CDW Canada’s Glenn Kaul on how AI is changing the solution provider business

Robert Dutt: Hello and welcome to In the Channel from ChannelBuzz.ca, bringing news and information to the Canadian IT channel for the last 16 years. I’m Robert Dutt, editor of ChannelBuzz.ca, and your host for the show.

Over the last week, we’ve been bringing you conversations from Cisco Partner Summit 2026 in Vancouver, where Cisco laid out its vision for how artificial intelligence, security, and an increasingly connected technology platform will reshape the opportunities available to its partners.

What does that vision look like from the perspective of a major solution provider? And how are the changes being driven by AI already affecting the way partners work with their customers and build their businesses?

Those are some of the questions I explored with Glenn Kaul at CDW Canada when we sat down at Partner Summit.

Glenn has a particularly interesting perspective on how AI is changing the market, and not always in the ways you might expect.

We discuss how demand for AI infrastructure has disrupted traditional IT purchasing patterns, how increasingly sophisticated security threats are forcing solution providers to rethink the way they deliver their services, and how AI itself could change the economics of professional services.

We also get into some of the emerging opportunities around managing AI consumption costs, the growing importance of data sovereignty for Canadian customers, and what Cisco’s ongoing integration of Splunk means for partners and their customers.

It’s a conversation that goes beyond the announcements of Partner Summit to explore what these changes actually mean for the business of a solution provider.

Let’s get right into it. My chat with Glenn Kaul from CDW Canada.

Robert Dutt: Glenn, thanks for taking the time. I appreciate it.

Glenn Kaul: Glad to be here. Thank you for inviting me.

Robert Dutt: Let’s start big picture. You’ve been here at Partner Summit the last couple of days, hearing the announcements, talking with Cisco leadership.

What’s standing out to you? What are the things you’re taking away from Vancouver that will matter the most to CDW going forward?

Glenn Kaul: Well, there’s a couple of things.

One is that our customers’ need for having access to a platform and not just a number of point solutions, that’s not news. That’s been a trend that’s been going on for a few years, particularly in the security space, but in general.

I’d say that’s accelerating now.

And it’s accelerating for lots of reasons, including pressures that our clients have with budgets, managing price increases because of some of the component prices.

Their budgets have stayed the same, having some of the needs they have to have all the skills they need to go and run multiple platforms and systems, having visibility.

So I think that’s accelerating, number one.

And number two, I think the importance of networking, network refresh, and secure networks for AI defence as a general statement, but as a broader statement, just to have resilience and sovereignty in a secure network, that’s more important now than it was a year ago.

And it’s dramatic.

Robert Dutt: Is that accelerating too, especially the sovereignty side of things?

Glenn Kaul: The conversations are accelerating.

I think defining what sovereignty means — is it physical sovereignty? What does it mean in Canada? What does it mean elsewhere? Is it access? Is it controls?

People say, “Well, it’s all of the above.”

And so, well, how do you solve for that?

It’s an important conversation. The exact meaning of it and the implications for each customer are going to be, I think, segment- and use-case-specific.

Robert Dutt: And that’s where you guys get to have fun defining and scoping all that and making it stand up and real.

Glenn Kaul: Right.

And asking our customers, interviewing them and doing assessments, understanding what’s important to them.

For some customers, if it’s healthcare, it’s going to be different.

But look, we had a small regional public school system — or a midsize regional public school system — recently that we were responding to an RFP on, and they said, “This has to be sovereign.”

And we said, “Well, what does that mean?”

“The provider has to be sovereign. It has to prove to us that everything’s in Canada and it’s never going to leave Canada.”

We said, “Well, hang on a second. That’s about eight conversations in one.”

They wouldn’t have even asked that before.

And this isn’t the federal government asking. This is a small school system.

Robert Dutt: Yeah, yeah.

The three big messages that I keep taking away from this week are the push around AI, security, and the evolution of the partner ecosystem and partner program.

When you look at those priorities through CDW’s lens, where do you see the biggest opportunities to grow the business right now?

Glenn Kaul: It’s secure networking and AI defence.

And these are related items.

The new partner program has been announced for some time.

We’re very fortunate, and I think the scale partners are fortunate, in that we have the resources and more specializations and master specializations than anyone in the country.

And so we have the breadth and depth of the resources we need.

It’s a question of, within that very broad portfolio, making sure we focus on the right things that are important to our customers and that fuel our growth.

Robert Dutt: One of the things that I keep hearing is that AI is changing what customers need from you guys in the partner base.

Jeetu Patel talked about the need to rethink business processes rather than just layering AI onto existing workflows, for example.

Is that consistent with what you’re seeing and what customers are either looking for or ultimately needing, whether they’re looking for it or not?

And how does what you’ve learned and heard here change the conversations you’re having with customers around that?

Glenn Kaul: Well, first of all, I’d say every time somebody says AI, which we hear frequently, I always ask, “What’s the word before or the word after AI?”

Because what are we talking about?

Are we talking about how we use AI internally? How Cisco uses AI internally? How the tools, and particularly the network and security tools, have AI embedded in them to make our lives easier and our customers’ lives easier?

Or are we talking about defence against AI threats?

Are we talking about their automation and their productivity?

And the answer is it’s all of the above.

So I think every time we get into one of these conversations, I always want to start to decompose it a little bit to understand what about AI we’re talking about.

At CDW, we think about AI in four different ways.

We think about it internally. We think about it for our customers. We think about it for our partners. And we think about it for our coworkers in terms of their own personal productivity.

We have four executive leaders with four different streams that are well coordinated as we move that forward.

The same thing applies to our customers, and the same thing applies to the partner quadrant of that.

And I’d also say it’s different by segment.

Our small business and smaller commercial customers are just exploring very basic AI productivity tool uses.

Whether they’re using Microsoft Copilot in many cases or, in the smallest businesses, using Google Gemini, they’re starting with the very basics of productivity.

And then some of our larger clients are doing more inference-style work and modelling.

And some of our really large clients are contributing to frontier AI models, but are probably more concerned in Canada about AI defence and sovereignty.

So the conversation is different depending not just on the size of the organization, but what kind of organization it is.

If it’s critical infrastructure or if it’s a high-privacy or high-security environment, the conversation changes a little bit.

And so that means our team needs to be really well educated and have enough expertise across the board to be able to engage in the initial conversation, but then to be able to very quickly bring in the deep subject-matter experts, depending on what the need is, to go and take it five levels deeper.

Robert Dutt: And I imagine in some cases, probably not the folks who are contributing to the frontier-level models, but for some of those customers, that’s going to be a matter of education too.

Glenn Kaul: It is.

It’s funny, people talk about Mythos still.

And I say, look, stop talking about Mythos, because it just leads to the possible retort or conversation about, “Well, there hasn’t been that massive breach because of Mythos.”

And that’s not the point.

The point is two or three things.

One is the ability to find vulnerabilities quickly.

And the cycle time to find them is — look at PHP last week.

That was discovered by friendly AI tools because they were able to scan environments and code quickly to understand what the vulnerability was.

So we’ll see more of them, and they’ll be found faster.

Secondly, they can be propagated more widely and quickly because they’ll be well known.

And PHP is another good example of that.

And third, the defences aren’t so obvious in every case.

And finally, I think maybe one of the more significant pieces is that you don’t have to be a super-engineer expert anymore to be a threat actor.

My technical skills are pretty limited, but I have enough information now to be able to ask a model for instructions to do something that might be nefarious.

And that wasn’t the case even a year ago.

Robert Dutt: Right. Especially the pace part of that.

The bad news is that these exploits are coming around faster or being discovered much faster. The good news is the defences are also coming online faster.

How is that changing how you guys think about your security business and the pace at which it has to iterate, respond, and meet customer needs?

Glenn Kaul: There’s a great quote in one of the sessions today that said, “We used to do annual system health checkups.”

And it’s like, well, annual is no longer, because now it’s daily, now it’s automated, now it’s ongoing.

If you think about that, again, in the last few months or year, we’ve gone from “Let’s do an annual assessment of our environments to see what the vulnerabilities might be,” or quarterly patching updates.

This is now a daily and hourly activity.

It has to be quite agile to respond to fast-moving new threat vectors.

Think about how fast that’s changed in just a few months.

So the way we approach things is less about designing things up front and long implementations.

We have to rapidly transform to being as agile as the threat landscape is.

Robert Dutt: And resilient, I would imagine. Or the customers have to be as resilient.

Glenn Kaul: Absolutely.

We talk about secure networking as being one of the big, significant opportunities for us, but defences for our customers and where we can bring some expertise to our clients.

Data resilience is a big part of that too.

And we’re seeing a significant increase in dialogue and demand around just the data resilience portion of it.

And even microsegmentation.

Just assuming somebody gets into somebody’s environment. We used to spend so much time making sure they didn’t.

Now we have to assume they’re going to.

Robert Dutt: That’s right.

Glenn Kaul: And now we say, when they do, how can we protect against lateral movement and east-west traffic?

How do we protect the data and make sure that we can be up and running quickly, even if we’ve isolated the environment?

Again, these are all important security principles, but it’s a significant departure in terms of importance and emphasis from what we had just a few months ago.

Robert Dutt: In conversations about partner economics, we’re hearing a lot about the emphasis on lifecycle services, customer experience, recurring revenue.

How do you see the balance between the traditional infrastructure sales business and ongoing services evolving at CDW, as everyone’s on this journey to find the right balance or the right business model around the combination of the two?

Glenn Kaul: Well, services have always been an important part of our business.

Six years ago, we acquired Scalar Decisions specifically because they had, in part, a very significant services business as well as a SOC and a NOC.

It’s been part of our strategy at CDW globally, and particularly in Canada, for seven or eight years.

It’s increasing in importance, although the services landscape is also changing.

Do we need to hire a solution integrator or technical experts to do firewall rule changes as much anymore? And how much of that’s going to be AI-driven?

So it will increase the reliance on companies like us for more of the design and the expert services than the basic services.

So it changes as well.

But it’s for us to provide value to our customers and help them with outcomes.

I would say we don’t sell technology at customers. We create buyers, and we create demand with our customers because they trust us.

And that doesn’t happen because we are selling products.

It happens because we’re involved in the ideation, the solution architecture, and the services implementation, and in many cases, the operations.

Robert Dutt: So back to your point about what do you mean when you say AI, this is one of those things that covers a couple of those aspects.

Do you find that AI is changing how you price and deliver services, or is it primarily creating new services to offer in different ways?

Glenn Kaul: It’s creating new services to offer.

We might retire some of our services if we decide that they are effectively commoditized.

We’re not at that point yet, but I could see that coming.

I think we need to be prepared for the fact that it’s changing.

At the moment, most of the services that we’re providing around AI are either enablement conversations for some of the smaller customers, or it’s data analysis, data reconciliation, and hygiene conversations to get data right to be able to do AI.

And increasingly, it’s about privacy, security, and sovereignty.

And I think the next one is going to be tokenomics.

It’s going to be: How do we help our clients understand and manage, as part of their data operating model, the token costs and when to use which technology environments for which purposes?

Because all of a sudden, just like a lot of consumption-based models, including cloud, you could wind up with a big bill at the end of six months or the end of three months and not realize how much it was costing versus your budget.

And for small organizations, that’s going to be hard.

We can get ahead of it with our customers by showing them, “This is how you’re going to consume these.”

I heard a great quote, and I have to credit it to one of our esteemed competitors, who said, “I’ve learned there’s a big difference between a free beer and a free puppy.”

Robert Dutt: That’s fair.

In terms of internally, I think you touched on this a little bit in terms of what services are going to be things that you offer versus what’s going to become a baked-in part.

For example, you gave firewall updates. That can be increasingly baked in.

How do you think about AI internally represented in the services that you bring to market?

Glenn Kaul: That’s something we’re exploring every day.

Our first act is to make sure we have all the right tools and training that our technical experts need to be able to explore it themselves.

And the second piece is where we are now.

We have 100 percent of our 1,100 Canadian coworkers, sales or technical or administrative, with at least some foundational training in the AI productivity tool suites.

Sometimes they’re very basic ones.

But we have 100 percent compliance in that, and that took some effort.

The next step now is how do we use it in our everyday work?

And for our service delivery professionals, who are naturally curious, and especially the security ones are particularly curious — that’s part of the hiring credentials, curiosity — they have been given great latitude to figure out, in the course of doing work as opposed to in a lab, how they can use the AI tools available to us to have either higher-quality or more efficient service delivery outcomes for our customers.

And we’re right in the middle of that.

There’s no destination to that. That’s going to be an ongoing evolution.

We have to make sure that every one of our coworkers, particularly in the service delivery part of the organization, is making that evolution and the constant modernization of workflows part of everyday life.

It’s not like, “Here’s a new way of doing things.”

It’s, “Here’s a new way of thinking about things.”

Robert Dutt: You touched on tokenomics a little earlier.

I’m curious where the customer conversation is at with that, particularly how it’s coming to you.

How often is it coming to you in the form of a customer saying, “We have a problem here,” versus you being in front of it and being able to guide expectations around token costs and managing that?

Glenn Kaul: It comes two or three different ways.

There is no question that there’s the, “I just started this, I got these bills, I need to figure out how to rationalize this.”

Robert Dutt: Suddenly Claude owns my house.

Glenn Kaul: Exactly.

And the CFO has a little chat with me about the budget depletion.

That’s one avenue.

A second one is, “I’m about to deploy these things. I’ve now got my licences, or I’ve now got the sandbox environment we’ve set up to do some of this experimentation. What do I need to think about?”

And so that’s part of the collaborative, proactive approach.

And the third is just general education as we’re educating them about the art of the possible with AI-enabled tool suites.

Robert Dutt: Let’s talk about what customers are actually spending money on right now.

There’s obviously enormous interest in AI, whatever that may mean to them, depending on where they’re at and what they’re doing.

But there’s a lot of pressure around budgets, security, and demonstrating ROI.

Where are you seeing customers move most from experimentation, dipping their toe in the water, into meaningful investments right now?

Glenn Kaul: I’d say what’s really interesting is that the year was not what we expected in terms of our calendar 2026 year.

A year ago, we knew that AI was an incoming trend.

What we didn’t know, even 10 months ago, was the impact it was going to have.

Looking back now, we could have predicted it, but the impact it was going to have on customers’ budgets.

It actually has nothing to do with AI directly, necessarily.

It was the unintended consequence of the huge demand from the large AI factories that were being built by the hyperscalers and neoclouds and others that drove DRAM prices up.

So what happened, starting in late December last year, but really early to mid-January, was that we started seeing the price increases coming in.

Eight percent here, 10 percent there, on every laptop, on every desktop, on every server.

And of course, anything that has a chip or memory in it was impacted, just not as much or as fast.

People always ask, “Well, did customers just go and pre-buy and buy in advance?”

Some of them did.

More importantly, I think customers changed their buying priorities in terms of the critical things they’re going to need over the next year or two, in part because of price increases and in part because of the subsequent concern about availability.

“If I need this capacity in storage, I need this capacity for compute, AI or not, AI-related or not, I had better get my orders in.”

And so some of the longer-range planning and outcome-based investments that our clients wanted to make were really put on hold a little bit because they had to repurpose not only the dollars, but also their time and the things they needed to make sure they could operate later in the year.

We saw six or seven months of rapidly shifting customer priorities.

Sometimes big projects that were for 2028 or 2029 needs got put on hold a little bit or deprioritized.

In some cases, some of their 2027 needs became a higher priority, in significant part because of prices, but also because of concerns about availability.

“If I just can’t operate unless I have this much storage available to me in December of the year, and we think there’s going to be supply chain constraints…”

Which haven’t materialized as much as people may have feared, but they’re still there.

The ETAs, the elapsed time from ordering even a custom-configured laptop, used to be two weeks. Now it’s 16 in some cases.

They had to be prioritized because of concerns about operational demands first.

And now we’re back into the conversation around what do you need to secure and harden, or to enable your business for future growth.

It was interesting how fast that changed over the last few months.

Robert Dutt: Has that pressure moderated in a way to a point where customers are taking stock of where they’re at and reprioritizing some of those things that got backburnered back when RAMageddon was really running wild?

Glenn Kaul: Well, RAMageddon, yes.

I think that’s true.

I think we’re in a place now where there are more thoughtful conversations and planned conversations.

We do a lot of executive briefings with our customers, where we bring in our customers from all across the country and sometimes internationally.

We’ll spend a day or two with them on whatever the agenda item is for them and bring in all of our experts.

Those are back on and have been rescheduled, and we’re doing three times as many now as we were doing a year ago.

So you can see us being back in the planning cycles with customers, whereas it was reaction and response in that February to probably May timeframe.

Robert Dutt: Kind of like COVID, but in a different way.

Glenn Kaul: Exactly, right?

And COVID was different. Everybody had to figure out how to work from home.

And then once we figured out how to get everybody working from home, we took the opportunity — our clients took the opportunity — to refresh the networks in their offices because nobody was there.

Those things happened within weeks of each other.

So we’re in the same kind of rapid-response and agility place now, but just for different reasons.

Robert Dutt: You’re in an interesting place in that you’re part of an international organization.

Obviously, CDW is active around the world in many different markets.

I’m curious if there are any specific-to-Canada peculiarities, opportunities, or challenges that you’re seeing in the Canadian market that are different from what you’re hearing from your peers around the world.

Glenn Kaul: I meet with my counterparts in both the U.S. and in the U.K., where our international operations outside of North America are managed.

We were all in Chicago last week doing 2027 planning, and we had a lot of conversations around this.

It was interesting.

As every segment of the company around the world started to present their 2027 plans, we saw the similarities all across the world.

Now, there are nuances in federal government in the U.S. versus federal government in Canada. There’s no question there’s nuance.

But there’s much more consistency.

I would say that security is important globally.

The sovereignty conversation, even though sovereignty is just as important in the U.S., is not as much part of the first conversation as it is in Europe and in Canada.

Robert Dutt: With Cisco bringing Splunk more deeply into its security and observability strategy, how do you guys think about Splunk and that business?

Are you very active there?

And do you see that creating new opportunities for CDW to bring together capabilities that customers have traditionally bought and managed separately?

Glenn Kaul: Absolutely.

Splunk is a fantastic technology.

It has expanded rapidly over the years through both development and acquisition and tucking in new capabilities.

And it was a material acquisition, both monetarily and strategically, for Cisco.

It’s complicated technology. It’s complicated licensing models.

And it doesn’t just tuck right into Cisco.

It’s taken them some time, but it’s been time well spent, and they’re being thoughtful about how they integrate it.

How do we get the customer-facing experts for both Splunk and Cisco to connect better so we’re not having, to your point, two different conversations? You’re having one related conversation.

And we’re at that juncture point now.

They still have some work to do, but you can see even in the product development, which is probably a little bit ahead of the field integration, there are more Splunk conversations happening as part of a Cisco conversation.

Whereas before, it was either a Cisco or Splunk, or a Cisco and Splunk conversation.

You can see it happening right now.

It’ll still take some time, but the product development is probably ahead of the field implementation on that.

Robert Dutt: What’s driving the unity of that conversation right now?

Is that a function of the integration being more mature? Is that a function of you guys getting out ahead of it, Cisco messaging, customer awareness?

Glenn Kaul: I think it’s all of the above.

But I’d say from a pure financial point of view for Cisco, making a massive investment in Splunk, there’s no point in doing that and having it as a separate operating entity or software stack or anything else.

Because as Cisco’s ecosystem turns increasingly into software and centrally managed and controlled and visible capabilities like Cloud Control and IQ and other things, data is a huge part of that, and Splunk is a huge part of data.

So commercially, it makes sense for them.

Logically, it makes sense for our customers, and therefore it makes sense to us.

Robert Dutt: From CDW’s point of view, what would be the next step that would unlock the most value for both Cisco and yourselves in bringing Splunk to market and making it a bigger part of the overall business?

Glenn Kaul: I’d say for our largest and medium-sized customers, including a lot of public sector customers, the enterprise agreements.

It’s been possible for the last year or so to include Splunk in an enterprise agreement, but it’s been manual work and not easy, understandably, because they’re very different licensing models.

But that’s now a matter of course, and our largest client is about to include Splunk in their enterprise agreement in the next few months.

I’d say that’s going to be the number-one piece of it: making sure that the enterprise agreement really means enterprise agreement, and it is the entire portfolio, including Splunk.

Robert Dutt: Yeah, that would make sense. That’s a big unlock.

Last one for me.

If we’re having this conversation again next year in Las Vegas at Partner Summit 2027, what do you hope to have accomplished over the next year?

What would tell you that CDW’s Cisco business has successfully capitalized on the opportunities Cisco’s talking about this week, and that Cisco has made them real for you guys as well, looking at both lenses?

Glenn Kaul: I think in the broadest sense, it’s the increased understanding by all of our professionals about the platform, and the truly connected platform.

Not just the lowercase-p platform, where it says we have, under the same brand, a number of different integratable capabilities.

To show that it really is a single pane of glass, and the visibility, and how all these technologies interoperate.

I think in general, we’ll see massive progress on that.

But I’ll be most pleased to see that we’ve helped our clients with Cisco harden their environments, have better visibility, and really have realized the first important next step in secure networking and all their related capabilities.

Robert Dutt: All right. Once again, thanks for taking the time, and I hope the rest of the stay in Vancouver is a good one for you.

Glenn Kaul: My pleasure. Thank you.

Robert Dutt: There you have it, Glenn Kaul from CDW Canada, speaking with me at Cisco Partner Summit 2026 in Vancouver.

I’d like to thank Glenn for his time, and thank you for listening today.

One of the things that stood out for me from this conversation is just how many different ways AI is already changing the solution provider business.

It’s creating new opportunities, certainly, but it’s also changing customer priorities, putting pressure on existing service models, and introducing new challenges around everything from security to consumption costs.

And as Glenn points out, those changes aren’t happening in isolation.

They’re playing out against a backdrop of evolving customer expectations around resilience, sovereignty, and the need to get more value from increasingly complex technology environments.

If you’re finding value in the podcast, please subscribe to or follow it wherever you get your shows.

You can find us on Apple Podcasts, Spotify, YouTube, all the usual players.

And of course, if you want to, leave a rating or review. We’d love that.

I’m Robert Dutt for ChannelBuzz.ca, and I’ll see you in the channel.