Subscribe Stay Updated with Insight On
Subscribe to our podcast today to get automatic notifications for new episodes. You can find Insight On on Amazon Music, Apple Podcasts, Spotify and YouTube.
By  Insight Editor / 11 Feb 2026 / Topics: IT optimization , Device lifecycle , Flex for Devices , Procurement
If you’re responsible for IT budgets, device procurement, or end-user experience, you’ve seen the headlines: RAM prices are spiking, OEMs are warning about cost increases, and some analysts predict device inflation of 30% or more in 2025. It’s understandable that CFOs and IT leaders are feeling the pressure.
In this episode of Insight On, host Jillian Viner talks with Ian Murray, who advises IT leaders on device strategy every day, about what’s actually happening — and what to do about it. Ian explains why RAM manufacturers are diverting capacity toward AI data centers where margins are higher, and why that’s driving up PC component costs with no clear end in sight.
But the bigger story isn’t the shortage itself — it’s the opportunity to rethink how you manage your device fleet. Ian makes a compelling case that modern PCs, with no moving parts besides a fan, can comfortably run for four to five years. That means the old three-year calendar-based refresh cycle may be costing organizations money they don’t need to spend.
Ian introduces Insight’s Flex for Devices program, which uses device telemetry — battery health, CPU consumption, RAM usage — to shift organizations from time-based to needs-based refresh. He also walks through persona mapping: matching the right device spec to the right role so you’re not buying workstation-grade hardware for someone who primarily uses email and PowerPoint.
You’ll also hear practical strategies for extending your fleet, including re-imaging older devices for lower-demand users, cascading hardware within your organization, and locking in today’s prices on devices you know you’ll need in the next 12 months. Ian addresses the security question head-on: age alone doesn’t make a device vulnerable — running current operating systems and patches is what matters.
Finally, Ian weighs in on mixed PC and Mac environments, noting that Apple’s total cost of ownership can be competitive thanks to higher residual values — and that Apple has so far absorbed RAM cost increases into new product launches rather than issuing direct price hikes.
If you need a clear-headed strategy for navigating the next 12–24 months of device spending, this is the episode.
If you liked this episode, share it with a colleague.
Have a topic you’d like us to discuss or question you want answered? Drop us a line at jillian.viner@insight.com
— Ian Murray, Director of Services, COE, Insight
Audio transcript:
Ian Murray:
I think that the, what we used to see as a three year refresh in PCs is probably obsolete now. I think that four years, four and a half, maybe even five years as a planned refresh cycle is perfectly normal and acceptable. Some customers I talk to, they have a six and seven year planned refresh. That's too long, right? That is too long. Things are falling off it after six and seven years.
Jillian Viner:
Welcome to Insight on, I'm your host Jillian Weiner, and if you're making technology decisions that impact people, budgets or outcomes, you're in the right place. Before we dive into today's episode, just a couple of quick housekeeping notes. As you know, technology moves fast, so to help make sure you've got the timely insights that you need, we're gonna be releasing a new episode every single Wednesday. We've got some great conversations ready for you, including one with the leading thought leader in ai Ethan Molik. We're gonna revisit what it means for organizations that are still navigating the VMware conversation, and we're gonna talk to leaders who have gone through AI adoption journeys, uh, data center modernization experiences, cloud migrations, building an AI factory, lots of great insights from, uh, leaders from different businesses. Also, if you've been enjoying the show so far, please consider giving us a five star review and share us with your network.
Jillian:
It really helps us grow the conversation. Alright, let's dive into today's episode. You've seen the headlines. You've probably had it come across your social feeds, but there is a serious memory shortage that is hitting the device market. OEMs are already being very transparent and warning customers about price hikes that are coming, and some experts are predicting that we could see an inflation of 30% of devices just this year. It's causing a lot of panic buying in the consumer market, and it's certainly, of course, hitting businesses, the enterprise. So we sat down with Ian Murray. Ian is talking to IT leaders every single week to help them come up with a really smart devices strategy. He's gonna give us really practical tips on what you can do right now to make sure that you don't overspend or end up in a device's lurch. Alright, let's go.
Speaker 3:
So one of the biggest bottlenecks right now is actually a memory shortage. A new issue,
Speaker 4:
Memory training AI models is memory storage, and it's becoming a big limitation.
Speaker 5:
As you're well aware. Rams stick prices are skyrocketing because there's a shortage because every AI data center is demanding all of them through memory.
Speaker 6:
Chips is so crazy that some retailers are actually selling it at market price, like its lobster. How
Speaker 7:
We could be in a RAM shortage through Q4 of 2027.
Speaker 8:
Everyone loves generative AI so much. The AI data centers are eating up all of the memory supply.
Speaker 9:
There's nowhere for the price to go except up
Jillian:
Ian. Get us up to speed. What is going on with these headlines?
Ian:
I think that what we're really seeing is, uh, obviously there's, there's profit chasing being done by the RAM manufacturing company.
Jillian:
Oh, that's so interesting that you bring that up already. Okay.
Ian:
Well, supply and demand laptops are pretty old hat, right? There isn't anything new around, um, supplying component parts for a commodity device, and that's a pc, right? So we're in a race to the bottom for pricing. Meanwhile, our electricity bills are going up because of all of the data centers that are getting built in the area, and everybody's jumping on the bandwagon to be the infrastructure that supports the data centers running massive AI content. So there's more profit to be made in there. It's a growing booming market. Lots of demand. Consequently, out of a limited manufacturing facility, those those companies are saying, well, I may as well make the RAM for the AI and we'll just let the PC RAM just fall by the wayside. There's less margin in there, all the technology, and I think that's what we're seeing. Um, and consequently, whatever little ram there is left is just being hiked in price. Um, so I think we're gonna continue to see that while there is this constant focus on AI explosive growth.
Jillian:
So the companies are prioritizing definitely the brand for the ai. Absolutely. All right. So next natural question is how long do we imagine this is going to continue? The headlines are screaming urgency, but if you're on Reddit or other threads, you'll notice that there is an undercurrent of skepticism and you even just kind of admitted it mm-hmm . That maybe there's some overhype here. So my question to you is, is this truly ramage in, or is this RAM shortage being overhyped perhaps to the financial gain of some companies?
Ian:
I don't know that the shortage per se, is being overhyped. My thought is how long will the demand actually exist? Because there are, there are other companies that are reprioritize reprioritization of, uh, manufacturing capabilities. So there, there's huge demand. I mean, absolutely there is huge demand. When will the AI bubble burst? I think that is really the underlying question. I think there's a lot of people that are being opportunistic that they're basically, you know, making, Hey, you all the sun shines and meeting that demand. There's a lot of incestuous investment that's going on as well, right? Between meta buying Oracle stock and Oracle buying Nvidia. And it's, it's kind of circular in terms of who's, who's funding what. So I think we're seeing a lot of that. I think that's some of this is in a hype cycle around the artificial price increases, and I think everybody's being a little bit opportunistic, but the shortages are real. We are, we're expecting the analysts say that we're expecting to see increases in PC pricing at least through 2027 into 2028. Um, if the AI bubble bursts prior to that , we're gonna have all set of other issues. But , but, you know, I, I don't know when stabilization is likely to happen. I do think that the inflated prices around ram, especially around PC ram, are here to stay for the foreseeable.
Jillian:
So you think this could be the new normal
Ian:
In the same way that the manufacturers are being opportunistic chasing the ai mm-hmm . Data center mm-hmm . RAM demands. If there is money to be made in pc, then when will that priority shift? Yeah. So I don't know if it's new normal, but it's certainly here for the next 12, 18, 24 months.
Jillian:
You keep saying you, you seem like it's a win, the bubble burst, not an if, but we'll move on from that. Um,
Ian:
That's a whole other podcast.
Jillian:
That's right. That's right. I'm gonna put, let's put us in the shoes of some of these business leaders because they've been dealing with a lot of pressures for the last couple years, right? So between the pressure of having to invest in AI and now there's this, you know, refresh devices before the prices go out, right? Um, I think there's a lot of executives, probably a lot of CFOs who are exhausted and really feeling the strain When you're talking to leaders of businesses of all different sizes all over the world. What's, how do you talk them kind of down from this financial ledge, if you will?
Ian:
Well, I think that a, a number of things have happened. Okay. So with some of the, uh, recent, recent ish, uh, changes in the way that PCs are being built, there's actually no moving parts inside a modern laptop computer. There's a fan. It's the only moving part was it used to be multiple parts of the device that would move and there would be mechanical failure inside of those devices. So you would have built-in obsolescence based around the processing power, the ram, the computational power that the PC could run mm-hmm . And that demanded an accelerated refresh, right? The workloads were getting more intensive. There wasn't enough infrastructure or horsepower inside of the devices and so forth. Therefore, you had to do a PC refresh. I feel that there is a lot of surplus capacity inside a modern pc. As long as you're running an I seven, an I five, or an I seven processor, 16 gig of ram mm-hmm .
Ian:
In a laptop or an end user compute device, it's probably gonna be okay for a while. And there is very little mechanical failure that happens inside these devices now. So long story short, I think that the, what we used to see as a three year refresh in PCs is probably obsolete now. I think that four years, four and a half, maybe even five years as a planned refresh cycle is perfectly normal and acceptable. Some customers I talked to, they have a six and seven year planned refresh. That's too long. Right? That is too long. Things are falling off it after six and seven years, but four and a half years, five years is probably the new acceptable, um, for the average employee if you're working in IT development. If you're working in design, yeah. I mean that, that's too long for that kind of user, but the average person, like you and me mm-hmm . We're doing email, we're browsing, we're making PowerPoint slides, right. We're making
Jillian:
Tables. My go 200,000 miles, not your standard 100,000 miles.
Ian:
Exactly. It's, yeah, exactly. We, we'll get by with a Civic, right? or a Camry, the Corvette drivers. Yeah. You know, not so much. They're gonna need a little bit more, uh, a little bit more care and feeding. Yeah. So I, I do think that looking at the overall lifecycle, um, sweating the assets that are already there, um, it's a bit like COVID, right? I mean, if you remember the early stages of the pandemic, all of a sudden we mobilized the entire workforce to go and work from home. It was impossible to buy a computer at that time. Yep. Any electronics. And then we ran into all of the supply chain issues with the shipping and, and everything that came along with that. It feels similar. There is increased scarcity. What you can get is going to be more expensive. So part of what we do with our Flex for Devices program is to help customers rather than do a calendar-based refresh, actually do a needs-based refresh using data that we gather from the machine to help organizations really plan for what can they retain, what can they cascade elsewhere in the organization, and what truly does need to be refreshed so they can plan more and operate more smartly about their device acquisition and retention policies.
Jillian:
So you're saying you're not just looking at, oh, Jillian's got a device that's four years old time for a refresh. You can actually look at the diagnostics of it and say, actually she's a pretty light user, doesn't hit the brakes too hard. It can extend a little bit longer.
Ian:
Absolutely. Yeah. Absolutely. Well, maybe even if you've outgrown your pc mm-hmm . Maybe your old PC would be right for me as a user, you get a new one, I can maybe get another 12 months out of yours. I like that plan. So let's move that elsewhere in the organization. Okay.
Jillian:
Alright. So for those who accept that they do need to move, the timing is a big question, right? Um, every CFO is weighing whether or not to buy right now or to wait this out. It kind of sounds like you're saying there is no weighted out. This is probably gonna end endure for a long time. Right. Um, you kind of mentioned flexor devices. What is the procurement strategies that actually work in this environment? Mm-hmm . Maybe some lessons that we learned from COVID, what kind of transfers over and what's a new strategy that teams need to be looking at today?
Ian:
Yeah, it's, it's interesting. So I was talking to a, a new customer that was just going through the flexible devices motion at the end of last year, and we hadn't, uh, maybe it was in Q3, but we hadn't really heard about these price increases on devices at that point. And we were talking about the benefits of future proofing specking 32 Giga RAM as the standard in the PC build, and then that way expecting to keep the device for five and six years. And of course now a 32 gig of RAM machine is a lot more expensive, so maybe you gotta undo some of that strategy. So I think right sizing, right persona based mapping, buying the right devices based on the workloads that we anticipate the users to be using has to be part of a purchase strategy. Then look at the actual health and performance of the devices.
Ian:
Come up with that, that mix of retain versus refresh. But then to the first part of your question, all of the OEMs, they're all saying that prices are going to be get more expensive mm-hmm . Um, as this year progresses. So I think it's really about how, how clear is your crystal ball? I would personally, if it was my budget, I would plan to at least take care of buying the hardware that's available at today's prices and lock that in. Mm-hmm . Um, just for the certainty just to, to be able to have the certainty for what I know that I'm going to need to deploy in the next six to 12 months.
Jillian:
Is there, uh, I don't know, like a, a financially savvy way of doing that other than just let me go buy a bunch of devices and put them in a storage unit until we actually need them?
Ian:
Yeah, I mean the, the, the MDM if you're using Intune or you're using some kind of MDM
Jillian:
Platform, MDM meaning, uh,
Ian:
The device management platform. Okay. So Intune or Jam for what, even SCCM, you can get telemetry data about the performance of that. We've partnered with a Dex platform digital employee experience platform, which is an agent based deployed to the end point, which gives a lot deeper analytics about the health and performance of all of the Windows and MAC devices in the environment. So there we can help with persona mapping for right sizing of devices, as well as getting individual device telemetry around how is it performing, how is it, how's its health, how many battery cycles has it been through? Mm-hmm . You know, what, what's the average CPU and RAM consumption, GPU as well, um, really to, to come up with data to make that decision rather than just, well, it's been four years time to go out and buy new devices. You
Jillian:
Keep seeing persona mapping. Mm-hmm . Is this a strategy that a lot of businesses do today or is this kind of a miss?
Ian:
Um, I think it's, I think that it's a journey that many customers are trying to get onto. So I think really what we've seen is that some companies say, yeah, we've got 42 different personas in our environment. We have other companies go, yeah, we really wanna start looking at different devices for different persona types. There is a sweet spot in there. It's probably somewhere between 10 and 15 different personas based on the, the type of, uh, customer, whether it's manufacturing or, you know, more data engineering mm-hmm . Um, but I don't, I, I think lots of people are interested in doing persona mapping. I don't think many are actually doing a good job of it today.
Jillian:
Why? What's, what's the advantage in how you get it wrong? Access to data access, access to data, access to just knowing who your users are and how have
Ian:
You used there knowing who used are, knowing what devices you have. Some, some customers just have a, we only buy this type of device and everybody gets the same thing and they're sure in this specialized group Mm. Other customers are more, well, let's have a catalog, let's have some employee choice between PC and Mac. So I think that up until this point, I do think that there's been a mix of customers that have really kind of embraced that it isn't a one tool fits everybody's solution. Mm-hmm . And I think a lot more, uh, organizations are really starting to get into coming up with the right sizing of the device based on the workload and the role.
Jillian:
Okay. So let's, let's dive into that strategy a little bit. So what's the best approach to avoid panic buying right now and make sure, make sure that we're doing those smart investments?
Ian:
It, it is, it's about using data to make that informed decision. It's really about moving away from a calendar based refresh, meaning that a certain number of years have elapsed, therefore it's time to go and buy new hardware, gathering data, making sure that you are buying based on need. And, and when I, when I talk about buying and re and refreshing devices based on need, I'm not saying wait until the device fails. Mm. Right. I mean, that's a horrible user experience. You're gonna annoy a lot of employees if the only way to get a new laptop is 'cause you're all one won't turn on anymore. That's horrible. So it is about having some anticipation built in, some predictive failure, um, and, and really coming up with a strategy around without compromising too much that user experience, being smart about spending capital dollars.
Jillian:
Is there a way to extend my current fleet?
Ian:
Yes. Yeah. Absolutely. So, um, two, two different thoughts come to mind for that. One is just keep it longer, right? Okay.
Jillian:
Quite obvious. It's, you know, slowing down, maybe seeing,
Ian:
And that's the thing, right? Yeah. So, so you, again, using data mm-hmm . Determine is it slowing down? Is it having an impact on that user experience below what's acceptable? Right.
Jillian:
Okay. Meanwhile, is productivity really different
Ian:
Productivity being impacted, or is it, you know, is it, is it adequate? Is it Okay, it's about, um, retaining, without impacting employee productivity or too adversely impacting user experience. It is about taking advantage of special offers and deals that might still be available. Mm-hmm . Um, and then also looking, is there a way to cascade within the organization with devices you already have mm-hmm . So for example, if you've got devices that you would otherwise refresh, but you're planning on bringing interns in, um, then
Jillian:
Give the interns the lower qualities,
Ian:
Slower re image the device, especially Windows PCs, they get bogged down with constant patches. Mm-hmm . Right? So being able to take that device away from the, the user giving them a, a refresh mm-hmm . Completely re-imaging that device, you'll probably get another six to 12 months out of that. And I, I, I picked on interns and you, you've raised a good point, , because I did have a conversation with a customer just a week ago and I said the same thing and they had your reaction said, no, the interns get the best equipment because they're trying to encourage 'em to come back after they graduate. Ah, thought, but they actually using this, yeah, it's thoughtful, but also the interns get the better equipment than the the loan 70
Jillian:
Employees. Alright. We're not gonna get advice on strategy of who gets the lead performing machines. Uh, but there is another risk I wanna talk about besides the timing issue, which obviously if you risk waiting, you're potentially risking higher prices. Right? But the other risk of waiting and maybe holding onto those devices longer is something that keeps our CISOs up at night, which is the security risks. Yes. The older the machine, the more vulnerable it becomes. How do you address that gap?
Ian:
Well, I think that's right to a degree. So as long as those devices are capable of running current releases of operating systems with the latest software and security patches mm-hmm . Applied against them, then age isn't necessarily a factor in security. Mm-hmm . The weak spot in end user compute security is you and me. Right. Right. It's the user more than it is the device. So keeping the device secure, um, obviously make sure that users are using VPNs appropriately, passwords, multifactor, authentication, protecting as far as you can mm-hmm . On the device. And again, then, then the age is important. Like you don't wanna be running super unpatched, old, obsolete, unsupported versions of operating systems, but honestly, age by itself probably isn't a factor. All all of that considered
Jillian:
Fair. Beyond the devices, are there other impacts of this memory shortage that leaders need to keep an eye on?
Ian:
That's a good question. Um, really it's just being, being opportunistic when it does come to spotting good deals and having somewhere that if you do get to pre-buy a discount mm-hmm . Or what will be, you know, viewed as a discount at some point in the future, then partnering somebody like Insight and having access to warehousing facilities. Sure. Right. Take advantage of those smart buy opportunities. There may be opportunities as well for, um, just doing component upgrades, right? If everything else within the chassis is solid mm-hmm . Then maybe there's opportunity to do a RAM or a, or a, um, a processor upgrade. But I don't see that as being particularly common. Okay.
Jillian:
Let's say I'm a business leader, I'm walking into my next meeting. I have to make an argument for either delaying our device refresh or fighting for some budget to do a device refresh. What are, what's like the most compelling piece of advice that you can give me walking into that meeting for either argument or maybe what are some arguments that you're hearing that you're having having to help clients navigate through?
Ian:
So I, I think that going with an either or that either buy now mm-hmm . Or don't buy at all. I think that those are two extreme. I think that there is some middle ground in there and that is be smart about which devices need to be refreshed, be smart about which ones can be retained, and then buying according to need, um, that, those that are gonna get you through for the next 12 to 18 months.
Jillian:
So if you need, if you recognize that you've got a group that really gonna need a refresh within the next 18, 12 months, you should go ahead and prioritize those now. I
Ian:
Would,
Jillian:
Is there an advantage to having a mixed environment of PCs and Apple computers?
Ian:
I'm an Apple user, so I'm gonna , I'm always gonna say yes for that. So I've been a Mac user since about 2015. I've never used, uh, or barely used Windows 10. Windows 11, um, employee choice. So you've got a, from a talent competition mm-hmm . Standpoint, the younger workforce has never used a Windows pc. They grew up on iPads. Maybe they got a Mac when
Jillian:
They went to college. Do mean the iPad generation? The iPad generation.
Ian:
Exactly. Exactly. We see them, we see them everywhere. Right. Every time we go to dinner, you can see the kids glued to an iPad.
Jillian:
To be fair though, those aren't the kids coming for the internship
Ian:
Next week. They're not. But they'll be, but they'll be in the future. Yeah. And while, while the kids that are now entering the workforce may not have gone through quite that mm-hmm . It's not that dissimilar. So you've got a whole generation that may have never used a pc, they went from the iPad to a Chromebook in high school, maybe to a Mac in college, and now they're entering the workforce. So for talent acquisition, talent retention, I do think that having a mixed environment is important. Obviously, uh, creative types tend to favor more around the apple. It's the, the hardware and the, the, the systems associated Yeah. Tend to suit more of the creatives. But also when an Apple obviously does a good job of talking about this, the actual total cost of ownership, even though Apple tends to have a higher purchase price, the total cost of ownership over its full life is equivalent or may even be less because of the high residual values on Apple. It may actually be less than a pc. Typically customers that are thinking about adopting or are starting to adopt that mixed environment, some of their reluctance to do it is that they've gotta re-skill the IT team to support both the PC and the MAC environment. Yeah.
Jillian:
But in terms of Ramig Getin,
Ian:
Well, apple hasn't announced any price increases associated with the RAM shortages. Okay. So they haven't yet. They haven't, however, haven't yet. However, they do have a new model lineup coming out. So I think that what we'll see is that any price increases will be buried within the launch pricing of the new hardware, which is super smart 'cause it's allowed them to buy some time without spooking the market. Interesting. They were also the first custom company to reach a trillion dollars of as well. So they do have some pretty deep pockets, but I do think they are going to be passing on those increases in the hardware cost. It'll just be, uh, mastered a little bit with that new hardware lineup launch.
Jillian:
Ian, thank you so much for explaining all this. It sounds like rather than going off and doing panic buying, which maybe a lot of consumers are doing, businesses can really be more strategic about this using data, working with a partner like Insight to really understand what their device needs are, prioritizing the part of their fleet that needs immediate refresh and, you know, make smart moves going forward no matter how long or how big this ram again becomes. And
Ian:
Let's hope it's not too long and that we return to normalcy very soon. But yeah. Is
Jillian:
Is there a new normal, I don't know, , but at least you're making us feel good about how to navigate the next, you know, 12, 18 months. So
Ian:
Yeah, it, it, it's not quite the crisis that I think some people might fear that it is. I do think that we can be levelheaded and stable about this. Yeah. And just operate smartly.
Jillian:
I mean that comparison to COVID though, I mean, I feel like that does kind of create a very relatable, tangible panic in, in folks because that, it was six years ago, but it really wasn't that long ago. Right. So it's very fresh. But Right. Again, you've given us some great strategies to move forward, so thank you. My
Ian:
Pleasure. Thanks for having me. Thanks
Speaker 10:
For listening to this episode of Insight on. If today's conversation sparked an idea or raised a challenge, you're facing head to insight.com. You'll find the resources, case studies, and real world solutions to help you lead with clarity. If you found this episode to be helpful, be sure to follow insight on, leave a review and share it with a colleague. It's how we grow the conversation and help more leaders make better tech decisions. Discover more@insight.com. The views and opinions expressed in this podcast are of those of the host and the guests, and do not necessarily reflect on the official policy or position of insight for its affiliates. This content is for informational purposes only, should not be considered as professional or legal advice.
Subscribe to our podcast today to get automatic notifications for new episodes. You can find Insight On on Amazon Music, Apple Podcasts, Spotify and YouTube.