Skip to content

By playing, you agree to uListen's Terms and Privacy Policy and YouTube's Terms of Service.

Product SchoolProduct School

HP President on Replacing Middle Management with AI Agents | Faisal Masud | E314

HP is best known for hardware and printers. Faisal Masud is turning it into a software and AI company, and he thinks the era of stacked management layers is ending.

Carlos González de VillaumbrosiahostFaisal Masudguest
Sep 30, 202630mWatch on YouTube ↗

Summaries, chapters, transcripts, and FAQs on this page are made by uListen, not YouTube.

EVERY SPOKEN WORD

  1. 0:00 – 0:50

    Trailer

    1. CV

      Now people are using AI in ways that it's not very great. So what is your take on that?

    2. FM

      If you lack ideas and you're relying heavily on AI for everything, you're not really gonna be that important.

    3. CV

      Faisal Masud, SVP and President of HP. I wonder if you found ways to achieve quality without getting through the nitty-gritty of getting some slop along the way.

    4. FM

      It's literally becoming hard to understand what's actually written by a human versus just being pure slop. In some cases, they're adding more work, not actually deducting.

    5. CV

      The million-dollar question for a lot of the CIOs that are being pitched software, hardware, hybrid solution, how close are you to the outcome?

    6. FM

      If we can't provide that return on their investment, we shouldn't exist. It's as simple as that.

    7. CV

      The problem that we are creating in the back end of all of this is who is going to manage this fleet of agents?

    8. FM

      So I think the, the flip is-

  2. 0:50 – 1:48

    Intro: Carlos welcomes Faisal Masud, President of HP Digital Services

    1. CV

      Hey, this is Carlos, CEO at Product School and your host on The Product Podcast. My guest today is Faisal Masud, SVP and President of HP Digital Services, which is HP's software business. His path there is odd in the best way. Six years at Amazon, then eBay, Groupon, CTO at Staples, then COO at Wing, Alphabet's drone delivery moonshot, then CEO of Fabric. Now he runs software inside a company most people still think of as printers, and his platform tells customers to buy less hardware. Things we'll cover in this episode: The irony of the AI manifesto and the slop it produced. AI is part of your strategy. It cannot be the strategy. Selling a product that tells HP's customers to buy fewer HP machines. Outcome-based pricing when you have fewer human seats to charge for. Why the middle management layer is the one AI actually removes. Let's get into it.

  3. 1:48 – 2:15

    Welcome, and Faisal's "very complicated" title explained

    1. CV

      Welcome to The Product Podcast, Faisal.

    2. FM

      Nice to meet you.

    3. CV

      So your title is very complicated. You have to help me here. It says President and SVP of HP Digital Services.

    4. FM

      That's right. Uh, I didn't, I didn't create the title, but it's the internal title for, I basically run, um, all of HP's commercial software business. So the digital part is essentially anything that's to do with software that we sell into our enterprises and channel partners.

  4. 2:15 – 3:05

    Why HP (hardware and printers) expanded into software

    1. CV

      That is kind of unique to HP. I mean, the first thing that someone could think of HP is hardware, printers, right? So what was the rationale behind e-expanding into software as a service?

    2. FM

      Yeah, it was, um, back when I joined, around that time, they were forming an organization, which later became Digital Services, that was primarily set up to be exactly that. A, a bit of a deviation away from the hardware, the separation on purpose so that it could be a standalone, uh, operating unit that builds software to support the hardware business. And the reason we did it is because, you know, it's very hard to just run a purely analog hard-hardware business. You have to have the data around the hardware. And the thesis around what I've been building with the team has been exactly that, to collect telemetry from what we serve our customers with, and then those insights serve

  5. 3:05 – 4:18

    From Amazon, eBay, and Alphabet's Wing to SaaS "by accident"

    1. FM

      our customers.

    2. CV

      So I was looking at your background, and I've seen you, you worked at Amazon, eBay, Google, but the drone delivery business, Wing, that was also really, really cool.

    3. FM

      Yeah.

    4. CV

      So for someone who comes mostly from commerce, e-commerce, agent e-commerce with a stint in drone delivery, how does some-- How did you end up in SaaS?

    5. FM

      Yeah. How I ended up in SaaS is, uh, completely by accident because the early, early days of this was started around my days at Fabric. After I left Alphabet and Wing, I joined some ex-colleagues from Staples and eBay who had started a company called Fabric, and I joined as, uh, CEO. And at that time, we were very, very early. We were still doing our seed round. So I got involved pretty early in the product. The interesting part about it was we had built something similar at Staples for our commerce platform, and we were trying to transition away from the legacy IBM WebSphere. And so this became a platform that we served everybody that wasn't Amazon, that wasn't, you know, does, does not have all the capabilities that Amazon had. And that was it. That's where my journey for SaaS started. I, I was sort of, most of our selling was founder-led selling for SaaS, and, uh, so took that company from series C to series E, and soon thereafter, stayed on the board for a bit and then, and then

  6. 4:18 – 6:03

    His tweets, and why AI FOMO is off the charts

    1. FM

      joined here, so.

    2. CV

      I was going down the rabbit hole on your X profile, and I found a, a couple of, uh, interesting tweets. I just want to bring them up and, and get your take here.

    3. FM

      Yeah.

    4. CV

      Uh, one was a response to, to Tobi, co-founder of Shopify. Uh, now he's all over the place again because he's talking about the AI grenade, right? But he was also the originator of this CEO AI manifesto to encourage people to use AI, and it seems like now people are using AI in ways that it's not very, uh, great. So what is your take on that?

    5. FM

      You know, my take on it is the, I wrote about this, the FOMO is sort of off the charts. I mean, two years ago it was completely off the charts, but now it's sort of settling down with all the token spend. But I always find that it's good to adopt, be an early adopter, but it's also good to have some data around the adoption versus just going all in. And I think the response you're, you're referring to might have been one around the slop grenades that he's talking about. And the irony of that, that I found was, you know, on the one hand, we're encouraging everybody that will replace every job with AI and don't come, you know, to a meeting if you haven't figured it out with AI, and encouraging that as much as possible to then, on the other hand, talking about how it's just slop central. And that's what's happening. It's, it's literally with the introduction of AI at work for those who are adopting, it's becoming hard to understand what's actually written by a human versus, you know, just being pure slop. And I find it ironic that, you know, you can't encourage an organization to go all in, and that's what they did with no guardrails around it, and then on the other hand, you have a problem with the massive amounts of slop that's, uh, you know, uh, in your organization. So you have to pick your, pick

  7. 6:03 – 6:31

    Pushing a team all-in on AI, and getting through the slop

    1. FM

      your sort of lane.

    2. CV

      I think maybe coming from a startup myself, I- Also encourage my team to go all in with AI, and of course, we had to get through slop. I actually had to get through slop in many ways, right? But I f- I, I sense, I, I thought of that as a necessary step in order to achieve quality. I wonder if you found ways to achieve quality without getting through the nitty-gritty of like, you know, getting some slop

  8. 6:31 – 8:01

    AI's dominant role in the enterprise: inform and consult

    1. CV

      along the way.

    2. FM

      Yeah. I think AI has a very dominant position in what we do at our enterprise, right? Outside of a startup environment, just even in an enterprise, which is AI is really good at the inform and consult part of the equation on whatever you're doing, to get information, to be able to consult that, with that information. But what it can't be is completely, you know, copy-paste, and, and I think that's where things get difficult is that human judgment's going to become really important here in this, where essentially i- if you lack ideas and you're relying heavily on AI for everything, you're not really gonna be that important. I think Nassim Taleb recently tweeted this, I don't know if you follow him, but I think this is something what he said, uh, those with high IQ and, uh, no, no original ideas, something like that, in the world of AI, are gonna find it very hard. And, and I, I, I tend to agree. You have to be, um, creative at the same time as utilizing AI along the way. So I think you did the right thing, and you probably expected the slop to sort of go up and at some point come back down.

    3. CV

      Exactly.

    4. FM

      I, I'm still waiting for the come back down part.

    5. CV

      Yeah. That's true. It's also easy to give up, you know? Or just stay in that slop stage and not push through enough, and I think that's where I found a lot of, um, lack of gains for, for most of the clients that we, we work with, that they're still doing all the right things on paper, right? They, they all sign the AI manifesto. They all bought the tools. They all, uh, somehow create some metrics around adoption, but clearly that's not enough.

  9. 8:01 – 8:51

    The flood of AI tools, and cutting through the noise

    1. FM

      Yeah. And, you know, we... I've seen enough tools now that are flooding the market of, you know, uh, knowledge graph this and being able to be integrated into all of your collaboration tools, getting alerts and alarms through these bots that are powered through, you know, the litany of providers. In some cases, they're adding more work. They're actually introducing more work into your daily workflow, not actually deducting. So where, where I have a challenge here is that I think AI is not your strategy. AI is part of your strategy. It's, it's going to help with your strategy. It cannot become the strategy. Then you've got a problem because it's a, it's a tool in the toolkit, right? It's not going to be the end-all. And I think leveraging it, uh, in the best way possible while utilizing human judgment, I think that's where the real impact

  10. 8:51 – 9:19

    The AI layer being built on top of SaaS

    1. FM

      is.

    2. CV

      I want to get your take also on this AI layer that we are seeing be built on top of a lot of SaaS. I mean, you come from e-commerce, so you've seen Muse or Instinct or Grok Bot, and how now anyone... I'm obsessed with Instinct, for example, right? I can throw my agents to go and do things on top of SaaS for personal work, uh, purposes. So how do you think of that as an actual tool for employee

  11. 9:19 – 10:52

    HP on the OpenAI frontier, and reading the model landscape

    1. CV

      productivity?

    2. FM

      You know, we're on Frontier with OpenAI, and it's fairly integrated here at HP with the latest models, et cetera. And just to answer your question on Instinct, like that's a product that you're very courageous. You can, you can, uh, you know, utilize it in the sense that give it all your logins, passwords, et cetera, and I, and I feel like privacy and security is gonna become a big obstacle along the way in that. It's not going to be trivial to just hand over all your credentials to everything to one particular bot that's gonna go do what it needs to do. People might be very selective in what they use it for. Muse being very similar, and you saw Amazon decline to work with them, just like they declined with Perplexity, and I, I, I, I don't know where I stand on that because I feel like Amazon's moat is Prime, and customers will continue to go to Amazon because of Prime. They don't need Muse. Not today, at least. When it comes to the enterprise, I think Microsoft has done a pretty decent job at this point. Early days were a bit more rocky, but with Copilot and the integration of Copilot and their soon-to-be coming CoWork, their claw for enterprise. I think it's a good product for those, um, that want to have the capabilities that AI can provide along with your work. I think the most integrated product right now is probably Microsoft. I don't find that, uh, same level of native capabilities out of Google Workspace. At least I haven't seen that. It's there, but it's not as integrated, whereas if you look at Microsoft, Copilot's actually made a lot

  12. 10:52 – 11:07

    HP's own role: AI for internal workflows

    1. FM

      of gains.

    2. CV

      So where do you play in this arena, right? Like you just... You're, you're running SaaS portfolio for HP, and now you have all of this agentic opportunity along the way. So what is your strategy?

    3. FM

      So our strategy is for internal

  13. 11:07 – 12:19

    How PMs are pushed on AI vs. human in the loop

    1. FM

      workflows, you know, I'm not running the IT for the organization. I sell into IT leaders. But for our internal organization, the product managers are, are pushed very hard on what can be done through AI should be done through AI, and what needs sort of the human in the loop, then you do that. So the, the PMs use all the tools available, as you can imagine, whether it's the early days of Chat PRD down to Claude now, and wherever they can use tools, they do. All of that work that starts from ideation to PRD and beyond, I think a lot of that is getting compressed, heavily compressed, down to even early demo videos, like, are so easy now. Everything is just so much more easier now, until you need to go to production. That's a whole different ballgame. What do we do internally is we're trying to make sure we can leverage as much of the capabilities that, A, out of the box Microsoft provides in every facet possible, and then, B, internally accelerate our overall cycle time for getting from You know, ideation all the way to deploying, uh, our features to the market. So our developers use Codex, Cloud Code, and our PMs all are pretty heavy users

  14. 12:19 – 14:39

    Moving fast on AI inside a large enterprise

    1. FM

      of, of AI as well.

    2. CV

      I'm curious to know to what extent, though, because as you know, in startup world, like, there is no backup plan, so everybody's forced to moving as fast as possible. While in a larger enterprise, you can still find ways to slow down, and you come from a massive company, obviously, uh, your company's more of a spinoff here, but still, it's not small, right? So how do you really push through bureaucracy and other barriers to, to, to make sure that, that your PMs are able to move as fast as possible?

    3. FM

      You know, large organizations are about relationships, and as much as it is friction along the way to get approvals for certain tools, for certain capabilities, which, which could be exposure for the organization, you have to build the internal relationships to make sure that you can bring people with you. We have... You know, early days, back when I joined, Figma was pretty popular. It wasn't really being utilized as much internally. Pushed very hard to make sure that whatever capabilities out of the box Figma, uh, provided, we were using for all of our design. That's now obviously moved forward to pretty much you can do everything, uh, with the tool set that you have. Internally, I've not seen a ton of challenges outside of when you're using a lot of these products, the tokens run out fairly quickly, and the budgets are challenging. So I think that's where it kind of comes down to where's the ROI for all of this, right? Because if you can't see the ROI for the amount of money you're spending on those tokens and, and utilizing AI wherever possible, you have to be able to do the math and be okay with it. When it comes to friction internally, I would say that we haven't faced the level of friction that I would assume from a large enterprise like ourselves in, in utilizing the tools available.

    4. CV

      I mean, one of the, the things I can imagine from the outside is because of you. Like, ultimately, a lot of these initiatives at this level have to happen from the very top, and having a leader who's not just talking the talk but walking the walk is necessary. Like, I've seen this so many times in, in some of the, our clients that they want to be more AI native, but they're not really committing to the, the pain that it takes to make it happen. And so you need to have someone who's really fighting from the very top, otherwise it just becomes a cute initiative for some people to kind of raise a, a trophy and say, "I, I checked my box."

  15. 14:39 – 16:00

    The blood, sweat, and tears behind Workforce Experience

    1. FM

      Yeah, I mean, I can tell you there was a lot of blood, sweat, and tears going into building the original thesis for Workforce Experience. You can imagine we're in a business where our platform provides insights on our customer's total surface area for their hardware and software and their OS and their network, et cetera. But at the same time, we also encourage, you know, buying less, which is counterproductive to a hardware business, right? So, um, why do we do that? Because ultimately we exist because we want to make that employee experience absolutely world-class, and you can't make a world-class employee experience if you've got a bias in the process. And, you know, from the top, you have to make sure that the bias is removed and your, your central sort of focus, your grounding principles are, "How do I improve the experience for those people that I'm selling this into?" And for that, you have to make some trade-offs. And to your point, it wasn't easy to convince everybody, but I think it's, it's less about convincing everyone. It's really important to convince yourself first, like, "Will this work? How, how relevant is it? Will it have the ROI for the people you're selling it into? Is there product market fit? Is this a PLG product, or are you selling it into these people?" Like, there's all these questions you've got to go through. And I think once we found that we had PMF, like, it was all roads leading to, "Let's go build this

  16. 16:00 – 16:14

    Go-to-market: leveraging HP's existing relationships

    1. FM

      faster."

    2. CV

      So as you talk about that product market fit, I'm, I'm curious to know what's your go-to-market motion? Like, how much are you leveraging your existing HP relationships to then introduce your product versus just going to market directly?

    3. FM

      We do both.

  17. 16:14 – 17:29

    From a point solution to a device- and OS-agnostic platform

    1. FM

      Um, and one, one big sort of friction point you can imagine early on, although it didn't really become an issue, was when I started, uh, pitching Workforce Experience as not a point solution, which we used to have back in the day, 'cause point solutions, like, are interesting. They don't really... They're not that relevant to the economic buyer long term. But when you're pitching a platform, platform has to be completely agnostic of your, your stuff that's involved in there. So we went device-agnostic and OS-agnostic. So we didn't care if the fleets we were selling into were our competitors' brands or completely different OSs. And our principled approach in this was, "We don't care what hardware you're running. We want to make sure we can help you run that environment better." And once you have those very specific principles outlined on why you're doing that and people are on board, it becomes a lot easier to move forward. To your question on how do we do the go-to-market, we rely heavily on our existing go-to-market rails. And you can imagine as a, as a very large, um, hardware, um, seller, we, an OEM, we are constantly in conversations where our,

  18. 17:29 – 18:16

    From reselling third-party software to owned-IP products

    1. FM

      um, account reps are dealing with customers who are buying competitor product. In the past, HP used to sell third-party software. We don't do that anymore. All the IP that we have built, we are selling our software. And so the transition was basically, "Well, that we're not gonna do anymore. We're gonna go ahead and sell, you know, our own home-built, uh, owned-IP product." And so in those conversations, what's evolved now is instead of selling hardware, we're selling the solution. What problem are you trying to solve? And our go-to-market relies heavily on CIOs and those RFPs coming in with explanations around, "We're trying to solve this problem," and Workforce Experience fits into that, versus Let's sell the software with the hardware. That just

  19. 18:16 – 18:37

    The CIO's question: how close are you to the outcome?

    1. FM

      doesn't scale.

    2. CV

      The million-dollar question for a lot of the CIOs that are being pitched software, hardware, hybrid solutions is, okay, but really, like, how close are you to the outcome versus just telling me that with this new software, my employees are going to increase productivity by X percentage?

    3. FM

      Yeah, we're very clear on that because,

  20. 18:37 – 20:41

    Dogfooding across 80,000 employees to earn trust

    1. FM

      um, when we started dogfooding the product inside of HP, it wasn't a low bar, it was a high bar to convince our own IT teams to use it. I mean, we had 80,000 devices and hundreds, if not thousands, of pieces of software that, uh, that are running in these environments. And we had to convince our internal teams that this is the path forward and this will help you manage HP's IT environment better. And not until we crossed that threshold of like, okay, we, we accomplished some success there, were we even going to market. So we had a very long alpha, pretty long beta, then we had an extended release. We didn't go GA for a long time, and in every stage we had 15 to 20 to 100 partners that were alongside with us looking at the ROI of what we deliver. So an example for you is we're not walking into our customers' offices saying that, "Oh, buy this and we'll promise you the world," and nothing will happen on the other side. We're actually telling them that their ticket volume will go down by 30 to 40% because Workforce Experience will tackle all those tickets behind the scenes. Well, that's pure ROI because all those tickets are managed by desk-side support. Desk-side support, you can imagine one ticket is, you know, I don't know, six, 10, $20, whatever it takes to resolve, and it's a headache for the employee. Second, we have a pro- a feature called Smart Refresh, which essentially evaluates all of your fleet and says what you should not be buying. So in the, in the IT industry, what ends up happening is hardware's bought every three- to four-year cycles, and it's bought in a really sort of a monolithic process where, "Oh, it's done this way and that's the way we're gonna do it again. Okay, let's go buy 100,000 machines." Well, we're breaking that mold and saying, "Actually, no, you shouldn't be buying 100,000 machines. You should actually be buying 55,000 machines. Upgrade these. But all these others, Workforce Experience can actually manage this for you in a perfect configuration so you can extend the life of your machine." Now, that comes at a very low cost to them because we can manage all of that. But think about that for a second, right? Uh, we don't have other people who are doing that.

    2. CV

      And look, as an executive myself, that's what I need to hear. Like, how is this solution either going to make me more money or save me, uh, some, some cost? Everything in between, it's kind of unclear.

  21. 20:41 – 21:30

    "If we can't provide ROI, we shouldn't exist"

    1. FM

      Yeah, and, uh, you know, if we can't provide that return on their investment, we shouldn't exist. It's as simple as that. The right to be in the room is only, you know, allowed until we can demonstrate the performance. We've had partners come back to us and tell us that not only has their ticket volume gone down, their mean average time to resolve the ticket's gone down by 30%. So what used to take six, seven days is now taking two, three days. Like, those are the types of things that we love hearing. We saw that, uh, when, when folks are refreshing their fleets, they're spending 20, 30 million. They're spending 25, 30% less. Well, in the end, that's all cash. That's all cash leaving their enterprise, and if Workforce Experience can help you with that, isn't that what you want? You want to reduce cost, be more secure, and have higher velocity in your IT, uh, environment.

  22. 21:30 – 23:37

    What AI does to headcount and management layers

    1. CV

      I mean, you have a front row seat here at workforce transformation. You are seeing companies go up in headcount, go down in headcount, especially now with AI, so really curious to know, what do you think it's going to, it's going to happen in the midterm due to the AI impact in business?

    2. FM

      You know, I mentioned this in the previous podcast we were talking about, about, uh, the layers. I think that AI, what AI is demonstrably doing immediately is that those additional layers of management are just not that necessary anymore longer term because a lot of that monitoring, management, and data collection, and sort of aggregation of that information can be done by individuals. And if you have to rely on a manager who's reporting to a manager, reporting to a director, like, all of those layers, I, I'm not sure long-term scale with AI, 'cause AI can do a lot of that work. So that's one. Second is what we're seeing a lot of requests for, and we've built a new alpha in our labs, we've built a product called AI Command Center, so when, when you buy Workforce Experience in the future, there's a capability that comes with our elite package that will look at your entire landscape of employees and their AI adoption, usage, and yield. So we can actually help you predict and manage your, not just your budgeting, but also how your employees are adopting all the technology that you're releasing into the market for them. And that is something we're hearing a lot more about, which is h- how can I make sure my employees are more efficient at what they do? The challenge has been most of these CIOs haven't seen the return from all of these things, and we wanna make sure that we can provide to them a clear, visible line of sight to here's the employees by department, by persona, by usage, using these models. How can you make them more efficient? How can you make sure that there's adoption in the right places? That's what people are looking for, and ultimately what that'll return in, to your question, is you'll still have people in these organizations, but fundamentally, managers will have to know a lot more than just managing.

  23. 23:37 – 25:28

    The hard part: who manages a fleet of agents?

    1. CV

      Yeah, and I think that's a real challenge right now. We are, at least in my case, deploying a bunch of agents, obsessing over it really, like really finding ways to use AI instead of a human when we have a problem. But the problem that we are creating in the, in the back end of all of this is who is going to manage this fleet of agents? How good are they? When do we actually need a human instead of an agent or in combination of an agent? And I still haven't seen a way to easily manage that type of relationship.

    2. FM

      Yeah. It's, it's not a trivial one to manage, right? Because back to your earlier p- point, agents just talking to agents all day long, I'm not sure. Ultimately someone has to make a decision, and I think that's where things kinda come to a close. When it's a higher fidelity decision, y- there's going to be human involved. Because agent's not gonna have any responsibility on those, right? So I could envision a place where in the, in the retail arena, if you take, if I take a retail example, we would have, uh, a retail, a traditional retailer have a buyer, a planner, a merchandiser who does all the marketing activity, and they would all have assistants next to them. Like, there'll be an assistant buyer, there'll be a planner level one, and then there'll be a site merchandiser level one. With AI, I think those second levels are agents, and those first level buyers are actually now not managers. They're actually ICs managing agents. That's the format I see because what is, uh, ultimately a buyer in e-commerce doing all day? They are negotiating prices. They are submitting POs. They're selecting product. They're selecting vendors. Those tasks don't seem like, you know, in the world of Muse and Instinct, like, they don't seem like they can't be done by AI. So I think the, the flip is where you were a manager, one could foresee this, now you're managing the

  24. 25:28 – 27:15

    What agents do to the SaaS pricing model

    1. FM

      agents.

    2. CV

      And so back to your point around, like, your current business, how does that impact your pricing model, right? Like, traditional SaaS, if you're basing your pricing, pricing on seats and now you have less human seats, what does that mean for you?

    3. FM

      The less human seats is totally fine by us because ultimately we're committing to ROI around your spend and your overall ticketing volume. Like, how much are you spending on desk side support? 'Cause that's a pretty large expense for large enterprises to manage that support 24/7. We see a world where those seats could exist or not exist, but move to more of a rev share model or, like, a cost reduction model that we, in the future state, can imagine a world where you had X number of tickets a year, we commit to this much reduction overall, and based on that, we are charging you a fee that's outcome-based. That doesn't seem out of, out of scope because I think CIOs are getting smarter about not just paying for dashboards, and those dashboards are not useful. They're paying for actual outcomes. So I think I see a clear path to you can be either seat-based with commitments on outcomes, or you could be purely outcome-based, and I think the skin in the game with the outcome-based is, like, then, then the equation's very equal. It's a win for you, it's a win for us. And, uh, what ... You know, the best negotiations in the world are when both parties are winning. If one party is losing, the deal doesn't work out that well long term. It's really good short term. You know, you can high-five all day long, but a year later you, you're churning that customer. So what AI's doing is a bit of a leveling of the playing field, that do I need all these pieces of software? What are they doing for me? Are they actually generating ROI? Is there a way to compute that? And if there isn't, I don't need it. And I, I kinda like that because we believe that if it were just a software business, not supporting the hardware business inside the environments, there's so many of those, but we have a unique position in

  25. 27:15 – 27:57

    The market shift: only what's accretive survives

    1. FM

      that.

    2. CV

      We have some extreme examples already in the market, not a lot. Like, Fin, for example, is a helpdesk platform just got acquired by Salesforce for 3.6 billion, I believe. We're all about, like, outcome-based pricing. I don't know how many dollars per ticket solved. And I think that the closer we can get our pricing to the outcome of the client, the better for both. I think it puts extra pressure on the vendor to actually deliver on what they promised, but it's actually a good thing.

    3. FM

      I think so. Like, at some point, right, why do you want anything inside your IT environment if it's not accretive to the bottom line of what you do? What's the purpose? Just for pretty dashboards? Like, that's meaningless.

  26. 27:57 – 30:11

    "No more dashboards": why AI feels liberating

    1. CV

      No more dashboards, please. I'm burning dashboards. I, [laughs] I don't want more meetings for meetings.

    2. FM

      Right.

    3. CV

      Actually, that to me, it's been liberating to know that I can do a lot of the things myself again, like when we were much smaller. Eh, it's huge. So I'm also curious how that applies to you at a scale at which you operate, right? You are the president of HP Digital Services, but you're still pretty much very in tune with what's going on in the market. So what are some of those pockets where you still want to feel a little bit more like an IC?

    4. FM

      I feel fairly liber- liberated with AI because a lot of the long tail tasks are being managed through AI, so, uh, I'm able to focus more. Like, I have an agent that runs overnight that looks at, you know, what, what is really important to me. I've pre-drafts ready. Like, I'm doing all of the stuff that you're probably doing, everybody's doing that is probably i- in sort of a similar job to mine. The usage of AI for executives is very different to usage of AI for ICs. Let's just be very clear about this. Because when you're an executive, you've got a support structure around you that eliminates the need for too much AI as well, because you've got human beings around you. But where you do need AI, it's there. Where I feel like I'm more liberated is that I want to spend as much time as possible with customers, and existing customers versus new customers. That's number one. Because I believe existing customers are more important than new customers, 'cause they trusted you, they brought you in. You gotta do everything possible to build a repeatable, predictable product that has ROI, number one. Number two, I, this AI command center that I talked about, I'm personally involved in architecting what we're building because ... And I find that so amazing for myself because I'm able to be in a room with the PMs, with the engineering teams, tackling issues of, like, where's that data gonna come from? What are we gonna do about this? How are we looking at synthetic data versus real data? How are we going to surface this, and what is the action that customers are gonna take? I love that part because that's the part that gets your, you know, juices going in your brain to really think about, okay, what's the ultimate goal here? How do we get to that? And I believe longer term, for us, selling more hardware is not the answer. It's, it's selling more actual value to the customer, and that could come in a SaaS product combined with hardware, or it could come with a either/or, but it ultimately has to be measurable. And so I'm spending a lot more time with upcoming product labs

  27. 30:11 – 30:31

    Wrap-up

    1. FM

      and customers.

    2. CV

      Asad, it's been awesome to connect with you and learn from your experience at a pretty large organization but operating in a pretty AI native way. Um, thank you for your time.

    3. FM

      Thank you. Thank you, and, uh, thanks for some great questions. [outro music]

Episode duration: 30:31

In the uListen app, the transcript lights up word by word as you listen. Get uListen

Transcript of episode gMjiHs8qvT4

Get more out of YouTube videos.

Read along on your phone, search it beside the video, or ask your favorite AI app what was said.

iOSAndroidClaudeChatGPTChrome