ServiceNow, Dynatrace, And The Future Of End-To-End IT Autonomy

[00:00:04] What if IT operations could spot issues before customers even feel them? And not only that, understand impact instantly and take action without waiting for the war room to fill up as that P1 hits. Well, today on Tech Talks Daily, I'm coming to you live from Dynatrace Perform here in Las Vegas, where the conversations on the show floor, they are moving fast.

[00:00:30] And the expectations around AI automation and resilience, they feel higher than ever. And this is where enterprise leaders, engineers and platform teams are all asking those hard questions about how IT actually needs to work in a world where downtime is no longer tolerated, all at the same time while complexity keeps piling up. So today we're going to go right into the heart of this challenge and I'm going to be joined by Pablo Stern.

[00:00:58] He's the EVP and GM of technology workflows at ServiceNow. We're recording today here at Dynatrace Perform just a couple of hours after he stepped off the main stage to talk about ServiceNow's deepening partnership with Dynatrace. So today we'll talk about that multi-year collaboration that was announced last year, what that means for enterprise IT teams and look beyond the headlines.

[00:01:24] Talk about observability and workflows, how they can come together to shorten the path from alert to action. And explore why autonomous IT operations have been so hard to move from theory into practice and where human judgment still matters as automation continues to scale.

[00:01:44] So if you are responsible for IT operations, service management, security, reliability or business impact, which I hope would be most of you listening out there, today's conversation will help connect the dots between those AI ambitions and operational reality.

[00:02:02] So buckle up and hold on tight because I'm going to beam your ears all the way to the show floor conversation here in Las Vegas and hear directly from Pablo and why this partnership between ServiceNow and Dynatrace is focused on outcomes rather than hype. So let's get him on now. As a warm welcome to the show, can you tell everyone listening just a little about who you are and what you do? So my name is Pablo Cern. I'm the EVP and GM of technology workflows at ServiceNow.

[00:02:33] And ServiceNow, our mission really is to enable AI and make it work for people. And in my role, I really focus on two core archetype personas, the CIO and the CISO. And what my team does is we build technology for CIO in their organization and for the CISO in their organization.

[00:02:57] So security, risk products, products for like ServiceS, the command center, asset teams and others. Like we ultimately build them. We try to bring them all together. And we're recording this at Dynatrace Perform in Vegas. You were on stage this morning talking about your partnership or ServiceNow's partnership with Dynatrace, which started last year, deepened your strategic collaboration to advanced autonomous IT operations.

[00:03:23] But for listeners who might not have followed that announcement closely because it was right in the heart of event season, so much going on. Tell me a little bit more about it. What problem were ServiceNow and Dynatrace hoping to solve together in enterprise IT that made a deeper multi-year partnership so necessary? Tell me about it. It's a great question. I've been at ServiceNow for almost 10 years. And we've been partnering with Dynatrace basically since I joined. So this is a longstanding partnership.

[00:03:53] And I will always start by saying the partnerships that matter always start with the customer. So the starting point for me on anything is can we solve a joint customer problem? Can we deliver on that promise? And if we can, and if our customers are asking for it, let's go focus on it. And so, you know, we've had a strong partnership, a lot of joint customers integrating with our ServiceNow, ITOM, AIOps capabilities, and the Dynatrace platform for a long time now.

[00:04:21] But I think one of the things that's changing in the last few years as the technology landscape is really shifting and the underpinnings of how we deliver technology is really transforming. There's a brand new opportunity to deliver value to customers. And what we've seen in the last few years is a lot of the building blocks have been coming to bear in a technology market.

[00:04:45] And for us at ServiceNow, and especially as we think about outcomes with our joint customers in Dynatrace, we see that these building blocks can actually enable our customers to do things that they could never do in the past.

[00:04:58] And so this idea of an autonomous future, autonomous IT, autonomous operations is really driven on a very simple notion for me, which is if you think about the evolution of AI in the last few years, we went from a world where AI was really accelerating insight.

[00:05:18] You could summarize something, you could create a poem, you could draft a knowledge article, you could take these like very like take mountains of data and like get to like some key insights. And that was a necessary building block. And then over the past couple of years, we started in layering action on top of that, driving workflows, being able to do tasks. I'm going to actually, I'm going to, I'm going to post that knowledge article or I'm going to write a communication.

[00:05:47] And that was sort of, that was a second step in the journey. And what really excites me as we think about where we're going with Dynatrace is the next step is you take those building blocks. You take all the data that exists, you can derive data from, you take these playbooks and these actions that you could then drive the deterministic view of what's happening in environments. And you connect those, and then you start enabling the technology to be able to autonomously drive end to end action. So you're no longer going from like a single task, you're actually driving system level actions.

[00:06:17] And the only way that you can actually do that is by connecting the ecosystem of different technologies. And we have so many joint customers that really want those outcomes. They're looking for how can we get to a world with no outages? How can we reduce our time to resolve issues? How can we get better adherence for SLAs? Those are common asks of our customers, and we know that together we can actually build that for customers.

[00:06:44] I think when many listeners and many organizations, they often talk about autonomous IT operations but struggle to move beyond theory. So in practical terms, how does combining Dynatrace, observability, and ServiceNow workflows, how does that change how incidents are detected, understood, and ultimately received every single day? Tell me more about that behind the curtain. I think that there's, as I mentioned, like a few building blocks that even if you take AI out of the picture are necessary.

[00:07:13] And those are actions that our customers are taking today and are going to build and have a virtuous cycle as we start identifying. So as an example, in the ServiceNow world, having prescriptive workflows, which are going to be very unique to an organization, will help you drive relief, will help you get to resolution faster. So it could be about who do you page, how do you get to them, and making sure the right people are on the call.

[00:07:39] It could be about failing something over or removing storage, like full space on your disk. There are a lot of different actions that can be taken. Now, if you take those building blocks and then you go on the Dynatrace platform from an AI-powered observability perspective, and then you can actually deterministically say, this is what's happening in an environment. Like go really understand what is happening that gives you confidence and trust.

[00:08:07] You can then start connecting these two use cases and trusting a machine to take them. And I think that that is the unlock that we're working towards, which is nobody out of the gate is going to say, I'm going to trust a machine to fail over my P1 production system to something else if it has some inkling that there's a problem.

[00:08:28] But it'll probably take the insight and it'd be happy to then run a deterministic playbook if somebody was willing to go and approve that change. But those different pieces over time, I think will become more automated. And the connective tissue for me is today, the world often in the standard of practice is not fully digitized end to end. So there's, you're playing telephone across multiple people, their manual steps or tribal knowledge.

[00:08:55] All those pieces create friction that an agentic system won't be able to go and drive. So if you can digitize that entire process, even if you're trusting the human along the way for each of the steps over time, you'll be able to trust machines to take some of those steps or to elevate to a human who will then approve. And that's, I think, the world that we're building towards. But those foundational pieces that I talked about are the necessary components.

[00:09:22] And there's enough technology now in place to be able to build those and to give enough insight and enough of a breadcrumb to actually start taking those actions. And root cause analysis, that's still one of the biggest time drains for IT teams. So anything you can share on how automation between the platforms here shortens the path from alert to action and where human judgment still plays a role? Anything you can add to that? Yeah. Yeah.

[00:09:49] So I'll give an example of this, of a place where we're partnering with Dynatrace and how I think that can help sort of produce mean time to resolve. So look at a world in which an event comes in. Now, there are tons of alerts, tons of events coming to the system and you're trying to figure out and prioritize and understand what's happening in those events.

[00:10:12] And so in a world where you're not waiting on a human to go and sort of triage through the list or prioritize them or try to drive correlation. But an agentic system can then go and do additional forensics, go from ServiceNow and query into Dynatrace and then try to look at what is happening, understand what the root cause of that event is, potentially, you know, understand what the blast radius, what's impacted, right?

[00:10:39] Try to assess both priority, impact, and potentially source. You then start removing potentially some of those multi-step processes that we hear often about where like people are on a major incident bridge and every team is like, it's not me, it's not me. And then, you know, you spend the first hour just trying to figure out where the cause is and you can potentially shorten that cycle time by doing some of that work ahead of time.

[00:11:07] And again, will it happen and solve it every time? No, it won't. But if you can either reduce the amount of time it takes to do the analysis and the forensics, if you can be right 20, 30% of the time, you can save hours of time in that diagnostic. And even in a world where like the, a lot of those steps are still taken by humans or relief, even the resolution pieces are done by humans.

[00:11:33] You can reduce the time that somebody sees an impactful event pretty dramatically. And so that's how we think about it. And that's a place where in partnership between ServiceNow and Dynatrace, we can actually deliver some of those outcomes and bring the best of the Dynatrace platform, the best of the ServiceNow platform to get to faster resolution. And self-healing systems sound incredibly compelling, almost remind me of the Christine movie all those years ago.

[00:12:00] But it can also feel risky to enterprise leaders. So what safeguards or design principles are essential to maintaining that trust in automation remediated at scale? Yeah, I'll say that I think like truly like full self-healing systems, like that's a ways away. Like we'll build towards it, but it'll take some time. What I think we're going to see is first relief.

[00:12:27] So I think as, and trust is such a big part of this, like you may trust a system to do a failover because you're already doing some of those fails. Failover load balancer, you know, if they're like, there's a bunch of things that you could do. And oftentimes a lot of those are actually scripted outcomes that are delivered. They're just, you're triggered by a human. So there's, there's a lot of places where the machine can then say, this is a suggested outcome.

[00:12:55] You still do the steps by a human. And so what, what I think we're going to see is you'll see a little bit more of that type of stuff. And I'll call it the low hanging fruit and it's around relief. Like, Hey, not yet. Like I'm actually going to figure out where in the code, the issue is back out the change, redeploy the change, get it all done into production and push it out. Do that all autonomously.

[00:13:19] Like that future, future that you'll build towards, but you know, is not like the future that you're going to see in the next six months. Um, but you can say, well, let me first just give relief so that like we get back up and running and then the team's going to do the diagnostic and somebody is going to for sure be spending time working on like, can we drive more autonomy in those higher level outcomes? But I think that that is, that's a destination of future.

[00:13:46] And I think of, you know, the AI world is like a generational change. Like this journey is not going to be done in a year or two. Like it will be decades in the journey driving change much like we are with cloud. Now we're going to make tremendous amounts of progress in the next few years. I think it's going to, it'll change how we work. But I also recognize that these changes take a long time and it's not just the technology, it's the people in process also that have to come along for the ride.

[00:14:13] Do you think we'll have a, an AI member like a virtual Rick, AI Rick attending the change advisory board in a few years? Do you think maybe? Uh, I know we will. Yeah. We will have, they may not look like Rick, but, uh, I can tell you that we will have members of change advisory board that are digital, that are helping get to those changes. I can guarantee it.

[00:14:38] And here at Dynatrace Performer, you've been referencing heavily agentic AI and intelligent automation. How do you distinguish meaningful autonomy from simple task automation? And why does that distinction matter for IT and service management teams? Yeah, I think it's a super critical one. And I think it's, I think that distinction is how I see the evolution of the autonomous world.

[00:15:02] Like the three layers are, I refer to it as step one of the maturity journey from like 22 to 2023, 24 was around insight and just getting those, you know, the needles in the haystack and understanding what it was. And it could be a summary of an issue. It could be bringing together knowledge. It could, could be enabling a team, but often like for human loop activities. That was step one.

[00:15:30] Step two was tying that to the action. And that that's where those task automations came in. Like, Hey, I can provide relief by failing overload balancer. I can page a team and take an action that brings the right people onto the call. And each of those building blocks I often will refer to as like, that was the agentic world that those are the agentify tasks. Like you have an agent that does task routing.

[00:15:58] You have an agent that drafts resolution notes or creates a KB article. And step three in that journey is where you end up driving the autonomy. And what that looks like is you take a lot of those agentic outcomes that are really tasks and you make them a system and you make them a system that can actually complete a job to be done.

[00:16:21] And that system leverages all the insight layer, which we will refer to as skills, the agentic layer, which are some of those tasks on top of the skills and then drive the autonomous outcome. So we're building towards that third stage. I would say even in the second stage of that journey, the automation, there's still a lot of work to be done by us and by our customers to put all those pieces in place. But these are obviously necessary building blocks.

[00:16:48] And watching you on stage today with Dynatrace and ServiceNow, it seemed to me from the audience point of view, of course, and me sat there, is it's not just a technical partnership, but operational too. And both companies using each other's platforms internally. So what insights have you gained from being customers of each other? And how has that helped shape your joint roadmap maybe? Yeah, I often say our internal customers. So ServiceNow's digital technology team, that's our IT organization.

[00:17:17] Is our best and toughest customer. They push us the hardest. They take our products before they're baked. They give us a lot of feedback around what's working and what's not. They don't pull any punches. And they ultimately have the same interest in mind. They want ServiceNow to be successful and they will challenge us and they will tell us and they will not hold back on it.

[00:17:40] And so for me, that's super important because they will always take our technology before we bring it out to customers. And so they really represent like the first pass of delivering really quality outcomes. So in a partnership like Dynatrace is the fact that our internal organization is using both technologies gives us a direct line to people within the company that we can partner with to make sure that the outcomes are actually being delivered.

[00:18:10] And if they're not, we can hear from them directly. And so it's incredibly powerful. The fact that we do this internally, the fact that Dynatrace is doing the same thing will deliver outcomes because every organization is different. They'll find some things we won't find. We'll find some things that they won't find, but it'll help us build. And I say this also recognizing that we have a bunch of customers or design partners are working also in parallel early. So you want to have as many of these as possible to understand like what the differences in requirements and environments.

[00:18:40] But like our internal teams using this is one of the best sort of canaries in the coal mine for the technology working. And if we zoom out a little from an industry perspective, what are the common operational challenges that you're seeing across enterprises right now from that traditional ITSM and ITOM approaches are no longer equipped to handle alone? Because it feels like there's a lot of changes going on at the moment. Yeah.

[00:19:05] So I'll say something that I've seen that's been pretty interesting to see, which is if you think about the outcomes an organization is trying to drive, if you talk to a CIO or CHRO or chief customer officer, all of which use ServiceNow technology or may not even maybe prospect customers, the outcomes that we hear that they're seeking are not different. There are accelerations that they're expecting, but the outcomes aren't different.

[00:19:33] So as an example, like a service desk and the service desk, they're expecting that they can take any questions are coming in from employees, be able to handle them, route them, and get the employee on with their day, working on the stuff that they need to do their job, not an IT ticket or an HR ticket or whatnot. Now that said, what we're hearing from customers is like, we want to do that. But now with AI, we want to really rethink how we can deliver that outcome.

[00:20:00] We can deliver that outcome more efficiently with a better experience, higher CSAT. And so really the outcome in and of itself is still a similar outcome, whether it's a service desk, whether it's command center, whether it's your asset management team, whether it's the EPMO or the PMO team that's working on like your planning. But the how is what's ultimately changing. And that how is meant to deliver the outcome more quickly, more efficiently with a better experience.

[00:20:28] And that's the journey that we are all on. And so for us at ServiceNow, that does require us reimagining the how of how we build the technology. Like we are not in our mindset of incrementalism to build AI just on top of what we have and assume that's going to be the answer. We really are looking at like if we want to deliver a 10x value prop expansion, like how does AI deliver that? What is the experience that we need to do?

[00:20:58] How do we leverage the pieces that we have, the system of record, our workflows to go drive it, make sure that we are also reinventing how we deliver that outcome. The good news with ServiceNow is we've done probably over a dozen AI acquisitions in the last eight years. A lot of it was came from brainchild of early research and LLMs and models. So we've been we've been spending a lot of time.

[00:21:24] We also have a lot of feedback from customers, but it is a journey that we're building towards. And again, I think the outcomes are are clear known. The fact that we want to do them 10 times better is what we are spending a lot of time internally to work and reimagine how we deliver it. And if I was to ask you to look into my virtual crystal ball, look ahead, assuming organizations and leaders listening get this right.

[00:21:47] What changes should listeners expect to see in the role of IT teams right in their own experiences across corporate America over the next few years, especially as operations become more autonomous? They become more resilient. What should they be seeing? How can they measure those improvements? Yeah, we started this conversation around customers.

[00:22:06] And so I always come back to the customer and, you know, being here at the conference today, I've run into probably a half dozen customers that I've I've known on the service outside who are here at Dynatrace perform. And I often think like, how do we make our customers successful? Like that to me is our mission, like we are building technology, but technology with the purpose of making our customers successful.

[00:22:26] We want our customers to use ServiceNow, to use Dynatrace, to use our technologies in seeking an outcome and being able to deliver that and be able to be successful for their organization, for their customers, customers. And so in that vein, as I think about like everything that we're doing, I come back to like for everybody who is in the audience today and, you know, we ask folks to raise their hand if they were a ServiceNow customer in Dynatrace audience.

[00:22:56] And there's probably three quarters of the audience with ServiceNow customer and Dynatrace customer. What we really look to do is they're going to have really high expectations of what AI can deliver. And those outcomes will translate to higher SLAs, delivering more efficiently, reducing time to resolve issues, shifting left and being able to find things before they ever hit environments. And so our mission is to be able to deliver those.

[00:23:22] And then the measurements around that mission, I don't think that they necessarily change. MTTR, SLOs, SLAs, you know, what are you driving in your air budgets? Like those will continue to be them, but the bar is higher in terms of what those things are. Your SLAs are going to increase, your availability time needs to be higher. Like all those things need to happen in order to deliver the outcome. So I think that's where we're going to see the evolution.

[00:23:48] And then it's our job to deliver on the how so that out of the box, the technology helps enable that. And something that's really critical in those is the technology pieces fit into both the organizational change that needs to happen and the process evolution that needs to happen at the customer in order to deliver on that promise. And one of the things I think that stands out at Dynatrace Performance, in fact, every tech conference I go to, is the power of the community there.

[00:24:16] Those conversations that you have, not just with customers, but with partners and also random people on the show floor that you would never be in the same room as. I'm curious. I know it's only day one that we're recording this, but from all the conversations you had, any themes, anything that's got you thinking that you'll be thinking about on that flight home? Yeah, absolutely. And this is why I love these conferences because I'm spending a lot of time with joint customers. I'll say a couple of things that came out.

[00:24:40] Like we were talking to a customer bank after the session and I'd mentioned this concept on stage around a pre-flight check. The idea that because a lot of our customers use ServiceNow for change and because change is often the leading cause of issues and outages in your environment.

[00:25:00] If we could help through ServiceNow and Diner Trace shift left and bring as part of the change process, understanding of what the blast radius of potential impact of that change is, a risk assessment, and actually filter that in earlier on. We can actually start potentially reducing the number of outages or at least putting those checks and approvals in the hands of a team that can actually go and really make an informed decision.

[00:25:30] And this bank customer came to me and said, we're trying to build this ourselves right now. Like we're weird between ServiceNow, we're pulling stuff out of the system and we're doing like our own, but like they're trying to do that. And so for me, as a technologist, I want to deliver that out of the box. I want to partner with this bank and I actually want to build that outcome together. And that's super powerful for me.

[00:25:55] And again, coming back to what I said around customers, like if we do our job right, we're helping solve for our customers. And that customer would say, you're giving this, you and Diner Trace are giving this to me out of the box. I have the outcome. So I've had a couple of those conversations around the future pieces. And the other thing I will say, every customer I spoke to raised their hand and was like, sign me up for design partnership. We will do this together. Like we want to partner with you. We're building this future together.

[00:26:23] And we think that between ServiceNow, Diner Trace and us, we can do this. So that's super reassuring because it tells me that there's a there there. The customers actually want this. I think that's a powerful moment to end on. But before I let you go, anyone listening want to find out more information, keep up to speed with announcements from ServiceNow, et cetera. Any way you'd like to point everyone? Yeah. So I'll say a couple of things. If you're a ServiceNow customer, reach out to your sales team.

[00:26:50] Like my team works really closely with our sales or pre-sales or other teams. Like they have all our enablement around where we are, where the roadmaps are. I would encourage ServiceNow customers, our user conferences and then world forums, which are in site day or two day long events, like bring a lot of content.

[00:27:11] And if you are a customer, you should make sure that you're using our briefings because that's how you get informed on your products, where you are, where we are and what the roadmap is. If you have any questions and you're not a ServiceNow customer, sales at ServiceNow.com. Reach out and we'll find the right person who the right contact is so that you know how to get started on the journey with us. I'll have links to everything you mentioned there. And I think very often people will hear that it's a tech podcast or a tech event. We're talking about AI, Agenda KI.

[00:27:41] But as you've said throughout this conversation, it's not even around the tech. It's about solving problems for customers. And I think that is a powerful moment to end on. But just thank you for joining me today. Thank you. One of the things that landed for me in this conversation with Pablo today is that autonomous IT is not just a simple switch that you can flip. It's a progression. Insight first, then action. Only then can systems coordinate those actions with confidence.

[00:28:08] And throughout that journey, trust, process and people, they matter just as much as the technology. So when we're hearing about partnerships like ServiceNow and Dynatrace, they work best when they start with real customer problems, not technology first. Only then can you reduce the time to resolution, improve service reliability and prevent issues before they escalate.

[00:28:34] Which is so much better than just a bunch of abstract goals. We're talking measurable outcomes that show up in SLAs, customer satisfaction and the day-to-day sanity of IT teams. And being here at Dynatrace, surrounded by customers, partners and practitioners all comparing notes, it really does reinforce the fact that something is changing here. Teams are no longer asking whether AI belongs in operations.

[00:29:02] They're asking, how can we use it responsibly? Where can we apply it first? And how can we build confidence in automation step by step? So I'll include links to everything Pablo mentioned, including how to learn more about ServiceNow's roadmap and how this partnership with Dynatrace continues to evolve.

[00:29:22] And if today's conversation made you reflect on how your organisation handles incidences, change control and resilience, I'd love to hear what stood out for you. So as you think about the next outage, the next change window or the next escalation call, what would it take for your operations to move from reactive to truly autonomous?

[00:29:47] You've heard from me today. You've heard from Pablo. I'm sure you've got lots of ideas, experiences and insights too. So please pop by techtalksnetwork.com. You can leave me an audio message, send me a written message or connect with me on socials. Let me know. But that's it for now. Time for me to hit the show floor again, but I'll be back again tomorrow. Bye for now.