Author of Hyperadaptive: Rewiring the Enterprise to Become AI-native – Part 2

Tom Stoneman:
Hi, I’m Tom Stoneman, and this is The Intelligent Enterprise, where every two weeks, we take a break from the chaos of enterprise life and get inside a big idea by getting outside of it. Each episode, we meet an industry expert who helps cut through the noise from all the updates and rollouts while exploring one of their favorite break time activities. It might be over a game of ping pong, a workout, or an afternoon on the water, something that gives them some headspace when they’re deep in a problem.
This is part two of my conversation with Melissa M. Reeve, author of Hyperadaptive: Rewiring to Become an AI-Native Organization. In part one, Melissa laid out the case that surviving the AI era isn’t just about adopting the right tools. It’s more about fundamentally rewiring how your organization operates. She also told me how getting out on the water in her kayak helps her to stay grounded, so to speak, when the pace of things gets to be a lot. If you haven’t listened to episode one, be sure to check it out.
Today, we get into the practical side, how leaders actually move their organizations forward, why middle management may be more important than ever, and what it really takes to build the kind of enterprise that doesn’t just keep up with AI but is built for it.
Melissa M. Reeve:
What doesn’t work is having a bunch of outside consultants come in, rewire our processes, and then hand them over to the people who have to do it on a day-to-day basis.
Tom Stoneman:
Let’s get inside the future of enterprises by stepping outside of them.
I’ve had a chance to go through your book a couple times now, and I have to say there were a few things about this book. I loved everything about it, by the way, and I’m usually pretty critical, and, really, this is a real good roadmap, a good outline of how do I implement AI? How does my organization do this? I think you really mapped it out in a nice organized way.
So one thing I want to mention here, because your book has stage one, stage two, stage three, personally for me, stage two felt it really hit home. And I know there’s leaders, and I see it sometimes just talking to my peers and others that there are leaders who feel stuck in the early stages, even at stage zero, which must be frightening to see all this happening. So what’s the first structural shift that you think they need to make that a leader would need to make or leadership would need to make to start that ball rolling to move forward to get to stage two?
Melissa M. Reeve:
Yeah, and I think that the miss here is that we’re treating this like a software installation and we’re like, “Oh, it’s like Microsoft Word rolled out, so let’s put it on everybody’s machine. Then let’s get people trained on it. And over time, they will gradually pick it up.” It’s actually more like putting the PC on somebody’s desk, this powerful new machine.
And somehow in the 1990s, we figured out that that requires infrastructure. We fired up our IT help desks. We fired up entire IT departments to support these machines, to support networking the machines, and we need to fire up support infrastructure for the humans in our organization. And I think people who have been in the transformation and change management space are intuiting their way to many of the concepts in the book.
So, for example, those who are familiar with John Kotter and Leading Change, they’re like, “Okay, we need to pilot this, [inaudible 00:04:14] wins. We need our AI leads.” And so they’ve appointed their AI champions to get those early wins, and they’re happening, but it’s not scaling. And one of the reasons it’s not scaling is because we’re not treating this programmatically. So many organizations that I talk to are appointing their AI leads or AI champions in name only.
And so they’re not necessarily giving them dedicated capacity, not only for their own learning, but to spread the learning to others. They’re not necessarily programmatically selecting their power users for their ability to change and influence others. And that’s a real difference. When you look to the agile world, and of course, I have roots in agile, we did a great job of that. We identified those early people who were embracing this new way of working. We identified the people who had either social influence or were natural trainers to lead the way for other people.
And in the book, I advocate for something similar is we need to treat that programmatically. And I advocate for something called an AI activation hub, and we can talk more about that, but it’s just a way for us, again, to programmatically support the humans in our organization and the spread of AI on an ongoing basis. Because unlike Microsoft Word, which you might get … Microsoft Word, I don’t know what version they’re up to now, 2016, that’s going to be good for a few years. AI just doesn’t stand still. And so we need a way to keep people up to date on an ongoing basis.
Tom Stoneman:
What about the AI collisions with legacy systems? So there’s the legacy owner’s reluctance and then the lack of peer learning systems. What do you see happening there?
Melissa M. Reeve:
I’m glad you raised this point, too, because it is a collision, and it’s a collision at multiple levels. You think about that leader who really sees the vision around agentic orchestration, and we want to get there as fast as possible. And I think what stage two invites us to do is take a breath, be deliberate, and build some skills on the way to that agentic orchestration. Because what I’m actually seeing is that the organizations that hit the gas too fast toward that agentic orchestration without building the foundation I start to articulate in stage two, they might get 60% or 80% of the way there, and they realize that they don’t quite have it.
Either the technology’s not quite there, they don’t have the internal skills, their data’s not quite ready. And now what they’ve done is they’ve burned through political capital, actual financial capital, and a lot of goodwill as they’ve charged ahead into this future. And so what I advocate that people do in stage two is really start with where you’re at and start with your legacy systems. Really start exercising muscles, muscles like process mapping, process optimization.
We learned from business process re-engineering that what doesn’t work is having a bunch of outside consultants come in, rewire our processes, and then hand them over to the people who have to do it on a day-to-day basis. And so what I start to advocate for in the model is how do you teach people to really look at their processes through this lens of AI? Because what we know for sure is that these processes will have to be reinvented over and over and over again.
And so we’re starting to lay this foundation to empower the frontline people to really look at those processes, think about how AI starts to layer into those. And we’re just starting to experiment with the legacy systems, with things like Claude Code, with the processes, learning what the boundaries of this new tool called AI look like. And that’s how we’re raising literacy. And from that foundation, we now have a springboard into which we can start thinking about and re-imagining what reinventing these legacy systems look like with local knowledge around what AI can do, what it can’t do, what we have to look out for. And I think it just provides a nice bridge from where we are now into the future.
Tom Stoneman:
There’s been a lot of talk lately about AI flattening organizational hierarchies, fewer layers, fewer managers, more autonomous teams, and new agentic teams. Melissa’s take is more nuanced than that. The layer most people assume is disappearing may actually be the one that determines whether AI transformation succeeds or fails.
Melissa M. Reeve:
This is where nuance is so important. As humans, we want things to simplify things. And so the simplification is, well, AI can do all this stuff that managers do, and so we don’t need managers and the hierarchy’s going to collapse. And so bye-bye middle management. Well, okay, first of all, that’s a pretty toxic message to all of your current middle managers.
Second of all, I mean, I do, in the book, describe this shift, and I do believe that AI is going to compress the hierarchy and that middle management layer is going to be more important than ever. It just might not look like what we call today the middle management layer. So Harvard research shows … It used to be how do we change the organization? And there was top-down, and then there was something called bottom-up. And so top-down … What the research from Harvard showed is top-down doesn’t always work because it’s disconnected from the frontline reality.
So when these top-down directives come in, they often don’t find footing because there’s real problems on the front line that keep them from being executed. And bottom-up struggles, because these ideas are good and they accommodate operational reality, but the folks who are doing them may not have the political capital or authority to really get them to spread across the organization. And so what their research found is that actually middle out led to more successful transformations. And it’s because your middle management layer can tie that strategic vision to the operational reality.
So they’re really the translators. So when we’re talking about AI, we want to activate that layer. And you alluded to at least one of those ways to activate them, which is what I call the AI activation hub. And what we don’t want happening in the organization is everybody spending hours a day trying to figure out what the heck is new with AI, how it impacts their job, what are the success patterns and best use cases? And so we consolidate that activity into these AI activation hubs, and these are fractal.
So if you’re a small organization, you might just have one or two activation hubs. If you’re a giant organization like Tata Consulting, you probably have an entire network of these activation hubs that are translating the advancements of AI into local context for that part of the business. So that’s one area that middle management can go. The other area, one of the other areas, is in the building, monitoring, and maintaining of these automations. Somebody who’s a middle manager might be like, “You know what? I’m kind of over-managing the people side, and I got promoted into this position, and I want to lean back into my roots and extend my capabilities into AI in a new way.”
I think a third way where we still need that middle management layer is the alignment. And I think people are just starting to wake up to the fact that when AI can do anything, that means there’s thousands more things that we have to decide between. And which of those thousands of things are most aligned with our strategic direction? What happens when you still need 10, 20, 30 people to build an orchestrated agent? What happens then? And who decides what the portfolio initiatives are?
So I think that we still need people to decide what work is going to get done, what’s the highest value work, where do we put our scarce resources? It’s the same as it ever was. And I don’t think there’ll be quite as many people needed, but there will still be some and there still are human issues. We still have humans in the loop, which means we have people who need to watch over your growth. We have people whose jobs are changing. We have people who need to be redeployed on a regular basis because of that change. And all of that is going to require the skills that managers have been building for decades.
Tom Stoneman:
In your book, I know you mentioned something about technical and social orchestration, which is interesting. So managing both the technical implementations and the people. What kind of shifts do you see in leadership? What do you think will happen here with organizations where, typically right now, communications come down, leadership down, and everybody understands that they go and they go about their business? What kind of changes do you see in communications now with AI in stage two?
Melissa M. Reeve:
I’ve actually yet to come across, I’m sure they’re out there, but I’ve yet to hear a really compelling story about an organization that invested heavily in an AI communications plan. And what I mean by that is they not only had what I call that AI north star. So this unifying theme about what we’re trying to do with AI, but then they know how to empower the leaders to deliver that message to their teams, their divisions, their parts of the organization. And there was just some recent research that came out. It was done by an analyst firm called RedThread Research.
And they took a look at how people learn in organizations. And what surprised me about this research is that people learn most commonly directly from their managers. And so it really begs the question, how equipped are the managers at any level to have these conversations around AI? And they’re tricky conversations because AI is really poking at people’s identities. If you have been a coder for 25 years and you love the art of code and, all of a sudden, somebody says, “Well, Claude Code can do that in a fraction of the time,” that really starts to disrupt who you anchor yourself as.
And so who’s equipped to have that conversation with you to say, “Well, it’s okay. Here’s our stance as an organization. We’re here to improve quality or really create amazing customer experiences. We’re going to be with you along this journey. And, yes, we anticipate there will be disruption. We are committed or not committed to helping you as your role changes.” And people just need to have that type of communication from their manager in order to really create that psychological safety that’s needed to embrace these new tools, new ways of working, and even new types of roles.
Tom Stoneman:
Who do you think is getting that right? Do you have a favorite?
Melissa M. Reeve:
I like to point to Unilever, and I also like MetLife. And I like those two examples. I know less about MetLife than I do about the Unilever program, but it’s this understanding that jobs are going to change. And it’s this understanding that we’ve hired good people with solid skills.
And so what they’re working to do is they’re working to, first of all, identify what drives their people. So you, as an individual, get to articulate your purpose. Why am I here? What dent do I hope to make in the universe? The next thing they do is they say, “What skills do you bring to the table?” And skills can be anything from, “I can code in Python,” to, “Hey, I’m really good prioritizing work,” to, “I know how to be calm when things get heated.” And then on the other side of what’s going on is they take a look at their initiatives, and they say, “What skills do we need to complete these initiatives?”
And they use an AI matching algorithm to match people to the opportunities. And I feel like what happens in an environment like that is we concentrate less on career ladders and more on career portfolios, because now I get to engage in work that gets me up in the morning and has interest to me and matches my skills. Whoa, that starts to sound like an entirely different workforce, doesn’t it?
Tom Stoneman:
It does. I love it. Yeah. I wish I were 30 years younger and could go through that.
AI is the most powerful tool most of us have used, but even those of us who can’t wait to sing its praises need to know when to put it down. We heard in part one how Melissa recharges out on the water in her kayak. To close our conversation, I asked her where else in her life she likes to keep AI out of the picture.
Melissa M. Reeve:
I’m in that process right now. It’s interesting you ask that. So let’s keep it off the kayak. Let’s keep it off my hike. Let’s reground in nature. I am starting to realize that AI feels as addictive as social media. You hit the end of a prompt, and there’s something else right there queuing you up for the next thing to do. And there’s so much that can be done, and it all feels simple.
And so the temptation is to keep going, keep going, keep going. So I’ve had to draw some pretty hard lines. When I’m done with work, I try not to open up the computer again, whether it’s 5:00, 6:00, whenever it is. I can check it on my phone, but I try and keep it out of there. I try and keep it out of family time, and I try and keep nature. I hold the boundaries at nature, too.
Tom Stoneman:
Boy, you were so right about the what’s next, what’s next, what’s next? I was doing something the other day on Claude, and I think I was on Gemini because Gemini does this a lot. And it says, “Would you like me to look this up? Would you like me to take a look at this other aspect or whatever?” There’s always a question, and it reminded me of streaming. There’s always a cliffhanger. Oh my God, I have to watch this next episode. I have to watch this next.
So I’ve had to stop myself, too, because I do, yeah, I do want to know what that next thing is and the next one, but I can’t sit there all day. One last question. Can you imagine something that AI will be able to do either in the near future or maybe further out that’s never been done before?
Melissa M. Reeve:
Yeah. So I have thought about this because somebody challenged me the other day with a similar question, and they were like, “Well, where do all these jobs go?” I’m famous for citing this World Economic Forum statistic that says there’s going to be a net new 78 million jobs that will be created. And so she challenged me, and she said, “Well, okay, where? What happens?” And I said, “Well, I spoke at this event tonight. And while I love exploring the world, for something like speaking where you’re on stage for an hour, it gets to be a lot. Time zones, all of that. So I would love for somebody to be able to create a holograph of me where I can be speaking in Chinese on a stage in Beijing and yet be here in my office in Colorado. We’re not that far from it.
These AI engines can create images. We’ve got the avatars going in software like HeyGen, the video avatars. We can reproduce voice. Certainly, we can cross distance with sound and video. And … So is that so far off? I don’t know. And it would be a whole new industry. It would be a bunch of new jobs. And, gosh, wouldn’t that be cool?
Tom Stoneman:
That’s amazing, Melissa. I love that. I could see it actually translating to almost anything where you want to broadcast to everybody. I could see music concerts being broadcast like that all over the world.
We’re going to close up for now. Hopefully, we’ll be able to get you back here to talk about stage three. But Melissa, thank you so much for coming in for part one and part two so far. This has been just enlightening, and I think, really, it’s going to be something a lot of people want to learn about. So appreciate your coming.
Melissa M. Reeve:
Thanks so much, Tom. It’s been such an interesting conversation, and I look forward to diving into stage three.
Tom Stoneman:
The hardest part of adopting AI isn’t the technology itself. It’s the people, workflows, and foundational structures you must establish for these changes to last. That’s Melissa’s argument. And after two conversations, it’s hard not to agree. The enterprises that will succeed with AI won’t be those rushing to adopt every tool available. True long-term winners are building the operational foundations, clean data, clear processes, and skilled teams. That’s what’s required to scale AI to last.
Stage three is still ahead, and we look forward to having Melissa back to map it out. Thank you for listening to The Intelligent Enterprise, a podcast where we get inside big ideas by getting outside of them. I’ve been your host, Tom Stoneman. Please remember to follow the podcast and leave a comment or review wherever you get your shows. See you next time.