Device Management Without The Drama

AI is everywhere in IT operations right now, but most sysadmins don’t need more buzzwords. We need fewer tickets, cleaner patching, faster deployments, and a clear view of what’s happening across every device. That’s why we sat down with Jaren Nichols, President and COO of PDQ to talk about modern enterprise device management the way admins actually live it: too many endpoints, too many apps, constant updates, and zero tolerance for security gaps.

We get into the real pressure point behind “software sprawl” as every department adopts new tools, including AI-driven apps, and IT inherits the responsibility for uptime, support, and security. Jaron breaks down three practical places AI can help right now: faster how-to research, smarter reporting that surfaces outdated versions and risks, and higher-level support for creating policies and workflows. We also dig into the part that matters most when automation gets powerful: transparency. If you can’t see permissions, execution order, and outcomes, you’re building a black box that will fail at the worst time.

From there, we zoom out to the bigger trends shaping endpoint management and AIOps: the shift toward a single pane of glass, the consolidation of roles across Windows, Mac, networking, DevOps, and security, and why legacy tools won’t disappear as fast as people claim. We close with what the next generation of sysadmins looks like when things go right: more visibility, policy-driven objectives, and faster execution without sacrificing control.

Subscribe for more conversations like this, share the episode with the admin who owns patch Tuesday, and leave a review if it helped. What’s the biggest “this shouldn’t be that hard” moment in your IT environment right now?

Transcript

Meet PDQ And The Mission

SPEAKER_00 0:02

Hey everybody. Today we're diving into how modern enterprise IT teams are managing devices and automation and AI-driven operations and scaling for admins by admins at PDQ. Jaron, how are you? I'm doing great, Evan.

SPEAKER_01 0:16

Glad to be with you. Glad to talk about the topic as well.

SPEAKER_00 0:19

Yeah, thanks for being here. Before that, maybe introduce yourself, your journey uh to PDQ, and how do you describe yourselves these days?

SPEAKER_01 0:27

Yeah, yeah, no, great question. So I've been with PDQ for about six years now. Um, it's been a fun journey. My whole career has been spent in tech. So love to dive in, love to help IT teams whenever, however, wherever I can. You know, PDQ, we would we would describe ourselves. Um, you know, I think our our probably our mission is maybe the best way to state it. Um we make device management simple, secure, and pretty damn quick. Uh the PDQ does stand for the quick in there. You know, I think going back from our roots, you know, is built by sysadmins. We keep that as for sysadmins, we keep that in our DNA. We want to make sure that we really give every tool to IT teams to help them be successful.

Why Patching Still Hurts

SPEAKER_00 1:05

Brilliant. Well, you make it sound so simple. Of course, it's very complex, and teams are struggling when it comes to patching and software updates, deployments, asset visibility, and on and on and on. Um, what are they working through most? What are the big challenges you you see out there today?

SPEAKER_01 1:23

Yeah, it's funny. I mean, so we started in 2001, and like if you a lot of the problems you have um back then are still existing today. It's just taking a different form or maybe even just a different scale. And so if you look at a kind of our origin story was um our founders were sysadmins and they and they they used tools like Tivoli that was really nice for enterprise to be able to do uh patch management, deploy applications, to get software on computers, but also have visibility of what's going on in an environment. And they really wanted to bring that into smaller businesses. And and so what was interesting is I would say tech teams or IT teams were pretty quick to adopt a way to do this manually. They went from what was more of a maybe a sneaker net where you walk from computer to computer to really kind of put in the new the new updates or the new applications to be able to do it centralized and automated, you know, they they adopted that automation as fast as they can. Today you're seeing you're seeing that extension. You're just seeing automations because if you have new problems that are emerging, or you're having maybe more applications than you have ever had to manage. I think AI is probably giving another software sprawl there as well. And so, really, it's how can we do this in an efficient way that gives them control, understanding, visibility, and success as they want to keep all the devices secure.

Three Real AI Wins For IT

SPEAKER_00 2:40

Brilliant. And so, automation, of course, and AI get thrown around uh a lot these days, but maybe talk about where it's delivering real value from your point of view for admins today.

SPEAKER_01 2:52

Yeah, and I'd say maybe, you know, admins probably have two ways in their mind. I would say maybe the first is a tax that I can we can hit on maybe a little bit more later, too, but um, or another time. But but if if another team, so let's take your marketing or let's take your finance team and they find a tool that they think really makes them more productive. So maybe an example would be hey, we want to, we we have an FPA tool, so financial planning tool that helps build our models for us using AI. Well, they're gonna rely on the sysadmin or the IT team to make sure that that software is up, running. If there's an issue, now the IT team has to become an expert in it. So, in some ways, like not helping the IT team is really just the kind of sprawl they have of more responsibility, software that maybe they weren't really as involved in the selection or in the implementation, but now have the responsibility for. So that's that's the tax that might be beneficial to the company, but really a tax on IT teams. Now, if you look at IT teams, like where can they use AI or where have they used AI to be successful? You know, I I think we kind of have the perspective, or I have the perspective, there's maybe three different areas that you can really kind of get that uh benefit from. One is just your your research, it's the how-to. I think this is where you and I use it LLM all the time. Um, hey, I want to know how to do this on the application. I want to do this faster. It's maybe faster than contacting support, reading the help docs myself. Um, the the second is maybe the the next unlock that we're really starting to hit. And that's more of the like the maybe the analyst skill set. So I want to build this report for me. I need to see every device that has an outdated version. And I don't want to do the if-then-else statements myself, but I want this to kind of really have a visibility into my environment. And that would be maybe like think of it as an analyst or help desk to really whether it's troubleshooting, building a report, something to help me better understand what's going on. And then the third unlock for our sysadmins or for IT teams is really more that executive function. Help me build this, help me do that, help me create a policy, help me create something. So I think that's how we're thinking of it. I think we're at different levels of those three. Um, but that's kind of that's that's where we're trying to trying to make sure that we support on all three of those levels.

SPEAKER_00 5:04

Makes sense. And when you talk to customers, what's the most common? This shouldn't be that hard moment they share with you.

Security And Avoiding Black Boxes

SPEAKER_01 5:12

Um, this shouldn't be that hard moment. So I would say, you know, you know, the thing that the thing that they want to make sure they get right and and needs to be right is is the both security and transparency. Um, and so whether it's using our tool, whether it's using other tools, like they want they want to make sure who has access to what information. Does the does the software have access to the right information? Does the user have access to the right information? Am I am I trying to fail-safe it so no one shares the wrong information? And so that's like a those are table stakes. Like, like maybe that is a hard problem to solve, but it's it's really a simple ideal or a simple policy that you want to make sure is right and need to have visibility to. So I'd say a lot of conversations will go around the okay, what can this do? What permissions does it have? And how do I make this not into a black box, but really see what is happening and know kind of the chain of command or the chain of execution orders? Interesting.

The Shift To Simpler Tooling

SPEAKER_00 6:07

Um, so many kinds of new devices out there, and people are distributed working from everywhere. Um, how do you think about bringing smarter automation into device management without overcomplicating it? Uh, what are your customers kind of asking for at the moment?

SPEAKER_01 6:25

Yeah. Um, so I would say a couple things we've seen, and and maybe if I take a step back that isn't just AI-centric, but is kind of shifts that we've seen over the last four to five years. So one has been um there was a time where you asked a sysadmin if you would rather have custom-built tools to do their specific job to be done. So let's take the example of um imaging. I want to image my Windows computer with a golden image so they all have the same ones, or maybe they all have the same ones depending on what department they're in. And then a different tool, maybe to do patching, and then maybe a different tool to do remote desktop or different tool to do monitoring and alerts. And I'd say there is a mix. Maybe half the group was saying, I want a pane of glass, and half the group is saying, hey, I actually want like custom built applications for each one. And so I'd say one shift that we're seeing is more and more are now starting to really lean in towards the single pane of glass. For them, they want this to be a simpler, a simpler UI, more power, more control at their fingertips. And as you start to kind of layer on AI, is it that you need to actually be able to, through an MCP or through other means, talk to um one AI tool that our our kind of agents will do work on behalf of a request from another one? Like that could be a natural um movement. So there's one. I I think the second area that we're kind of seeing the shift is um you used to have maybe a little more specialization too within IT teams. So five, 10 years ago, you you might have you know three Windows admins, so sysadmin probably is what you'd call them. And then you'd have your Mac admin. And and they didn't always communicate, they weren't always doing the same things. And I think you're starting to see a consolidation. I think you've started to see that already, uh, very strongly, a consolidation of I need to be a little bit of an expert on every type of thing. And that's not just device types, it's not just MDM, it's not just OS, um, but it also might be I need to be an expert on networking or at least have some competency to know when to pull in network or pull in an expert, or I need to have competencies around even DevOps. You're starting to see IT teams and DevOps really come closer together. So as you think about the number of devices, you you have to think of both like we're trying to do more jobs to be done, but you also have a higher expectation for your IT teams to really understand each different, uh, each different type of device, each different OS, and each different execution of what I need to accomplish with that device. Interesting.

SPEAKER_00 8:55

So, you know, you talk a lot about truly admin-friendly tools and what's behind that in terms of UI, UX, onboarding support, yeah, and other kinds of uh uh you know pillars as well.

SPEAKER_01 9:10

Yeah. You know, great question. I'd say, I'd say this. So when I when I came to um PDQ, I kind of I paused for a long time decided, is this is this the next step I want to take in my journey? And it might not be that different than when someone chooses a software tool that you're making a big investment in. You do a little research, right? And one of the things that was interesting is um, I don't know, Evan, if you go a lot on Reddit, but I'm sure a lot of people who listen to this do uh. Yeah, and and there tends to be a little bit of a skeptical skewing of perception on Reddit. So whatever topic you're talking about, you get a healthy dose of skepticism. And and then if you go into the IT communities within Reddit, I think that amplifies a little bit too. And um, you know, PDQ was kind of had almost like a special reverence of hey, a lot of promoters for it. And I think one of the reasons and what we believe, yes, you know, I could talk a second about our UI and about how we make things simpler. Um, but really, I think the quick is really what makes it a promoter. I think that the fast feedback loop that it gets the job done and you see it completed rather than it's a set. And I have no clue when this is really getting done, is something that's going to give us um it's gonna give us two advantages. One, it gives the confidence to the user that the job is actually being done. But two, it also gives them the ability to tell us when it's not working because they saw it perform quickly. So, like we are very much committed to being as fast as we can at finding any uh outcome that they want, at showing them that outcome and really getting there. Now, how do we keep it simple? I think that's going to be something that that's going to be a lot of um staying close to our community, getting their feedback as well, similar to the job completion. Um, but we we don't we don't want someone to say they have to hire a PDQ expert in order to do it. And I think you're seeing you're seeing that general spectrum in consumer apps 10, 15 years ago with Apple. And I think you're seeing that more in in technology apps too. Someone can pick it up and run with it, and that's that's what our hope is for sure.

Where AI Is Still Hype

SPEAKER_00 11:11

Brilliant. Lovely approach. Um, the whole the whole industry has gone uh all in on AI for IT operations, no secret. A lot of it's uh hype and marketing, of course. But you know, where is AI, do you think, you know, still more hype than marketing, even if you look out this year, where we where is the industry pushing uh is an unrealistic expectation?

SPEAKER_01 11:37

Yeah, yeah. And you know, we we do uh an annual state of the sysadmin survey, and and and you know, so that's a that's an interesting data point for us. Um we see we see a lot of kind of um a lot of mixture of two feelings that you just stated. You see some optimism of like, wow, what could this be? And and if we we came back, 94% of sys admins right now see ways that AI could help them. And and they again, this is this is to get 94% of any group uh to say something like that is is pretty pretty high for sure. Um and and then if you double click on that, they say, hey, look, if I were to categorize how much of let's just take endpoint management, how much of that could be automated ideally, so in the in the long run, um or maybe in the short run, in the five years, um, is kind of how we frame that. 73% of their work, they believed, could, or not their work, 73% of endpoint management potentially could be. So you have the question of like, A, where are we at today? And B, what's left in that 25 to 30 percent that we just don't want to automate at this point or don't see a way? You know, I I think it is AI is really good at saying, here's a lot of help documentation. Can you now have a one-on-one conversation with me to see how am I doing this and why am I doing it incorrectly? Or, or it's that second category of like, let's analyze a specific instance. It's not great at analyzing holistic ones in my environment. But if I say, hey, I try to deploy this application and here are my error logs, can you use someone else's help desk or maybe someone, maybe it's PDQs, maybe it's someone else, and diagnose what is wrong and where it happened. So I don't need to spend the time going through the logs. That's where AI is going to crush it. When it says where we want to go next, though, and when I say we, I hopefully mean as an industry, um, is hey, I want to be able to set a policy and I want AI to think of all the if-then else statements to make that policy accurate. And so my policy might be I need to have you know, maybe it's a security posture, right? Like I need to make sure that my applications that are high security, high vulnerability are if it's has it needs to be up to date within 24 hours of a patch, but there's some that maybe need to be less. And and and and I don't want to articulate which software goes in which bucket. In the old old today, you would actually have to. You create literally a policy with every single software and and every iteration, say which one it is and what what it is. Hopefully in the future, it's I just want it to kind of categorize on its own and I need to review it. And I think we're not there yet. I think people believe we're there, but I think it takes a lot of of um of really you almost spend as much time arguing with the AI as you do of getting it to the outcome you want. And so I think there's there's a first opportunity. Maybe I'll touch on um a second one is um, you know, when if you wanted to build a script, if it's a simple script, you know, sometimes it's as good as my inputs, a simple script say, hey, this is what I want it to do, another outcome-based statement, it can build a script for you. If it's complex, it isn't. And sometimes you don't really know is it the AI can't do it? Am I not articulating it well? And so when you have tools like Claude that walk you through every step, you can start to diagnose the problem. But I think until we get better at really saying where are we either miscommunicating or where is its capacity or can't it do it? You know, I think building is still going to be taking ambiguous items and building and trying to execute on policies is an area that we have more hype than we actually have results at this point. Diagnosing diagnosing issues is an area that we probably are still strong.

Legacy Tools And Hybrid Reality

SPEAKER_00 14:58

Interesting. So legacy tools um have been around forever and they just seem to go the way of the fax machine. They live on and on and on. But what where do they start to break down uh given the new environments we're in? And um, do you see a time when we'll deal with some of that technical debt, as it were?

SPEAKER_01 15:18

Yeah, um it's funny, you know. I think of so so there used to be a conversation of when is everything going to move to the cloud? And everybody had their timelines. And then you get into a world of saying, like, you know what, I think hybrid is here for a lot longer than I really anticipated. And I it's going to be the same way with software. Like legacy tools are going to have a much longer life than everybody believes. At the same time, though, there's gonna be some faster, um, shorter life cycles for some software than we probably anticipate. But but what I would say is I expect a hybrid approach there. You are going to find your key software tools that you just rely on are tried and true, are low effort, are high reliability, high security, and you're gonna probably not change those out for a long time. Conversely, you're gonna find some that just have leaps and bounds, better improvements, and you're gonna be willing to make that shift to that new product and and lead your legacy apart. But but I'd say both things are true here, where some things will be shorter than you have. But I suspect hybrid, hybrid between legacy and and new software is going to be around for a very long time.

The Next-Gen Sysadmin Role

SPEAKER_00 16:24

Makes sense. So if we were to have this conversation in a year or two or three, what would the next generation of IT operations look like uh from a sysadmin standpoint if if you get it right?

SPEAKER_01 16:37

Yeah, um, I mean, if we get it right, there's a couple things you're doing. One is you're going to be able to have more visibility into things and in a way that you can be a policy-driven world or say, I want to accomplish this objective. You know, I see the sysadmin as more of like the chief of uh chief executive for devices. I want to make sure that these things are what's happening. And I have AI telling me, A, this is a time it looks like it's not going correctly, B, here are the ways that we have fixed it already. And C, this is like where we're uncertain. Can you help clarify the objective for us? And in a world that in five to 10 years, we get to that. And so they IT teams are chief executives. They're able to better uh execute fast, they're able to have visibility and be able to make decisions based on information a lot more clear. The second thing I think IT teams will do is I just think they're like, I think now is the best time to become a sysadmin or IT, because I think their sphere of influence is just growing. It's growing, you know. Again, we went from being I was a sysadmin or I was just a help desk person to I was a sysadmin. Now I'm a have to do Mac admin, now I have to do networking. I just see that continuing to sprawl. I think by nature, IT teams are looked at as most people who went into IT were known, they went to IT, they almost fell into IT because they were a curious person. And I think I think that happens to be the center of belief in most organizations, where hey, whether it's them helping with DevOps more, obviously SecOps is something we're seeing merge already as well. You know, I just see that their sphere of influence will become grander. They'll be able to say whether it's a production soft, like, you know, I'm a software company, whether they're the ones who are more looking at, hey, how about our applications? How are they doing? I see IT really focusing more on that as well. Uh, you know, I think their sphere of influence will grow as they're able to get visibility, execute better, and really kind of set the objectives for the organization. Brilliant.

SPEAKER_00 18:29

So just to wrap up, um, what are you excited about over the next weeks, months, as we head into the year? What's on your mind event-wise, uh roadmap-wise, team-wise, what's uh what's on the radar?

SPEAKER_01 18:40

Yeah, no, we're we're we're pretty excited. So we have um we we cut we have a product that is our our newer product that we're um we're we're really heavily investing in for for our AI kind of tooling. And so so one of the couple things that we named already. So we're going to be able to um start helping people build their own scripts. So right now, people do it on their own. You know, we're excited about that. It's already kind of in process. Um, we we're we're moving over to your point, Evan, on legacy products. So we actually have a product that I wouldn't call it legacy, but it is our on-prem. We're just seeing an adoption to our cloud base being so significant because of integrations, because of AI. So we're really excited about that. The the kind of the fervor in our community is probably stronger uh than it's ever been for those who are adopting our new tooling. And then for me personally, like I'm always enjoying the good weather. You know, I got out a little bit to uh to Universal Studios last weekend, so excited for for some uh good weather on that. But but PDQ proper itself, like this is a time where you're going to see inflection points in software that we look back on and say this is maybe at a race once in a lifetime, potentially, right? Or at least at least once in a generation. And so um I think the next three to five years has our team more enthusiastic about what our software capabilities could do. And I and I think we're gonna execute on that in the next six months where you're really saying, hey, this is this is how I can use help to make sure that my environment is more secure, is up to date, and make sure device management is simple, secure, and pretty damn quick.

SPEAKER_00 20:09

Wow. Well, that's a mic drop moment. So on that note, congratulations on all the success and uh onwards and upwards.

SPEAKER_01 20:16

Appreciate it. Thanks, Evan.

SPEAKER_00 20:17

Thanks. And thanks everyone for listening, watching. Also check out our TV show, techimpact.tv on Bloomberg Television and Fox Business. Thanks, everyone. Thanks, Jaron.