and GM of Field Service Management for ServiceNow. Bulent I know that you are dealing with field service organizations every day. How do you see AI playing a role in the services industry? And what are some of the business drivers and initiatives affecting field service?

Thank you, John. Let me start with the business drivers because new technologies come and go. There's always companies that are looking ways to improve their bottom and top line. But the fundamentals always remain the same. So what are the field service business drivers we're all trying to contribute to cost and profit margins while keeping employees and customers happy. This is easier said and done. And to be honest, you can never say you're done with this because this is a better, better, never done kind of situation. There is always room for improvement. Any percentage improvement you make might mean millions of dollars for organizations. And of course, when you look at these business drivers, decreasing costs, keeping employees healthy, customers happy, AI can play a role in each of these drivers. Of course, based on your company objectives, this is a delicate balance across multiple initiatives at your enterprise. And again, before I say, "Okay, let's inject AI to your business processes," I want to talk about the fundamentals because in order to take full advantage of the AI and the benefits you can get, first, you need to put the foundation right way, right? This is most of the time how we see the problem in the service industry. So companies use CRM applications, beautifully engage with their customers, and they use ERP transactional systems to fine tune the back end, but there is this human middleware at the centre. So no matter how beautiful you engage with your customers, if you cannot connect up the silo systems, no automation and people need to rely on messages and e-mails to solve the customer problems. This is not a fine-tuned system. Of course, you can inject AI and get some benefit, but you're not going to get the full benefit out of this. So I guess the first thing companies that they think about, how do they streamline this process, right? So this is where ServiceNow comes and helps organizations first and foremost, all the way from your customer experience, your own experience to partner ecosystem experience, you can really streamline this. And if your flow end-to-end experience like this, then let's talk about AI because there's tons of opportunities to inject AI at the self-service level, customer experience for your employees experience and for your partner ecosystem experience and automation also plays a big role. And when we talk about AI, we shouldn't be only thinking about Gen AI. There is this good old traditional machine learning and AI, you can inject and play a role here as well.

So we know that there are dozens of transformation projects going on across enterprises as well as in field service. And I'm wondering how are leaders supposed to prioritize that. And I'm curious if you're hearing AI as a normal part of the conversation in the field service industry? And if so, what are the specific use cases you see as the most promising? I'm hoping you're seeing a focus on all areas cost, productivity, improving the customer experience, improving the field tech experience. But what's happening in the real world?

Yes. So let me share some data actually. This is from service console, which I'm an advisory Board member. This is back in December 2022. So state of the service survey results as you can see for service organizations, the top 4 priorities and also they are planning to implement within the next 12 months is all field service relevant related, field service, mobile apps, scheduling, routing, AI/ML. But one thing happened since that time, OpenAI is the fastest to 200 million users and suddenly took over the conversation and everybody interested in because it's relevant. It's like human, you're talking. People are adopting in their personal life. So the adoption barrier went down and ServiceNow Council actually refreshed the survey in September. So as you can see, AI/ML part with the push of Gen AI went up 20%. So definitely, companies are prioritizing Gen AI, AI/ML and it's becoming a top priority. And you can see vendors like us, we're prioritizing, we're taking actions, bring this technology in the right way to our customers out of the box, so they can start using it. So how this would benefit field service organizations or generally enterprise? I want to share one slide with the audience. Actually, I would like to give the credit to TED talk of Khan Academy. So it caught my attention there as you know Khan Academy, provides online training to students and they're all about improving and how to make it better. So this is based on a study done back in 1980s, it's called 2 Sigma problem. 2 Sigma problem is that the first light blue is the bell curve, the performance baseline in a 300-people classroom. And that study says, "Hey, if you apply mastery learning to the classroom, meaning that you move to the next topic, once you master the previous topic, then the baseline moves 80-plus percent of students are successful." And if you apply on top of it, one-on-one tutoring then the baseline moves 98%. So 2 Sigma shift happens but also studies say this is not scalable because you cannot really provide one-on-one tutoring to every student in every classroom. But if you apply this in today's online situation and if you apply this to your enterprise, imagine every one of your employees has a digital assistance and they get one-on-one tutoring for learning purposes. Suddenly, you're improving your employee performance. So this is how it resonated with me and it applies in my opinion, to how Gen AI can improve the employee satisfaction and reducing costs, all these business drivers that we talked about. So if you look at all the use cases, first low-hanging fruit, summarization and all those stuff, but it's becoming more and more sophisticated, use cases as well, this is becoming true, give them the digital assistance so they can get things done faster and give them the tutoring so they can learn not just follow the instructions as well.

Well, I've got a follow-up question there Bulent. I think this is something Brian can chime in on as well. What are the challenges that field service companies should be aware of when they're implementing AI? And what are you seeing out there as potential barriers?

Yes, definitely. Every new technology has its own challenges, but with AI, especially Gen AI, the early news was about hallucinations and the company confidential information leaking on the Internet kind of thing. So I'll talk about the business blockers, and I'll let Brian talk about the user blockers since he, in his previous life worked as a field technician as well. So it resonates with them more than unfortunately, I didn't work that way. So one is cost. How much is this going to cost? How can I predict how many calls I'm going to make? This technology is not cheap. And as I said, data concerns, what's going to happen to my data is my information being used to train the models so that my competition is going to take advantage of what I'm providing. And if the recommendations are bad, so is it going to hurt my reputation as well. So there are a lot of data governance, the controls around it, the security and the kind of blockers out there from a company's point of view.

Yes. From a user's perspective, technicians are always the smartest person in the room. So they don't trust machines, they trust themselves and their fellow technicians because we're all members of the club. So the first hurdle you've got to get over is, do I trust the data that's coming to me from the machine. The other thing is if the responses I get back from AI or even just simple knowledge search, don't meet my needs. I'm very rapidly going to abandon that process because I can do it better myself, at least that's my attitude when I was in the field. And then lastly, if we're not aligning the content and the answers to the work that we're actually doing and it is in the context of the work that I'm doing in the field, it becomes just another one of those technologies that it doesn't make my life easier as a technician, I'm just going to ignore it. So there is a lot of showing benefit to the users in the field as this is being rolled out to help generate that virtuous cycle. It helps me the more I will use it.

So I think next up, we were going to talk about broad and secure. As you said, there's been a lot of concerns about hallucination and sharing customer data, could we go into some of the security aspects of this?

Yes. So if these are the blockers, vendors are trying to help the companies to overcome these blockers. And from ServiceNow point of view, there are a couple of things that we're doing. On the organizational concerns, the security, data privacy point of view, this is what we are doing and implemented in certain cases. For general purpose, Gen AI, so if it is not specific to your company data, your information, then we're providing at the platform level, at the workflow level, like integrating with API, easily integrate to OpenAI and Microsoft AI right now available, and we're working on the Google Cloud AI as well. So with that, you can easily start using these technologies and experiment and build your solutions. On the company domain specific data. So this is where your data is not going to leave the data centre, ServiceNow data centres where your instance resides in. We're partnered with in media, and we built this Gen AI infrastructure, and we're building ServiceNow LLM, train on your data, and we're providing this Gen AI functionalities as out of the box into our products. Every product is providing these functionalities, runs in our data centres, train data based on your information, LLM and with that, really providing, okay, if you have domain specific, here is out of the box, here is the infrastructure. If you want to go generic, here's the secure way of integrating. But also on the user side as well. So this is something new for everyone. We're learning, we're experimenting as well. So we provide the central governors, governance from administration point of view, right? So the companies can control, who can use, what kind of questions can be asked? And the secure platform way of integrity third-party systems and using on our data centres, but also we developed the U.S. guidelines as well. So anything that we provide or our partner's provide will look the same. End users will have no confusion, what is recommended by Gen AI or generated by Gen AI versus in other ways, right? And again, we're also using the controlled phase releases. So we announced that we're going to do monthly releases starting in. September, we're doing monthly releases and every product team is contributing certain use cases, and we're working with the customers. So this is something that you don't want to do a big bank, pick a use case, make it successful, and apply the learnings from that experience to the next use case. And people always underestimate what overestimate what you do in the first release and underestimate what you can iterate on top of it, already tons of Gen AI related functionality being release. For example, this November release field service team, we're providing our work order summarization functionality as part of that monthly release, and there will be more the next month and the month after.

Well, going to monthly release cycle is great for customers, but a lot of internal work involved there. So my hats off to you for doing that.

Yes. I guess we talked a lot about what people can do kind of thing. So Brian, do you want to show an example from field service ticking point of view, how Gen AI can help technicians and how our customers using our platform, do what you're going to show now as of today.

Sure. I thought about AI and how it assist their technicians. If you consider that an adult reads at about 238 words per minute, that seems pretty fast, but when we look at all of the data that's associated with this work task that our technician is reviewing on their device, when you consider notes, comments, customer interactions, knowledge, content, articles, guided instruction sets, task instructions and troubleshooting guides, it just go service history, it goes on and on and on and on. There's about 50,000 words associated with this test. So assuming I read at a normal rate of 238 words per minute, take me about 4 hours to go through all of this stuff. And that's assuming that I'm going to glean any wisdom from it. So let's start a conversation with AI and ask simple questions and let AI mine the data for us. So like when was the last time this fan belt was changed? So rather than me going and looking at service history, it can then go through and say, "Here's the information that you're looking for," faster than I could possibly do this. And then this also helps me from a perspective of, if I'm not familiar with this machine or this particular process that I'm going to do right now because I'm relatively new, I can then go in and say, "Show me the procedure or give me the steps for replacing a return air filter." And then to show you the power of AI, I want to translate it into Spanish. Now I don't speak Spanish, so this would never be able for me to be able to do this no matter how much time I spend on it. But literally, in seconds, it went through mine that information pulled it in and gave it to me right at my fingertips. So if you think about what Bulent was talking about with the 2 Sigma problem, I now have a tutor in real time when I need it in the context of when I need it. So it does make it a lot easier for me as a technician to get at the data that I need when I need it, doing well and doing my job. And if you think about retention of that data, it leads to much better retention because I've done it literally at the time that they need to consume it.

This is just one example of what companies can do. And we're providing these with the monthly releases, these functionalities out of the box. And of course, as we showed the slide, you can integrate with the OpenAI, Microsoft AI builds your own solutions as well. But there are a lot of low-hanging fruit use cases that are already on our road map, for example, work order summaries coming customer debris going to come before you go to customer side. Get a quick summary history about that customer and not just the first program of what happened, the summary about it and then the knowledge recommendation, knowledge article generation, these are the things, really simple use cases and can be done really smoothly with Gen AI and we're more and more thinking about more sophisticated use cases. For example, in workforce optimization that we released 2 lease ago with Gen AI and now we can take a look at what are the skill gaps instead of managers reading, trying to find out if Jim Joe or Jane has a skill gap or not. Gen AI, can identify what skill gaps are and even take the action from there saying "Okay, here are the courses or let me one-on-one coat you on the skill gaps as well." There are tons of opportunities that Gen AI can really come here and AI come here and help. But again, let's do not forget the end-to-end process needs to be connected. If it is silo, then you're going to get siloed improvements, not to get the full benefit.

Yes. I love that the demo you showed was on a mobile device because that's obviously where field techs live and breathe. And he showed some data earlier from Service Council about mobility. I believe that our benchmark data shows that the average age of a field service mobile application is 3 to 5 years. And if you think about all the advancements in mobile capabilities and now the new phones coming out that have AI and Gen AI and then, boy, you people in the audience, if you're field service mobile apps are 3 years old or older, you seriously need to look at what is available today because you're probably really hampering the productivity of your field techs with some of those older apps.

Yes, field mobile application is the main interface for majority of our users.

So before we close out, I know we've got a lot of questions coming in from the audience. Everybody loves hearing about potential for AI but they really want to understand the business impact. So could you talk about some of the benefits and considerations that you're seeing from your customers to leave the audience with today?

Sure. Even though there are a lot of blockers and people are hesitant, but a lot of companies are using, trying and because they see the real benefit. I mean, here are some data, even the end users more than 3/4 of frontline professionals saying, "AI helps them." And nearly half of the organizations are already using AI and customer engagement one way or another. And the ones are using, they are reporting up to 45% value, they get an ROI from these investments. And again, the success depends on your approach as well. So there are a lot of ways you can take a look at like a best practice. But when I look at it, I guess, whether you work with us or you go on this journey on your own. There are a couple of things that we recommend as a best practice even before you start implementing. The first thing is address the data health. We all know the bad data, you're going to get bad results. So once you have your data clean-up, that you're at least foundationally ready to start experimenting. And now the risks. So that will help you to decide to go with ServiceNow or vendors like us or on your own and establish the guardrails. And if you remember the slide that I show what we have done, we're clearly helping customers to do this. We provide the guardrails. We're reducing, eliminating the risk for them, the data be secure and given them the tools to address their data health. And then once you are ready for that, pick a use case rather than boiling the ocean, pick one use case, make it successful and iterate and move to the next use case. Again, exactly what we're doing. We're doing monthly releases, we're picking the use cases, every application group, providing out of the box, working with the customers, learning from that, make it a GA and then moving to the next use case. So, we're applying this to our own processes, whether you go with us or you on your own, I recommend this as a practical way of approaching it.

Well, Vanessa, I know there's a lot of questions coming in from the audience. Do we want to start working our way through some of those?

[Operator Instructions] I'm going to go ahead and start with Benjamin's question, and they ask what impact does AI have on workforce management or field technician training?

The way we are approaching it, and Brian, feel free to add colour as well. From a workforce management point of view, as I said, we released our workforce optimization product 2 years ago and there is a version for that you can use for ITSM, and there is a version for call centre and field service management. It's all about injecting the AI to help whether it's forecasting. What would be the customer demand X months down the road? And then whether it's helping the capacity management, the shift management or inject in AI. We're also using maybe not Gen AI, but the traditional AI for simulations. So if this will be the demand, if this is a workload and if this is a territory, show me how it's going to perform. We're getting great feedback working with the customers on those simulation topic. And if you move forward, like, okay, you know the demand, you know your workforce, either the capacity management, and you might be at the decision that I don't have enough people or I have people but not the right skills. So this is where our workforce optimization and employee workflows, higher on-boarding, training comes in the picture. And AI, again, plays a big role, find skill gaps, one-on-one coaching, training and all this closed loop system comes back and when people gain those skills, then you can re-simulate and do the capacity plan. And again, or you might say, "Hey, I am not going to hire, and I'm not going to train at this point, but I want to start deflecting. I want to do self-service. I want to do remote assistance." And again, AI can play a big role on these areas as well. So especially right now, Gen AI is more text based, but the video and the image and the sound portions of this Generative AI is coming in the picture and that's going to play more and more role as well.

Yes. I think the other thing, Bulent like you mentioned, is using AI to help me focus. So if I'm a service manager, I have a day job. It's running the organization. I don't have time to be an analyst to go out and try and draw out which and all the content is missing, what training gaps we have. If we can use the AI and ML that's built into the system to say, "Focus your energy here. Here are some locations where you need more content or your technicians need more training." And then the other side of this is we cannot underestimate the power of replacing phone a friend with having a mobile device in my hand, while I'm standing in front of a machine, trying to figure out what's going on to be able to get that instantaneous feedback on, this is the direction that you need to go. So if you look at it that way, you can take an old [indiscernible] like me who had been in the field for 20 years, and it helps you take advantage of some of that tribal knowledge that's in your organization, but then push it out to the newbies that are coming up the ranks. But again, it's right in their hand. They have their phone with them. They don't have to go to a training class to get that knowledge and then forget about it the next time they go out in the field. So using AI to drive better results, make your technicians more effective, and make your overall organization more effect. And it also provides you the ability to provide enablement in bite-size pieces which we all know the smaller the enablement module, the faster and easier it is to consume and redeem.

Our next question here comes from Sarah and ask what specific change management tactics do you recommend for the successful adoption of an AI tool?

Yes. It goes back to -- maybe I can put that slide back, the best practices, right? Once foundational you're ready, your data is healthy, and you have the guardrails. It's all about picking a use case. It could be the use case you already have and you want to measure the improvement you're getting on top of it. In fact, AI is more about augmenting the existing processes or augmenting the users to make them more productive and maybe reduce the cost. So add the measurable metrics saying that, "Okay, we implement AI on this use case," "What improvements we're targeting, what improvements we're getting." And if you're not getting what you're targeting for example, in ServiceNow field service, we also released our process mining capabilities in the last release as well, KPIs usually tell you what happened, but it doesn't tell you why it happened. So with process mining, you can go down and see if the customer satisfaction is going not the right direction, what are the blockers and how can you improve that. In fact, some of the platform capabilities that we're working on in the future is really bringing AI to help you on that instead of human manager drilling down, figuring out what the bottlenecks are, AI can easily start telling you what the bottlenecks are by using this data mining tools and also create the workflow for you to automate that process to make it more efficient as well. The change management is always make sure the scope is well defined and again, don't make the goal line, keep moving and deliver that use case, make it successful before moving to the next one. Sorry, Brian, go ahead.

Yes. I was just going to add that Bulent. Number one, don't overpromise. You're not going to solve world peace and world hunger on day 1. And then probably the more effective way of doing it is involves a sub-segment of your users, especially technicians early in the process, get them involved in UAT so they become change agents. And isn't seen as something that's being forced upon them, but it's more of a, "Hey, I can actually use this." A technician talking to other technicians about how great this is. It is worth a lot more than any marketing material you can produce.

If I could just add one other comment at a slightly higher level, something we heard loud and clear envision of them a couple of weeks ago. There's a lot of scary headlines out there about AI is taking jobs away. And there are a lot of employees that are very nervous that when they hear about these aggressive road maps of AI and Gen AI what it's going to mean for them. And first of all, I think we're a long way from eliminating field service techs. So we actually have robots who can drive out into the field and do things. But there is some potential for impacting maybe some dispatching jobs. So the big topic that we heard is upskilling that if you are doing a lot of investment in AI, you need to let your employees know that you're taking away the tedious parts of their job so they can focus on the important value-added work and where necessary, you will be training and cross-training employees to take on more and more strategic work as AI comes in. So don't forget that as part of the change management. I personally think people being replaced by AI is a little overblown but employees are concerned about it, so include some sort of upskilling message in your change management plans.

And then let's go ahead and take one more question here and it's from Mike and they say has the demand for predictive AI dropped a lot since the Gen AI revolution?

I don't see that, to be honest, there is always demand for predictive AI. And again, going back to the business drivers, you're always trying to reduce the cost, right? So contribute to profit margins. So whether it's recommendations, next best actions or offers or even part recommendations. So it's great, it can reduce the truck roll and your cost. From that point of view, in some use cases, maybe offer recommendation Gen AI start playing a role, but overall demand for prediction is higher. Now people can see that can be done easily and they're demanding for it. I don't see the demand is going down, it's actually going up.

We have Conor who's speaking about the future, John, he asked, how do you envision the future of field service considering the ongoing advancements in AI and automation?

Well, this is an area moving real fast, to be honest. If you guys remember, I showed the state of the service survey results December versus September. There is a huge demand. And the state of Gen AI back in December versus now there's a huge difference as well. So a very fast-moving domain and people are already using in their personal lives. I have 2 teenagers. They're not cheating on their own work, but they are using it. And sometimes you use it to perfect some of my e-mails as well. So that kind of familiarity is helping adoption on the enterprise side as well. And we're doing monthly releases. So again, we're learning along with our customers, providing them the right tools, the use cases, but it is a really fast-moving area. And again, more and more video part of it, the image part of it, the sound part of it is going to come and we'll always find ways to incorporate and provide these capabilities because, again, business drivers better never done. So there is always room for improvement.

Okay. Well, with that, we are actually out of time for today's webinar, but I know we have quite a few questions. We weren't able to answer here live. So don't worry, though, I just want to say we haven't forgotten about you, and we absolutely will make sure to follow up with you. And since we have come to the conclusion of the webinar, a few quick reminders before we sign off for today. There will be an exit survey. And if you could take a few minutes to provide your feedback on the content and your experience by filling out that brief survey and know that a link to the recorded version of today's webinar will be sent out within the next 24 hours. I'd now like to take this time to thank our presenters, John. Bulent, and Brian for delivering an outstanding session and thank you to everyone for taking the time out of your busy schedules to join us for today's live webinar using AI to enhance field service technicians productivity and efficiency brought to you by Technology & Services Industry Association and sponsored by ServiceNow. We look forward to seeing you at our next TSIA webinar. Take care, everyone.