In this insightful episode of the Tech Sessions podcast, e360's Field CTO Barry Hen sits down with Roy Douber, Sr. DevOps Expert and Observability Lead, to discuss their key takeaways from Dynatrace Perform 2025. The conversation explores how Dynatrace is evolving its platform to meet the growing demands of modern IT operations through AI advancements, predictive analytics, and comprehensive tool consolidation.
Roy provides a unique analogy that captures Dynatrace's approach—likening it to a chef with all the right tools in the kitchen, organized and ready for use at exactly the right moment. This holistic approach reflects how Dynatrace is positioning itself as more than just an observability tool but as a comprehensive platform that serves various stakeholders across the organization, from developers to C-suite executives.
This episode offers valuable insights for IT leaders considering their observability strategy, as well as teams looking to enhance their operational capabilities through platform consolidation and AI-powered analytics.
1. The All-in-One Platform Approach Delivers Real Value.
2. AI is Evolving from Reactive to Proactive
3. Tool Consolidation Offers Both Hard and Soft Cost Benefits
4. Developer Productivity Can Be Dramatically Improved
5. The Observability Market is Expanding Rapidly
6. Expertise Multiplication Through Consolidation
[00:00:36] Barry Hen: Hello, everyone. Welcome to another episode of e360's Tech Sessions podcast. Today, we're going to be discussing Dynatrace Perform 2025. I'm Barry Henfield, CTO here at e360, and I'd also like to introduce Roy Douber, who's here with me to further discuss what we learned and found exciting. Roy?
[00:00:54] Roy Douber: Hi, my name is Roy Douber.
[00:00:55] Roy Douber: I'm the observability leader here at e360. We had an awesome time at Dynatrace Perform 2025, and we wanted to give a short recap of everything we saw that was, was a really exciting conference, and we really enjoyed it. And there's lots of features to discuss, and we want to give you a brief overview.
[00:01:17] Barry Hen: Excellent. Yeah, there's a lot that I found exciting as well. Roy, what were some of the highlights, and what did you find most exciting that you want to talk about first?
[00:01:26] Roy Douber: So I think, I think there, there's a lot of excitement because I think with, the onset of AI, we, are seeing that the addressable market for some of these tools has grown tremendously.
[00:01:44] Roy Douber: The code that engineers are writing is being created at a faster pace. And so as you release more quickly, you need better checks and balances to ensure that everything you're doing and the code that is distributed across your application is working in the way you'd expect. And so, and I think Dynatrace is really executing well on its vision of it's kind of like, how do you, what's the analogy here?
[00:02:18] Roy Douber: The analogy is, like a chef that has all these tools in the kitchen and the right place, dynatrace is ready for the engineers to have all the tools they need at the right place at the right time to triage an outage. And so I think, I think that that's, that's what makes Dynatrace have such a kind of holistic feel to it.
[00:02:44] Roy Douber: It's really,
[00:02:45]
[00:02:45] Barry Hen: yeah, that's exactly right. And I think, what we learned about some of the new announcements at perform, what I found particularly impressive is they're really rounding that out.
[00:02:55] Barry Hen: They're adding in things like security modules, CSPM [00:03:00] KSPM. So Kubernetes security, posture management, and, the AI for AI, I mean, Imagine in the year 2025, we're getting to see AI as a topic of conversation at a conference. Who's surprised to see that? So, what else within the AI space did you learn about with, Dynatrace this time around?
[00:03:24] Roy Douber: Yeah. So, you know, if you've used Dynatrace or maybe you haven't, I'll just assume we haven't , they were kind of one of the originators of using AI in the Operations space with their Davis AI. And that was prior to generative AI. So they had their causal AI and they continue to make improvements in this space.
[00:03:47] Roy Douber: They released. An enhanced version of Davis that basically not only does it give you the causal AI that you've had previously, but also it tries to give you answers based on, you know, language models to the problems that your system is experiencing in real time. So it really brings both of your alerting storyline of the company to bear while also trying to give you the knowledgebase of the internet, which is generally what engineers tend to do anyways. It's like, Oh yeah, you know, you, you know, the meme with the Google picture, here's what I do for work every day. And everybody goes to the search bar for Google. Well, it seems like they've tried to bake, bake, you know, a copilot of their own into Dynatrace.
[00:04:40] Roy Douber: And it gives pretty good answers from the breakout sessions that I've been in. I was like, I was pretty impressed and you know, all of that leads to a reduction in meantime, meantime, the innocence, meantime, the repair. And, you know, moving the, you know, reducing the amount of time you spend in an outage scenario, which is what businesses bought the product for.
[00:05:03] Barry Hen: Yeah, that's definitely one of the things that I found particularly impressive was seeing how Davis can be leveraged in order to perform a root cause analysis in real time to when the actual outage began, which is almost unheard of in operational worlds today. Were there any other announcements, Roy, that you found to be, extremely interesting or, helpful within the current climate of technology?
[00:05:31] Roy Douber: Yeah. So I think, I think in the AI space, obviously, what I found interesting is everybody's trying to leverage these large language models, but there's less interest in capturing what are the questions that are being asked by the users that are using those. And that's what observability for AI will provide you.
[00:05:53] Roy Douber: and it will even try to tell you whether the answer was good or [00:06:00] bad. Per the, per the customer feedback. So say you get a question often, you're going to be able to see it in Dynatrace's observability and say, well, let's look at this answer. This answer looks like it's too shallow of an answer, or it's too, it's not, it's not a good enough answer.
[00:06:21] Roy Douber: So let's create the knowledge base so that now, when Davis comes back the next time, you know, we know exactly the answer that it will receive and they're going to receive a good answer that's going to help them move forward. The other thing that they were talking about, and you know, I've yet to see this in practice, you know, we had some breakout sessions, but really moving the conversation from a reactive to a proactive...
[00:06:48] Roy Douber: leveraging AI.
[00:06:50] Roy Douber: So Dynatrace is continuously looking at the data coming in and because you have all the data in context, it's going to let you know, not when it's an outage, but prior to the outage, when it subsumes there is an outage coming. And you know, that's kind of the Holy Grail. We've always talked about it.
[00:07:09] Roy Douber: Or how do we move from our own set of alerts that we've created, our curated data set that we've created to something that can come to us, like before it all happens.
[00:07:22] Roy Douber: I don't want the outage at all. I want to prevent the outage. And so they've, they've modeled a way to try and figure out when the outage is coming, like looking at, looking at those, info messages or early warning messages and trying to correlate them and causate them together to lead us to an answer.
[00:07:47] Roy Douber: So that we can take action prior to everything failing or degrading, which is kind of the, the Holy Grail. So I
[00:07:57] Roy Douber: think
[00:07:57] Barry Hen: So speaking of Grail, Grail is an actual Dynatrace product for anybody that's not aware. Can you talk a little bit more about some of the advancements over the last year plus since Grail has become a Dynatrace platform?
[00:08:14] Roy Douber: Yeah. Yeah. So I think, the platform needed you know, a logging solution, they released Grail, I believe it's, it might be like a year and a half ago, and it is, it now competes with the best tools in the industry from a search perspective, it has its own query language, which, you know, I don't necessarily love.
[00:08:38] Roy Douber: You know, there's plenty of languages that could have been used, I would say that are commonplace, but they also have the AI engine that helps you generate the queries, right? Like, so again, leveraging AI within the platform to make the user experience improve. And I think, I think that's one of the, kind of the, the best [00:09:00] things about Dynatrace is the user experience is very, it has a luxurious feel to it.
[00:09:05] Roy Douber: You jump, you have all your tools in one place. You can jump from. You know your security to your APM to your traces to your database transactions all within one system. And it's a very clean experience.
[00:09:22] Barry Hen: Yeah, I think that's absolutely fantastic. And that's definitely something I found particularly impressive as well.
[00:09:29] Barry Hen: And now that we've looked at everything it can do from the observed side with predictive analytics and being able to achieve root cause analysis and really short order. the other part that they've really been doubling down on an investments is the automation engine, which will help you do auto remediation and things of that nature.
[00:09:49] Barry Hen: What did you learn at perform that kind of related to their automation engine and some of the capabilities there?
[00:09:57] Roy Douber: I think again. It's a very easy to use platform. You basically have a bunch of boxes and the flow that you can execute against. You can execute approvals against, you can execute API calls, you can run Ansible playbooks.
[00:10:13] Roy Douber: There is quite a few integrations out there that they've baked into the product where they might not be a first class citizen within Dynatrace necessarily. So like, they're not trying to be an automation platform for spinning up resources. So they say, Oh yeah, you can integrate with Terraform.
[00:10:31] Roy Douber: You can integrate with Ansible. You can integrate with whatever platform of choice you have internally. And there's quite a few of those. The other thing they're not trying to do, which I found interesting is they're not really trying to go into the you know, incident space as much, so they're, they're relegating that pager duty to a Service Now to, X matters.
[00:10:58] Roy Douber: so very tight integrations with that as well. And I think it's, again, everything that you need in one place to take a set of alerts from something that you view on a day to day basis and automate your work away. And that, you know, follows all the principles of dev ops, like reducing toil, not doing things more than once.
[00:11:24] Roy Douber: If you don't have to.. Minimizing the time it takes to resolve issues, knowledge transfer... all within one platform.
[00:11:34] Barry Hen: Yeah. I mean, you started mentioning about developers and the developer experience. can we talk a little bit about what they'd have announced in order to kind of bring more to the shift left approach of how to enable more in developers hands?
[00:11:49] Roy Douber: Yeah. So, I think it's not even just developers, I'll touch on developers, but I think it's the entire suite of users at an enterprise, right? They're really [00:12:00] pitching a platform for everything. And I think it's really, really interesting. It's kind of like, call it like a reverse snowflake or something like that, where, rather than developing the data platform first, they develop the use cases on top of the data platform and they continue to expand their offerings.
[00:12:17] Roy Douber: So they now have, you know, an offering for a security professional and offering for an operator and offering for an engineer and offering for DevOps and offering for the C suite, and offering for the director level. So, you know, they're looking at business metrics. And business events and integrations with SAP and products that really, really, are close to how the businesses report their revenue, top line, bottom line, how they calculate it, how they measure it in real time.
[00:12:49] Roy Douber: Very, very interesting approach. And I think from my vantage point, it's a winner. it's, it's a winning approach. Yeah. It, yeah. Sorry.
[00:12:59] Roy Douber: To jump into the dev side, they acquired Rookout, I think it was a year ago. And that's the, the live debugging tool. And there's quite a few. Small contenders in this space, like we've watched a demo of Lightrun in the past.
[00:13:20] Roy Douber: There's a few others in this space.. very interesting and very useful, especially when you're dealing with distributed applications. The idea here is that you put a probe into the development, the running production environment, and then you can inspect the code in real time and solve the problem in real time, you know, with, with the data right in front of you, it makes it much easier than trying to say, okay, I'm going to go into staging. I'm going to go into my development environment and try to replicate this. Now, in this case, you can, you can almost see everything in production and take it back and either work on it, right, right there live, or you can, you know, then take your findings to the developer or as a developer, develop it yourself.
[00:14:15] Roy Douber: I think that's, that's one key addition. So live debugging, and a lot of advanced analytics that are specific to the software engineers that are working on, you know, trying to increase their productivity. So one of the key metric that they gave, I think, Dell was there giving a presentation and they're like, our engineers were like before Dynatrace and all the tooling 30 percent utilized.
[00:14:46] Roy Douber: So they're only writing code 30 percent of the time. And with all these new features we introduced, we were able to bump that up to like 70 or 80%, which is way above industry standard.
[00:14:58] Barry Hen: [00:15:00] It's definitely an impressive metric and it's definitely, what's, what I find more impressive is that there's a platform out there that could enable that, with pretty much ease.
[00:15:12] Barry Hen: Cause Dynatrace is also extremely easy to deploy, quick time to value, and very easy to make pervasive within any environment in short order.
[00:15:23] Roy Douber: There was one slide that was really interesting that they showed. During our partner summit, they basically had like a company that was the before Dynatrace and after Dynatrace image, and they showed the kind of the tool rationalization graphic.
[00:15:40] Roy Douber: I mean, we, we talk about this all the time with our clients. They went from 36 different tools to mostly Dynatrace and some tools where it made sense. And, and that, that was a really, if I find it. When we do release this episode, I'll have a link in the bottom where you could view those slides. I thought that that was a really cool, cool slide.
[00:16:06] Roy Douber: So if you're looking at rationalizing your observability story and trying to reduce your costs, there's no better way than saying, okay, we can remove, There's 80 percent overlap between, you know, 10, 20 of our tools and Dynatrace. Let's get rid of as many of those as possible and try to fill in the gaps where we can.
[00:16:30] Barry Hen: Well, that definitely is because there's oftentimes two different cost models. There's the hard costs, meaning Dynatrace and all of these other potential tools cost me X and Y amount of money. But then there's the soft costs, which is often overlooked. And those are the costs of the operational overhead.
[00:16:48] Barry Hen: How many people does it take me to operate and manage this platform? What is the time it takes my team to learn how to use this tool and extract the value of the platform? And both of those are highly important because sometimes the soft costs could be more expensive than the hard costs, especially when you start looking at observability and everything surrounding observability and productivity.
[00:17:13] Roy Douber: Yeah, when you have, when you have one user that's managing, you know, one user per tool managing, 36 tools and you could bring all that knowledge or most of that knowledge into one platform, that's a really powerful, adder to the business, right? Like you're multiplying the value of the tools and you're multiplying your expertise,
[00:17:36] Roy Douber: in a given tool set. And I think that's often overlooked, like rather than, you know, rather than adding value to the business, you've got all these, this, this massive fragmentation of tooling everywhere. And, it's very hard to get to, 50, 60, 70 percent [00:18:00] adoption per tool. Whereas if you consolidate to one set, to one tool, all of those engineers can come together and try to increase the adoption of that tool, build around that tool, quickly execute when, when the time comes to execute, right? There is less of an argument of we should do this here versus we should do this there. Let's just get the data that we need in front of our clients, stakeholders, executives, and the tool provide, you know, I think over time, the tool that provides the most value in this arena is going to win. It's going to be easier and easier to justify. Let's just consolidate on one platform. The old way of doing things no longer works. I need my data in context across all facets of of our industry, right?
[00:18:57] Roy Douber: Like, whatever it is we're doing, we need we need security data alongside networking data alongside advanced observability alongside our live debugging efforts to really improve our user experience. And that's what, that's what it's all about.
[00:19:14] Barry Hen: Agreed. And I guess my next question, Roy, is what do you think is next for Dynatrace?
[00:19:20] Barry Hen: Like looking at the logical progression of all the amazing announcements, where do they go from here? Where do you think the rest of the 2025 and beyond is looking like for a platform is prolific?
[00:19:33] Roy Douber: So, I think they've taken kind of the Facebook meta model to the way they're running their business, which I think is interesting.
[00:19:41] Roy Douber: so what we're seeing is that these competitors are coming in the market. They have creative ideas and Dynatrace is coming in and implementing some of those ideas. So like. Or, or acquiring some of those ideas in the case of RoCal. So, live debugging, acquired feature, security, direct competitor to some of the CSPM tools out there, particularly the ones that are, you know, directly provided by the hypervisors.
[00:20:12] Roy Douber: So, like your Google, your Google SCCE, your, you know, your Azure offering for security, your AWS offer for security. they're coming after, they're coming after those, you know, subsets of the subsections of observability and tying them all together. And I think it's powerful because you don't have to have that second tool.
[00:20:39] Roy Douber: You could just say now, I'm just going to add it. And it's a marginal increase in cost. I've already owned the platform. I already have the engineers that know how to execute on the platform. Now I have my data where I need it to be. and there's a lot of value that's going to continue. [00:21:00]
[00:21:00] Barry Hen: Yes.
[00:21:01] Barry Hen: Yeah, there is to that point a lot of value and not having to switch from one tool to another to see your observability data layered in with your security posture, layered in with your logging, all in one easy to find place where you can overlay them together and gain insights that you otherwise couldn't in multiple disparate.
[00:21:21] Roy Douber: And, and I think there, there's a, you know, a growing extension ecosystem. So there's hundreds and hundreds of integrations and that continues to grow. And even the developer experience there means that smaller companies like us can come in and develop an extension for one of our clients. for a feature that may not exist today and that know how the knowledge can be, you know, distributed via marketplace within Dynatrace.
[00:21:50] Roy Douber: And that's also powerful. So, there is, there is value. There's value everywhere. And I think, you know, I think this is the year or these next two years. with the speed of software engineering increasing tremendously, right? Like if you, if you go out there and develop a piece of software now with cursor, with windsurf, with some of these new ideas that exist, they make the development process a lot faster.
[00:22:22] Roy Douber: you know, from my experience when I'm writing software now, it's oftentimes some of the boiler plate, and some of the things that would have taken me, you know, a few days to develop, I can now get mostly done within hours. and, it's just, it makes it very, very simple to, take on projects that you wouldn't even have taken in the first place.
[00:22:53] Roy Douber: So there are a lot of really good ideas out there that you may have said before, Hey, you know, the time to value on this idea is not, it's not good enough. Well, when now with AI, maybe, maybe the time to value and the engineering time saved is good enough and you might have extra time to spend on that project.
[00:23:16] Roy Douber: I think, I think there is increasing value in observability, as, as they've shown us. I think Barclays, did a small presentation there. They, they increased the addressable market, from 18 billion to 50 billion for observability. They also mentioned that the top three observability, vendors only account for, I think it's 10 percent of the market share.
[00:23:43] Roy Douber: So I think there's room to consolidate. I think we're going to see more acquisitions. I think it's going to be really, really interesting. I think Dynatrace is going to continue to, trade blows one-for-one with Datadog, [00:24:00] from a feature perspective. And it's, I think, I think that's good.
[00:24:04] Roy Douber: I think that's good for the industry.
[00:24:07] Barry Hen: I agree. Nothing like some good healthy competition to realize what innovation can do.
[00:24:12] Roy, before we adjourn today, I did want to ask, what do you wish you saw being announced at Perform? If you could request one feature, one thing that Dynatrace could do that it doesn't do today that they didn't announce, what do you think that would be?
[00:24:27] Roy Douber: I think there that one of the weaknesses that I see with Dynatrace that often comes up is on the networking side. I would expect that some kind of acquisition or some kind of improved integration happens in a networking space with Dynatrace, and I would expect that to be announced by probably next year if I had to guess.
[00:24:54] Roy Douber: But yeah, I think, I think there's a lot of good players in the network space that are specialized that might be a really good and compelling acquisition for Dynatrace so that they can complete. Their network integration suite, not to say that it doesn't exist, but I think it's, it's one of those, weaker points, and I think they could improve there.
[00:25:16] Barry Hen: Agreed.
[00:25:18] Barry Hen: Thank you for that incredible insight. Roy definitely appreciate getting your expert vantage point on everything from perform 2025. Again, lots of great announcements from Dynatrace. everyone, we invite you to, join us and listen to future sessions of, our tech sessions podcast here at e360. And thank you for joining us.
[00:25:41] Roy Douber: Thanks everyone.