Why AI Value Depends on People, Strategy, and Escaping the Activity Trap
Prompt 16 - Why AI Value Depends on People, Strategy, and Escaping the Activity Trap
===
[00:00:28] Tom Adams: Hey, hey, welcome to Prompt and Circumstance. We are glad you're here. Mark, hey, how are you?
[00:00:35] Mark Redgrave: Hey, Tom. Good afternoon, good morning, wherever we are in this sunny world of ours. Um,
[00:00:39] Tom Adams: Yes.
[00:00:39] Mark Redgrave: in Southern California
[00:00:41] Tom Adams: yes. And, uh, it's just after lunch, and so, uh, it's just you and me today. The other guys bailed apparently, so it's just you and me, and I thought, in terms of a conversation today, you live in a really interesting sphere that you're playing in, and I thought it'd be kinda cool to, to, uh, talk a little bit about, um, your, your view of the world, how you're seeing it, some of the stuff that's happening for you.
Uh, but maybe before we get there, um, anything kinda grabbing your attention this week or this last week or so since we last had our, last conversation? Anything that's showing up for you right now that you're going, "Oh, that's interesting," or, "That's big," or, "That's appealing," or, "Oh, that's crazy"? '
[00:01:20] Mark Redgrave: it feels like everywhere I turn today, so what is it? It's Friday the 12th. it feels like everywhere I turn is just a SpaceX IPO thing,
[00:01:28] Tom Adams: Yeah
[00:01:28] Mark Redgrave: and welders that are about to have 5 million bucks, you know, because they welded a space rocket like, uh, in 2019. And I'm like, it is awesome, right?
Like, you gotta, you gotta love that stuff. So, um, but the Anthropic releases this week, big news. I haven't actually dug into those yet. You'll, you'll be able to, you'll be able to fill in some of the gaps for me on that, I'm sure. But, um, so seems like, but, um, the news cycle feels like IPO, like at a scale, IPOs at scale we've never seen before, incredible wealth being created for people who were early, and the San Francisco housing market is about to go on another rip.
That's what, that's, that's kind of what I, that's kind of what I'm reading.
[00:02:08] Tom Adams: Right. weirdly I'm less interested in the IPOs, and I've been playing with, you know, the new, uh, Anthropic, uh, one that came out, and I don't know. I'm not, uh, you know, I, I think, uh, I have, I have gotten in a rhythm. I'm building a piece of software right now, and, uh, I've gotten in a rhythm with Opus 4.8, which is really good, and this new one, uh, comes out, and I'm a little scared to throw it at my existing project, so I've been playing with it, uh, on other stuff, and I mean, it's quite good.
In- incredibly good. Uh, it's, it's just sometimes you go every... Like, I didn't like, uh, on Anthropic, I didn't like 4.7, but I love 4.8. Um, and so it like weirdly you get attached to certain
ones. It's
[00:02:50] Mark Redgrave: when you say you didn't, didn't like, you know, you love 4.7, not sure about 4.8, what are you seeing? Like what, how does that, how does that, yeah, how do you draw that conclusion from like what it gives you?
[00:03:00] Tom Adams: Yeah, it's, it's weirdly, uh, sometimes it's just how it responds in the moment to stuff. Like you ask it a question that you've been used to asking it from the previous, uh, version, and you move to the new version and you ask it a question that you've asked it before, and it comes back with a completely different way of...
It's kinda like you- you're, you're working with, um, a person who's, like, has a multi-personality.
[00:03:25] Mark Redgrave: Yeah, yeah, yeah. Like a bipolar
[00:03:27] Tom Adams: And so, so you have personality one, and then all of a sudden you go to personality two and, like, Anthropic has, or Claude has historically been more verbose than, say, OpenAI. OpenAI, the, the most recent Codex ones, and I do a lot of stuff in terminal, which is kind of the, not, not the app.
I do it in the, um, Claude Code or Codex in terminal. And Codex just goes, "Yes, right. We'll do it. Go ahead." Gemini is kind of weirdly in the middle, but, but Codex is, is generally quite nice to you. And then the next version comes back and it's like, where did that thing come from? All of a sudden it's not as nice and it, it responds differently.
So it feels like you've got this multi-personality disorder that you're working with, but they're also smarter every time. So it's like you're dealing with a crazy engineer, right? Who's, who's, doesn't really know human skills, but tries to pretend they know them, and they sound weird in them. It's just kind of funny
[00:04:24] Mark Redgrave: Yeah, and you, and you, you know you should be, you know you should be paying attention 'cause they're really smart, but they're not making it very easy for you. You know? Like, I, I'm not sure I wanna hang out with you 'cause you're like, you're, you're a bit weird
[00:04:34] Tom Adams: Yeah, but I have to live with you all day. Like, I work beside you, so I get used to working... You know how you, with s- with certain people, you get used to working with their style and you go, "Okay, I'm gonna get..." But then that person has a different style and you go, "Hmm, okay. All right.
We'll just learn, we'll, we'll learn the next version of you." So,
[00:04:52] Mark Redgrave: We've, we've, we've talked about this though, right? But like, like, uh, l- like one of the fundamental, like, um, building blocks of trust is consistency, and what you're experiencing is an inconsistent style, and that means, like, that, that questions trust. Which is interesting, right? Because it's like, I mean, I actually, like this last, this last kind of couple of weeks, I've had some really deep work to do around, you know, like, like, like just deep work to do with my own business and also to do with some immigration stuff that I'm having to go through, and we've got really into it with Claude, and I find it tremendously, like, supportive to sit down at my computer and go, "Hey, we're gonna launch- we're gonna dig into this."
And it goes, "Mark, hey, yeah, let's go." I'm like, I find that as someone who works primarily alone, I find it super supportive. I'm like, "We're gonna do this thing together," you know? And you kind of go, I'm like, "That's weird." But it's like I, I, I genuinely feel like I've got a collaborator, uh, I've got someone, strange
concept, who I can work with on really difficult stuff, and it's like, yeah,
[00:05:55] Tom Adams: yeah, and the other interesting, the other interesting I find is once you work with Claude, in this case enough, or ChatGPT or Gemini actually, uh, all of them tend to go, "Oh, I get to know you and I get to know kind of what you think." And I, oh, like all of a sudden it says, "But you said this, um, and does that a- a- apply here?"
And so you get this sense that this, this entity starts to know you as well. It's not just you liking the relation with this entity, it starts to get to know you, and it's weirdly human.
[00:06:29] Mark Redgrave: Yeah. Yeah, yeah. I, I, I mean, I find it, I find it really supportive. You know, it's, it's like we all need... Some of these things we do are quite hard, like intellectually, emotionally, and, and I'm like, I've got somebody to work with, and I'm a, I'm a, I'm actually a guy who I'm, I'm sort of in my previous life when I was, you know, client, like company side in, in executive teams of large, you know, like six, five, 6,000 person companies.
I loved the team bit. I loved it. Yeah. I wanna be around people. I love people. I love working with people. So it's, it's kind of, it's kind of cool, I think
[00:07:01] Tom Adams: Yeah. So, uh, this last few weeks, uh, and w- we haven't really talked about this before because a lot of times there's three of us or four of us on the call, but, I've been watching your LinkedIn, uh, posting, and y- you've really been focused on, uh, you know, AI and how you can help companies. But I wanted to talk to you less about just what you can do for companies and more about your thinking around it.
And one of the things you keep saying is AI is a people problem, not a technology problem, and I wonder what you mean by that. Because to me, AI is, is technology, but what do you mean by people problem, not technology problem?
[00:07:42] Mark Redgrave: Yeah, so and, I, this is like the building block of my complete belief system here when we, when we start to think about, you know, how do we enable AI to create business value, right? 'Cause that's what this is all about. We're not, we're not spending millions of dollars, um, to have fun here. We're investing significant amounts of capital in like i- in, in pursuit of business performance, in pursuit of whatever that may be for your company or situation, right?
But what we know from the past is whether it's, whether, whether it's like in the era of databases or CRM solutions or ERP, it's like the technology bit is the easy bit. So what I mean by that, by this is a people problem, not a technology problem, is like the tools, Tom, the tools will reach parity quickly if they haven't already.
And, and for companies, this incredibly powerful technology will be available to them in a number of different flavors and at a number of different price points. They're all gonna be amazing, okay? Now, the difference is gonna be how you leverage those tools to create value, and that is gonna require people and organizations and teams changing.
So this is what I mean when I say the phrase like, "This is a people challenge, not a technology challenge," because we've gotta get people to adopt, to adapt, to change, and this is a challenge as old as time, right? We've been doing this for 100 years as every technology wave has come through. The hardest part of all is getting the people around the technology to accept change.
[00:09:21] Tom Adams: Yeah. it's interesting 'cause, um, w- one of my clients, um, uh, he just wrote something recently and he talked about, he runs a Microsoft MSP shop, and one of the things he talks about is, is the struggle is always people love the default, right? Like, there's this, this sort of, there's this... Once you, once a lot of people figure this thing out, well, that's the default.
That's the way we do it. That's how it's done. But what's happening and what you're saying is, um, that, that the evolution of AI is now forcing us out of the defaults, whatever those were, whether they're SaaS or they're toolkit or whatever. Um, so while the technology is what's driving it, the default is what's keeping people sort of locked in to where they were.
[00:10:08] Mark Redgrave: Yeah, and like it's change, and change is hard. I mean, this is, these are not new concepts, right? It's like, you know, we've been talking about like, like how difficult change is, and for at least the last 15 years, the narrative has been swirling around, which is like, "This is the slowest pace of change we're gonna see.
It's only gonna get faster." And then we fast-forward to like 2026 and like change is ripping.
So, so you kind of go, so this is the situation we're in, and it's like, and, and companies need to understand that like to leverage these tools, to leverage the technology, and to create true durable business value, people have gotta do stuff, right?
And it, and it, and it's like, and that's where I, because of my skill set and my belief system, that's where I'm focusing on this because, um, you know, the implementation of AI tools is a skill set that is essential, but like you gotta get the people part right, okay? Before you start implementing things, the organization's gotta be ready
[00:11:09] Tom Adams: So how do you, how do you get there? Like, like how do, how do you... Like, you go into these big organizations and y- you say what you just said there. People have to get there. People have to be ready. Like, what do you, what's the, what's the process that unfolds to get that to happen?
[00:11:26] Mark Redgrave: So I, I think the, you know, like there's some foundational components that are universally true of any company in any size across any industry, right? And those things are, so for example, we talk a lot about this, but people have to understand the why if you are going to change something. Now, this is not the Simon, Simon Sinek why, that's a different why.
I'm talking about the why about like, "Hey Tom, you're about to like... feels like you're about to change everything about how I do my work and what my job might be. I need to know why we're doing it." So because if I don't see that and understand that and believe that, like I ain't gonna change nothing. So, so, so like at-- that, that's one component is like understand the why.
Now, the why does something super interesting. When people are connected to the why and they understand it and they believe it, okay, you get effort, engagement, right? You get like this concept of discretionary effort, which when you're trying to change a company becomes very valuable, right? It's like imagine you say, "Hey, we, we need to, we need to innovate across our, uh...
across the way we do this thing as a company." It's like we need people to dig in. You hear CEOs say that, right? "I need my guys to dig in and really get stuck into this." It's like, well, if they don't understand why, they ain't getting stuck into it.
So, so that's that's one example, right? Then, like it's about clarity, and this is again, this is like you'd argue, "But Mark, isn't this like leadership 101?"
Yes, yes. Leadership 101, clarity and alignment. Are my team with me? So I see this with clients. The CEO could be totally on board, um, members of his or her leadership team not so. Okay, fix that.
Because if you are trying to change your business, you need a leadership team who are aligned and ready. So a couple of examples there, right?
But when I say like this is what you've got to do because change without those things is even harder. And, and, and you know, and, and we know these are, this is the key to making this, uh, being successful in this
[00:13:42] Tom Adams: So, another side of what you're saying though, which is really interesting to me is, i- is, um, uh, we- we've gotta make it a people issue, or I mean, it's gotta be a people thing. Leaders have to get that. But then on the other side of the equation is, uh, something you call the AI activity trap. So like there's action, but what, tell me about how you see, especially in mid-market or larger, um, mid-market companies working, um, what's the activity trap?
What are you seeing? How are you counterbalancing leaders have to be engaged and do things and teach their people to do things and be on board, and the activity trap that people get into?
[00:14:23] Mark Redgrave: Yeah. And, and I think, Tom, this is super interesting paradox, right? It's like this is t- this is really interesting. And I don't claim to have the answers, but all we can do is like, is l- is continue to use our, like our skills and our insights to, to help people through this, right? So on one side, like I, I will never discourage experimentation because we've said, haven't we, on this podcast before, like people have to use these things to start to understand them.
And when they understand them, they build belief. And when you build belief systems, it's like now you're on a j- a change journey. Awesome, right? So, so experimentation is really key part of early market adoption of technology. Now, the issue is increasingly like this experimentation is not really delivering results.
And that's why we see in the media a lot of things like these, these terms of like pilot purgatory and, you know, like why, why are we doing-- Like everyone's doing so much stuff, but I don't really know why. And I think that is the era we're moving into quickly. I think H2 2026 is gonna be the time many, many mid-market companies go, "Hold, guys, hold on, hold on.
Just, just stop the train for a second. Stop the bus," right? "We have invested 100, 200, $500,000 in this experimentation, but like where are we?"
Like that's what I call the AI activity trap because that, that, um, that frenzy of activity is valuable to a point, but we now have to stop. And if you're a leader, a CEO, an owner of a business, I think you now need to be much clearer about what A- what, what role AI plays in your business, how it creates value, and how you will continue to invest in these a- in these efforts.
[00:16:19] Tom Adams: So what, like, uh, so there's activity, right? So we've talked about those last couple of years, people are going, "Oh, we gotta, we gotta deliver value. We gotta use it. It's gonna help us. It's gonna make optimized performance," whatever that is. Um, and so we've got activity, um, but then we're creeping into activity without a result, just for the sake of activity.
So, um, what's the difference between activity and maybe performance? Uh, like, what's the distinction there? Because, because there are obviously initiatives that are producing a performance, and maybe they're both happening in the same company. I don't, I don't know. But how do you see that distinction between activity that you get stuck in a trap and performance that's actually producing results?
[00:17:07] Mark Redgrave: Yeah. So, uh, that's an awesome question, and I think this is the, this is the kind of foundation of the activity trap kind of concept. So to me, AI in support of strategy, okay? That's the only way this can be. So if you are a company, like you have a strategy, you have goals, you have things you need to achieve, right?
H- The question should be: how do we use this amazing new technology to achieve those goals, to f- to meet, to meet and beat our strategic obligations? Okay? this is really challenging because for, for many companies, not a criticism, but for many companies, like they don't actually understand their strategy.
So again, it's like, oh, so that's tricky 'cause it's like, yes, it's tricky 'cause strategy, remember, right, is not like many people, when I talk to some clients, they go, you know, you say, "What's your strategy?" They go, "Oh, we've got a strategy. It's, it's a 10, 10, 10, and it's our 2027 strategy." And you go, "What does that mean?"
"Well, we, uh, t- $10 million more revenue, 10 more points of margin, and 10 more employees in, in, you know, in Belgium." I don't know. And you go like, "Okay, that's not a strategy," right? That's like a goal. And it's, and it's good. You need that, but it's not a strategy. The strategy is like, is like... A strategy is facing into to what can kill you.
That's strategy in my mind, yeah? So, so when you start having those conversations about, okay, what's our strategy, like our competitive landscape, where have we got true differentiation? Where are we, where are we at risk in terms of across our product sets, in terms... And it's like, now how do you use AI to support defense or achievement of strategy?
That's important. So, so like when you do that, right, like the AI conversation in your company changes completely because it s- stops being about like these little things we're trying, and it starts to be owned by the CEO who g- who goes, "Right, for us to achieve this strategy for our business, this is the role AI will play.
Okay, game on. Now we're, now we're on it," right?
[00:19:21] Tom Adams: Well, and it's, and, and, and it feels to me as you describe that, it's, it's the way we would hire people or make acquisitions or anything, which is we have a, a, from a, a leadership perspective, we know where we're clearly going. Uh, I like, I like the model of, um, where are we gonna play? How are we gonna win as a strategy structure, um, that came out of, what was his name?
Um, forget his name, but, uh, an author or professor or something. But I, I always like that, that model of strategy. Um, but, but the interesting thing there is we think about people, we think about acquisitions, we think about, um, um, those kind of things in the same way. But weirdly, AI, when you get stuck in the trap, is just this cool tool, and we just employ it all over the place.
Um, and we throw, throw it. So I, I get the distinction there between activity and, um, and performance then. And performance is in service of strategy. Activity is generally in, in service of we gotta do shit.
[00:20:27] Mark Redgrave: Yeah. And I, and I think, and, and as I said before, like this is challenging conversation, right? And no one's trivializing any of this. Like we talk about this, don't we, Tom? It's like, it's like when we talk to leaders, it's like no one's sat there going, "Oh, oh, you guys, ugh, you really don't know what you're doing."
It's like, no, this is challenging, right? It's like, it's like w- are we clear on our strategy? It's like, um, a similar analogy, and I know we've spoken about this before, is like you actually don't need AI, okay, to look at where value is leaking from your business. If you, if you wanna talk about efficiency and if you wanna talk about like, like how do we impro- increase the pace of our business?
How do we reduce the number of handovers? How do we get faster to market? You don't need AI to have that conversation, guys. Yeah? But it's, it's difficult work because it's like, it's like, okay, do, do-- When, when I say to CEOs, "Do you understand clearly where value is leaking, like across your organization?"
None of th- nobody can answer that question, right? Literally, like I have a 0% success rate because it's like it's complicated. And, and I think so, so, so now, now the conversation becomes super interesting 'cause you go, AI is giving us incredible tools to address some of this. But in my mind, the conversation must start with, do you have clarity on what you're trying to do?
Yeah? Do you see where the opportunity is in your business? Now let's talk about technology. So that's what I mean, right? Stop, stop frenetically doing things, and let's get back to the conversation around clarity about your business.
[00:22:03] Tom Adams: So somewhere in there, and somewhere in that leak of value, AI tends to get pushed down the org chart into VP or director levels instead of CEO priority. Why, why do you see that happening? What's, what's causing that, or why is that occurring, or what's the, what's the thing that you see under the hood that many people miss?
[00:22:28] Mark Redgrave: Yeah. So I, I, I, I actually think the question there is not why is it occurring? I think the question is, why does it matter? And let me tell you why it matters, right? so well-meaning, well-intentioned people who own budget are doing AI pilots and experimentation, which as I said before, should not be discouraged.
But here's the problem, okay? For companies, you know, every company I've ever worked in or worked for, there are a few things, very few things the CEO genuinely cares about. These are like what I call metrics that matter for the business, right? Like, that is what the CEO and the leadership team are investing in and invested in.
Now, like, this activity that's happening at director level across these companies is not on their radar, and that's a problem, okay? Because, because investment for that ultimately will, it will fizzle. It will... Because if you wanna get their attention, you have to, you have to be working on the things the CEO cares about.
Sounds obvious, right? Not obvious. Not obvious. Because I see it all the time. It's like, "Oh, we have, we have 30 AI projects in the company." Okay, great. How many of them are actually touching the stuff you, CEO, care about? "Uh, well, I don't really know what any of them are." Well, then there's your answer to your question.
Yeah? Asked and answered. So, so I, I think like there's, there's, there's this concept that I'm sort of testing with people, which is this concept of a CEO mandate, which is like, do you understand the CEO mandate for AI in your organization? Because if that's the alignment, so, so I think that's the right question, yeah?
It's l- it's, it's, it's, um, these things are happening and we don't need to, we don't need to really diagnose why. It's like, it's like, but how do we create this strategic alignment and a CEO mandate that leads to clarity around what AI is and isn't in your organization?
[00:24:32] Tom Adams: So when that, when you see that happening, I'm not I'm not sure you're at a, fully seeing it happen, but when you see that move towards that, what changes? How does it move up the, uh, up the, the org chart to actually be, uh, a CEO mandate about AI versus just being VP or director level? Because if we're not gonna ask why that's happening, but, like, then maybe how does it, how does, when a CEO mandates it, how does it move out of just being these random initiatives into something that produces
value?
Like, how do you see that unfolding?
[00:25:09] Mark Redgrave: Yeah. So I had a, I had a great-- I was at, um, I spoke at an IT, like senior IT event like a couple of weeks ago, and a question I asked was like, what, what... And I just said, "Show of hands in the room, okay, how many of you are working on like, uh, uh, like AI projects internally that you're really excited about?"
Big show of hands, yeah. And then my follow-on question was, "How many of these does the CEO know about?" Much less hands.
Yeah. And then I'm like, "And how many of you are frustrated..." Question three: "How many of you are frustrated that you don't have the support you need to go bigger, further, faster?" Every hand in the room went up.
Okay. So like this is how this is playing out, right? In real time in these companies. So when you get alignment and when the CEO goes, "Right, I see now." Okay? 'Cause remember, CEOs are not specialists in this stuff, okay? They shouldn't be. They don't need to be. They have a job and it's called, you know, creating a vision for this business that people believe in, and a strategy that will help w- that will mean we can win.
That's their job. When they, when, when you align that for them and they go, "Right, awesome. So there's three things we're gonna use AI for over the next three years to help us crush our strategy," then investment follows, support follows, yeah? And suddenly, like the organization aligns and, and remember, there's no...
It's just not useful for organizations to be doing work that is not in support of the stuff that matters. So I've seen organizations like galvanize around the clarity, and I think that's the, like that's the defining moment, Tom, that you're describing. That's the difference. And, and what, what does it mean?
It means people have to go, "Oh, so this thing I'm working on is like, maybe that's not the thing, but the thing over there, that's the thing." And everyone goes, "Wicked. Let's do the thing."
[00:27:09] Tom Adams: So, yeah, so, uh, I mean, I'm just trying to process how, how... So you're generally brought into the boardroom, right? Or the board or the ex- the C-suite, and you're going, "You have to get clarity. You have to define a mandate if you want that." And at the same time, there's all these initiatives happening. What's the-- Once, once you get clarity, and once the CEO and the C-suite is behind this in service of the strategy and the mission, what, what, what's it take now to move that, um, from just being, uh, experimentation into some level of accountability scrutiny?
Because, like, somehow you gotta move from where we've been into this model that you're suggesting. But like, like, what-- how does that transpire? Or, because big organizations, again, we go back to it's a people thing. So I mean, now it's complicated 'cause now you've got a CEO or a C-suite who, who has a, a mandate.
You have people who've been testing stuff and it's not really getting anywhere, and now you gotta move an organization. How do you do, like how do you do that? What's
that process?
[00:28:20] Mark Redgrave: Yeah. People and change. So it's like we've, you know, we're looping back to the conversation. So you nailed it. It's about people and change. So the way this goes, right, is, is you create, you create the clarity with your executive team, okay? That then informs like programs and work streams that become strategically important, okay?
We-- I've definitely mentioned this on the podcast before. If something is important, make it important. So now we have a situation where at these companies, we have AI projects aligned to strategic priorities that are important, okay? They then get funded and we organize around them. Now, the activity that happens, like, like that doesn't all stop because sometimes you're like, "No, this is great.
Keep doing that. That's fine." At, at this team level, in this department, it's like, that's great. See if you can fig- figure that out. But the most important stuff gets prioritized in terms of investment, effort, focus, ownership. Then the CEO and the, you know, well, the CEO ultimately has to, has to create accountability across those work streams.
So, but everything we're describing, this was true before AI, Tom, and it's still true. So it's like, if you wanna get things done in your company, you have to organize around it. That requires people having clarity, inspiration, motivation, and engagement around like, why the heck you're asking me to do this thing.
So it's like, but, but this is, this is fascinating for me because like, like I've, I've been involved in, in transformation and, and performance and moving companies for the last 15 years intensively, and I'm like, this is, this is just like now we've, you know, it's, we, we're, we've, we've taken vitamins and we're going harder and faster than before, but it's the same
[00:30:12] Tom Adams: yeah, yeah. yeah. So one of the things I, I read you posted was, um, something about, uh,
so, we get the, we get the structure in place, right? We, we got a mandate from the C-suite, CEO C-suite. Uh, we understand that it's a people management change. We understand that we gotta bring people aboard. Uh, we start to, uh, have context for how it delivers value, but then we run smack dab into some of the challenges we've already talked about.
And one of the things you've posted about is this sense of the operating model becomes a bottleneck somehow to creating AI value. So what do you mean by that? What are
you seeing? What are some examples of that maybe?
[00:30:53] Mark Redgrave: Yeah. So I think, like we've, we've discussed a few of them actually on the way through, 'cause this is all I'm talking about, right? It's like for me, the operating model of your company is like, is okay, like, um, to, to-- how do you as a leadership team create the clarity and alignment the organization needs?
How are your teams designed? How do we do work, right? That's, that's operating model, a combination of these things. Culture, operating model. Okay? So, so like it's a, a central part of an operating model is the culture that underpins like how we work together. So a lot of what we described sits firmly in that bucket, right?
But I think it's, it's really interesting, like when we think about value creation in any organization of any size, it's like most organizations are still very functionally designed, right? By departments. They're very vertically designed. And increasingly we f- we're, we're, you know, we're, we're finding and thinking, and I definitely sit in this team, it's like that, that, uh, cross-functional team design is better.
So like when we think about end-to-end delivery of something, that's really important. And we've just seen this, we've been doing this work for many, many years, and to me it's conclusively true. So, so what AI's doing from an operating model point of view is saying, "Hey, like let's just..." If, if your organization is rigidly stuck in these departments, okay?
When you have an innovation opportunity that looks like AI, like those rigid departmental lines will stop you from driving rapid innovation and change through your business if you, if what's driving you is like some outcome, some customer-centric thing you want to do better. So, so that's a really big component of this too.
So when we're talking to clients about like trying to, trying to leverage the value of AI in those organizations, it's like it quickly gets to a conversation around team design, which is operating model. So I mean, it's, it's kind of, I think it's a, it's a little bit ambiguous term operating model, but everything we've just discussed for the last 32 minutes, right?
To me is operating
model.
[00:33:14] Tom Adams: So I mean, that, that's an interesting, I'd, I'd love to double-click on that one because so many organizations are in that model of sort of siloed, um, structure. Uh, how do you... Because if you can't derive value from this change that is being, uh, being brought on us by AI, even if it's mandated from the top, but you've got the wrong operating model, um, how does that get stuck?
Like, like, how do you see that getting stuck? 'Cause ul- ultimately it sounds like you gotta change that before
you can make the other thing work. It,
am I missing
something?
[00:33:49] Mark Redgrave: No, you're not. No. But, but like, so, uh, and there's a-- I was thinking when you were talking before, um, when, when I'm having conversations with leaders about this AI kind of challenge and opportunity, and we go down a path of this is not a technology challenge, it's a people challenge, the pressure comes off, right?
Because a lot of executives go, "Phew," because like, I-- how can I possibly understand the technical components of this? And you're like, "You don't need to." The good news is there's plenty of super smart people that can do that for you. Like, your success will depend on how you organize your business and your people around it.
So pressure comes off. And I think that's analogous to what you're just describing, which is, you know, like you don't need to change your whole company tomorrow overnight, okay? To be able to capture this value opportunity. But you should start thinking about like, how do we create cross-functional teams to execute the important stuff?
So, so, you know, if you're-- When I say operating model, if you're very functionally aligned, what you should do is you should sit down and go, "Right, our strategy looks like this. The opportunity for AI in our company is for it to support this strategic pillar for us." Yeah? "And that means we need to do this thing."
Now, end to end, what does that mean, right? And what you can do, and I've seen it done super successfully, is you create cross-functional teams to deal with that. And it might be that 90% of your organization is operating in silos the old way, but 10% of it you reorganize to capture value, okay? In the new way.
And, and you can call them anything you want. You can call it skunkworks, you can call it front-runner teams, you can call it this. It doesn't matter what you call it. But like, try working in a different way. And for most people, when they try that, they go, "Oh, this is very interesting because things are not getting stuck anymore."
Right? The whole concept is when you create cross-functional teams, is that you have people representing each step of the journey around the same table, right? With the same mission, and they're dealing with it in real time, face to face, in the moment.
No more, "I'm waiting for product for-- I'm waiting six weeks to get an answer back from product."
No more, "Oh, it's with the IT team on QA." No more, "Well, we need the marketing team to do this thing." It's like, no, everyone's there. So like you're all in the same team working together, and you're fixing it like in an hour, not in four weeks. So like, so this conceptually, like that's what I mean. You don't have to completely turn your business upside down.
[00:36:34] Tom Adams: Well, the, the interesting thing too in cross-functional is you begin to understand how work actually gets done much more connectedly. Like the, the handoff, like you said, between IT and product or IT and admin or, or finance and, and, um, you know, product development. Whatever those structures are when you're cross-functional, you begin to see how, process works, um, and then automation comes out of that.
Uh, and that's how the work
gets done.
[00:37:08] Mark Redgrave: Bingo. And you know what? Yeah, like statistically, right, why this is so important is, like all those handoffs and all of the inefficiencies that sit between teams in your company, and if you're a CEO listening to this, I'm telling you this is true, right? Account for 70% of the elapsed time. 70.
So
[00:37:26] Tom Adams: Hmm.
[00:37:26] Mark Redgrave: like only 30% of the elapsed time of any project in your organization is anyone actually working on it. For 70% of the time, it's in someone's inbox somewhere.
[00:37:37] Tom Adams: Right.
[00:37:38] Mark Redgrave: So it's like, it's like, come on, man. Like what, what, what, what CEO or executive team doesn't wanna talk about breaking that
[00:37:46] Tom Adams: Yeah. Yeah. Yeah, I, I, uh, I, I spend a lot of time in theory of constraints thinking, right? Uh, the theory of constraints and you go, well, they- the, the whole theory of constraints model is manufacturing, but ultimately it leads, where's the bottleneck
here?
[00:38:01] Mark Redgrave: Yep
[00:38:02] Tom Adams: If 70% of it is stuck in an inbox, like some of that kind of stuff, when you understand process flow, um, you understand how that works, and you can add automation to it, you can add AI to it, but you remove it from where it's being stuck.
If that's the thing that, that sometimes the tool
brings value in, that's where it gets kind of interesting and
fun
[00:38:22] Mark Redgrave: Yeah. And, and theory of constraint, I mean, like, if anyone's interested, read about it, 'cause it's like you-- it will make, make so much sense to you as a business owner. It's like it's such a powerful concept. And when we're talking about like, so, so we have a strategy, we know exactly what AI will do for us and what it won't, 'cause that's important.
'Cause part of the thing right now is people don't-- people have focused a lot on what it won't, like things won't ultimately help them. But it's like once you know that, it's like how do we create, like get the right people around the table, yeah, to be able to implement and prove this thing quickly? And I'm telling you, that does not like look like a functional siloed handoff over six months.
It looks like put a team together, we got eight weeks to crush it.
[00:39:08] Tom Adams: Right.
[00:39:08] Mark Redgrave: what it looks like
[00:39:10] Tom Adams: Yeah. Yeah, I love that. I love that. So, um, the, the people you work with, the companies you work with, without disclosing it, how are you seeing, uh, the ability to do what you just said, these cross-functional teams that are like other skunkworks or, or, or kind of experimental project teams that are cross-functional?
How do you see those start to unfold as you observe after you go in and, and kinda leverage, you know, give, give people a way to think about it? What do you see happening? Like, what... How does that work? Like, can people actually do things in eight weeks? Do you see that happening? Um, do you have any kind of experimental thing that you've seen occur
that you're particularly excited about how it's unfolded?
[00:39:57] Mark Redgrave: Yeah. I mean, so, so I do, you know, a portion of my work, as we spoke about before, is like I'm a, I'm a, a senior advisor for McKinsey, so, um, and that's, that's public knowledge. So, so I actually have the great pleasure of sitting across from some of the biggest, most exciting businesses in the planet doing really large scale transformation, right?
And the reason why that's so interesting is because I get to see how this gets done. And my mission, like with my work with Shift here on mid-market, is to take what I see that is applicable, okay, and, and bring that to bear into the mid-market. And what, like, motivates me hugely is that when you, when you give people permission, okay, to break some of these, um, age-old structural limitations in the company, people love it.
It's like, wow, like, like, well, b- well, I don't think that's a big surprise, you know? It's like people... What we normally see, large scale and small scale, is people, employees going, "Y- c- can I, can I do that?" Like, "Do you really? What, so I can be g- we're gonna make this thing, and like I can go and do that with these people?"
It's like, "Is that okay?" And you go, "That's okay. That's what we want you to do." And they mostly, well, I would say like nine times out of 10, the people are like, "That's super cool because that thing, that's really dumb. What we do over there, that's dumb." Yes. So, so I think the biggest, like the thing everyone should really take comfort in is like back to the people thing, right?
People want to work in a different way. They don't want to be stuck in their functional silo where h- work is hard. AI has turned up and everyone goes, "Look what this can do for us." And if people are, if people believe that, right? Now they're like, "Let's do it." And, and let's do it means if there's a way for us to think about this that's innovative and different, like it, I'm in.
And that's, we, we universally see that, Tom. Now it doesn't mean it's a walk in the park, but we universally see people go, "Oh yeah, man, like, like let me... Can I work with these people? Because that w- we can do it. If I, if I can work with these people, we can get this done." Yes, go.
[00:42:16] Tom Adams: so so related to that though, 'cause you see this unfold, um, so let, let's say you've got the s- Skunk Works team or a, an initial pilot team, and they do it. I- is it, it, is part of the people change the cultural, like, impact of that group of people who have just done something and everybody else goes, "I want what they got"?
Is that, that kind of
the, uh, thing that starts to unfold that
you see?
[00:42:43] Mark Redgrave: Yeah. I mean, you know, like it, it, it sort of looks a little bit different at very large scale because at very large scale, like you've got, um, you've got all these... You've got quite a lot of sophisticated structural mechanisms for change management and cult- So, so, but, but that's exactly what happens. And, and people, everyone in every company I've ever worked with, like wants s- progress.
Yeah? Like they wanna see things change. I, I literally have never met an employee that sat there and went, "Oh," like, "It'd be really good if we just continued to suck at this." Right? It's like, it's like ev- everyone goes-- M- most people you talk to, you go, "Yeah, I don't know why this is so hard, Mark." Like, because man, we make, we make this so difficult for ourselves.
And you go, "Yeah, okay, we can fix that." So it's like this, so but that's culture and, and, you know, don't expect to do it in five minutes, but like it can organically become like a really powerful thing. Just do some stuff, prove it works, and if the biggest problem you've got as a CEO or a leader is your organization is pressuring you to replicate that success, you got a good problem,
let me tell you.
[00:43:55] Tom Adams: Yeah. Yeah, that's really cool. So, um, uh, let's assume we, we sort of sprinkle magic fairy dust and we're two years out from now, and you and I are having this conversation. Um, what do you think we'll have discovered that really mattered the most in terms of successful AI adoption, value creation? Uh, what, what's, what's in your, you know, um, crystal ball that
you see unfolding?
[00:44:24] Mark Redgrave: Yeah. For me, it's this clarity upfront at CEO and executive team level on like what AI is and is not for my business. Because we're not there yet, Tom. We're not there yet. Like at the moment, I see most leaders kind of sitting and looking over the balcony at this frenzy of activity, and they're like, they're-- most of them are going, "I don't really know what this is."
So, so, so I would love to think that when we come back two years' time, it's like the companies, you know, they have tremendous value realization because they have thought long and hard upfront about how AI supports their strategy, because their strategy is the thing that matters. So, so that for me would be the, that's, that's the big determining factor in, in, um, I think the challenge that we see today.
But it's exciting, right? Like I'm not, we're, we're not, this, this is-- I mean, I think there's some very important decisions people need to make, and those decisions we've talked about this are coming fast, right? Like, don't sit like, like, um, don't sleep on this, ladies and gentlemen. Like, there's decisions you need to make.
But, um, so o- o- on balance, super exciting, but we're gonna get through this early frenzy quickly, and I think boards and, you know, CEOs are gonna be asking themselves, "Am I putting, am I putting another 500 grand, $2 million into this thing? Why, why am I doing that?" That's the que- that, that's the scrutiny's coming, right?
And, and I think it's gonna be here in the second half of this year.
[00:46:06] Tom Adams: Wow. Very exciting. So, uh, if you were to, uh... A CEO's listening to us right now, and we were to sort of distill all this down, 'cause I feel like you've been very clear about what you're seeing in the marketplace, both in the largest companies through your McKinsey work and, and your mid-market work that you're doing i- independently.
Um, uh, let's, let's talk to your, like, let's talk to a mid-market CEO now. What, what's your, your, what's your current advice for them today? Like, what do you, what do you tell them today, just as a, as a way to wrap this conversation?
[00:46:43] Mark Redgrave: have a conversation with your leadership team around how AI initiatives in the company are supporting your strategy. That's, that's the question. Now, if, if you can't really get clarity on that, that's what you need to do. That's your next step. Find clarity in that question. But it, like, that's all you need to ask.
And, and don't be... And you-- And w- I'm not asking that, like, to intimidate or to trick. I'm asking that because, like, that's the question you need to answer, right? It's like, um... And, and I did this thing, um, I, I, I actually didn't coin this, but I love the term, like, ready, fire, aim, right? Which is like a beautiful, like, saying.
Like, we're like, "Guys, we're, we're ready, fire, aiming right now." Yeah? Don't do that. Stop pulling triggers, okay? Like, let's ready, aim, fire. Much better
idea
[00:47:39] Tom Adams: Love it. Love it. Well, brother, this has been great. Uh, thank you for sharing your, your wisdom. It was kinda cool to be able to sort of do a deep dive with you individually, uh, on this. And, uh, I, I love what you're doing in the world and, and, uh, hopefully anyone who is listening, um, can, uh, derive some really cool value and take that question and go, "What is our strategy and how, how does it relate to, to AI, and how can we implement AI to support our strategy?"
Which I think is ultimately the, uh, the
question that everyone needs to ask. So,
[00:48:13] Mark Redgrave: Yeah. I, um, I, I was thinking, "Oh, so where are the, where are the other guys?" But I've just had fun for 45 minutes, so I'm like, good. I'd like maybe, maybe, uh, maybe this was an awesome way to spend Friday morning. So, uh, wherever they are, whatever they're doing, thank you for giving me the time with you, Tom, 'cause
it's been great.
[00:48:27] Tom Adams: Yeah. Fabulous. All right, brother, we'll talk soon
[00:48:30] Mark Redgrave: Okay. See you, man
