Bolt On AI or Reimagine the Business? Blaine Mathieu on Agentic AI Strategy

Mark Evans: Hi, it's Mark Evans, and you're listening to Marketing Spark. Today I'm talking with Blaine Matthew, the author of The River Doesn't Wait, a Senior Executive's Guide to Navigating the Surge in Agentic AI. Blaine has spent decades at the intersection of technology, strategy, product, and executive leadership, including roles as a CEO, CMO, startup founder, and advisor to enterprise leadership teams. His new book arrives at a key moment.

We're moving into a world where agenc AI systems can take action, make decisions, coordinate work, and reshape how companies operate. Now, for senior leaders, one of the key decisions is whether they should bolt agentic AI onto the way their businesses already work, or use it as a reason to rethink how work gets done in the first place. And that's where Blaine's River metaphor becomes useful. Leaders have to make decisions.

with incomplete information and decide how much autonomy they're prepared to trust. In this conversation, Blaine and I discuss why agentic AI poses a distinct leadership challenge, how executives can avoid both panic and paralysis, and what it means to navigate this shifts deliberately as the river keeps flowing. Welcome to Marking Spark. Thanks, great to speak to you, Mark.

And by the way, that intro was fantastic. I think my work here is okay. Thank you for that. As someone who has written three books, I know that they are mostly labors of love as opposed to commercial entities.

And of course, I have to ask you about the book. What was the inspiration and motivation? Where did the river metaphor come from? It certainly captures the ⁓ the spotlight.

And what made you reach for that image? Instead of the usual wave or disruption language that everyone else uses for AI. Great question. Let me deal with the motivation and inspiration first.

I've been wanting for that business for 30 years. It's really a book on corporate and business strategy. That's really what I fundamentally wanted to do. But over the literal decades, I never really find a hook the driver which would make this thing really interesting and meaningful.

And finally, mid-2025, I began to work on a summit on agentic AI that was held at the Stanford Research Institute actually in February of twenty-six. I began working on the project in mid-2025. Just when the Agentic AI thing starting to become real, I'd be on just chat thoughts and prompts and into more autonomous systems. And I started prep prevents, I actually put out an interview series of Quad leaders and technologists and others that are working at the leading edge of a genetic AI.

And as I was talking to these people and doing these interviews and f ultimately hosting the summit, I realized, wow, this is the driver, this is the inflection point. We'll get to the wave and the river in a second. But this is the thing which has to for enterprise leaders to think about their core operating strategy and their strategy in general. differently.

They have to be thinking differently. And this is how I can take 30 years of working even with enterprises as a Gartner analyst, playing Mark Mintel just at Adobe, being a startup leader, all almost entirely working with enterprise clients, take all that experience and all those learnings and really rethink it in this new world of agentic AI. To pretty to be very frank, the book is not an AI book. corporate strategy but above using the genetic AI as the driver, as the inflection point or the wave.

Let's talk about the wave question, because it's a great question. I seem to recall this was months ago when I began writing the book in earnest. I was originally thinking of things like waves and and obviously the ⁓ technology in general and market disruption and crossing the chasm and all that good stuff. We know quite well.

In fact, to be honest, I ran some of these early ideas past my son, who's an amazing writer and has been a technology marketer for his entire career. He saxes said that maybe this is not a wave because a wave is more of a one time we think of it as a one time big event or a disruption is being more of a one time thing that happened. Isn't this more water that's continually moving? And that was a huge aha moment for me because I realized that okay, this is not a way, this is a current.

You're navigating to to a toward your goals, you're navigating toward your corporate goals, and the current never stops moving. It changes speed sometime, and sometimes there are surges in the current talk about in the book. And overall it just seems a much more apt analogy because this is not about a one time event what's happening system we agenda AI. It's a continually moving series of events that we need to be able to respond to strategically appropriately.

So that's how this whole thing began. So I do want to get into the tsunami of Intec AI, particularly given the fact that we're three years away from Chat GPT launching. But one question that I want to ask is around strategy. My Thinking is that many companies have jumped into the ⁓ AI waters feet first when it comes to tactics, the tools, the efficiency, the productivity aspirations.

And it probably explains why a lot of the research houses are suggesting that many companies are not getting ROI from AI. Wanted to bounce that thesis off you. Do you think that companies are not doing enough strategic thinking when it comes to how they want to deploy and take advantage of AI? Is that a is that something that I'm misreading the landscape, or is it a problem that many companies are facing when they think about where's the ROI?

Where are the benefits that everyone promised? First of all, AI is in the desperate days. Gen AI for three years or so. Jettic AI, the ability to have these autonomous systems running and doing tasks similar and maybe in some cases to how a human team might do them is quite new.

And so it's not surprising that enterprises, organizations are doing a lot of experimenting, a lot of learning, or spending a lot of money on this stuff because they see potential impacts, but they haven't seen the ROI. Given that this whole space is maybe nine months old. depending on what inflection point you use as the start of it, it's not surprising that ROI is not that high. So having said that, ⁓ the most fundamental framework in the book is this notion of using agentic AI to retrofit the work you're doing.

And that's really at the more tactical end. So make your processes and systems maybe faster or somewhat more efficient, but fundamentally do the work in the same way to achieve the same goal. Or at the other end of the spectrum is a reimagination of the work that you're doing, reimagining potentially how you're doing it and even the goals that are possible to achieve using these new technologies and systems. Obviously, the closer you get to the reimagination end of that spectrum, the more strategic you might say the questions become that you need to be thinking about.

But having said that, I want to be clear, and the book spends a bit of time on this, there's no right answer. You know, the solution today for an organization is not to say, okay, we must reimagine everything. No, there's actually a lot of efficiencies and opportunities for retrofitting existing processing processes and systems with these new technologies of Gentic AI. But I think in the longer run, differentiation in the market, competitive advantage is gonna come from a reimagination.

At the retrofit side, at the tactical side, yes, you could cut a process by twenty or fifty percent. But then all your competitors can easily retrofit their own similar processes the same way the same technologies. So it's a very limited timescale for competitive advantage. It's on the reimagination side where we have the potential for a much longer term competitive advantage.

You're a senior leader, you recognize that AI Is something that you have to embrace. You have to figure out strategically and tactically how to integrate it into your company. What are the steps or processes that you need to go through to determine what parts of the organization should be retrofitted in the short term and what other parts need to be reimagined? It's a balancing act, but it also requires some deep thought.

And planning to make sure that you're making the right moves at the right time. Because a lot of companies can go crazy in a particular operational pillar and then realize that it's the wrong place to focus because there was better ROI in other places. What are those steps that a leadership team needs to take? It's a great question and probably the most important question because most of the book lays out very easy to understand and ingest.

frameworks that a senior executive can use to to help think about their strategy in the face of this change. And by the way, I just want a quick aside, this book was written for non-technical senior executive leaders. It's not written for the CTO or the CIO. It's not a technology book.

It's a strategy book for people that are faced with this technology that and having to make the strategic choices you're talking about. I think the most important chapter in the book is near the end. It's about how to effectively read the river. How to re-reset the navigation of your organization, your boat.

I describe it as your boat that's that you're navigating through these waters, and how to filter appropriately for the information you need to make the right choices. It's another reason I wrote the book, and it's actually why I have on the books companion website just a monthly briefing series where I provide updates on surges in the water. These are significant changes in the space of agentic AI that might cause you to want to sit back with your team a second and think, okay, should I be reimagining how the work is done?

Should I be rethinking the goals that I'm navigating towards from organization? Because this surge in the water is big enough that book that chapter in the book is effectively the outline of a workshop that a team could use to help figure that out. How to do that watching for surges. And then how to think about work through the strategic changes.

I've got a whole chapter about that. And the watching is especially hard right now because I guarantee any enterprise leader is getting just overwhelmed with the torts and substacks and all of this information about AI. And how do you which is really relevant to your business, which could trigger a reimagination, and which is just noise. probably ninety-eight percent of what's out there is noise.

So that's partly why I I s have this briefing series free. Anybody can sign up for it on the website, river doesn't wait dot com. And I go through and try to filter out the 99% noise and just hit on the two or three real surges that might have happened or that did happen in the last month. So that you can take that back to your team and say, okay, does this allow us to reassess Where we're going or how we get there?

We live in a world in which people want instant gratifications and overnight success. So I suspect that bolting AI onto existing operations is the easiest path forward. It's the way that companies and leadership teams can demonstrate to their investors or the board that we understand the impact of AI and we're on it. We got this cover.

But it also There's also a risk of making strategic mistakes or missteps. And I'd like to ask you about the most common mistakes that you're seeing senior executives make when it comes to bolting AI onto their operations. Are they overreacting? What are the mistakes that you see?

And how can senior leaders avoid making those mistakes? I'll characterize it as biggest challenges that senior leaders have. And if they don't meet the challenges, then that would be a mistake, obviously. But I think one interesting I think one of the most interesting elements I discussed in the book is this notion of organizational context.

And giving agents proper context is a quasi-technical term that people building AI systems have been talking about for a while now. But it's very much an organizational discussion. Because the domain or the scope of control and information that an executive owns is effectively that executive's context. And forget about Gen TIC AI.

If you want your human team to be working effectively, obviously you want to give them all the necessary context to make the appropriate decisions. Right. And that doesn't just mean access to the databases and the systems and the customer records. It means understanding the goals.

of your organization. It's the long term strategic plan. And so they can make the hard judgment calls that the regular rule set doesn't, the formal rule set doesn't necessarily apply to. The challenge with agentic AI is because most enterprise leaders think about technology as being deterministic.

You said if this, then that, the this is the concrete set of rules applied to this database, go. Agentic AI using based on generative AI technologies enables a different kind of decision making, a different kind of action taking. That's what agentic AI is all about. But and to do that, you have to provide the same kind of context that you would provide to any team that you're running.

And so a challenge and maybe a mistake that some leaders are making is thinking that this kind of technology is just a tool. And if I give it a very narrow context, it's going to be safe and effective. And in fact, in many cases, it's the exact opposite. The more context you give it, the better it understands your overall goals and objectives, the better it understands the place you're navigating your boat toward and the parlance of the book, the more likely it is it's going to do the right thing at the right time.

In conjunction with the rest of your organization. So that's one of the key challenges, and it's a really interesting fact, I think. Another thing that I think about is biases and assumptions based on experience and expertise. You have senior leaders assuming they're in their 40s, 50s, and 60s, they're digital creatures, but they're certainly not AI creatures.

My 19-year-old son is a digital and AI teacher. Creature. He thinks of AI as just another way to do work. And it's not something new.

It's just, it is what it is. But a lot of senior executives come in with prejudices, or we've always done things this way. This is the way that technology is implemented. This is the way that it rolls out.

How do senior executives overcome those biases when they're thinking about something that is so new? And to your point, it's a river. It's flowing, it's moving, it surges, it calms down. How do you strategically make the right decisions when you may be operating with something that's beyond your domain expertise or skill set?

First of all, that's specifically why I wrote the book for that audience to help that audience, maybe the 50-something senior executive VP-ish or C level exec in a mid to large enterprise, how to be able to think about this. These changes that are happening right now to help address that is really exactly why I wrote the book. But that also relates to my comment about the workshop at the end. When you're working through this, what I'm encouraging is that senior executive who now is a great think about this, to involve her wider team.

Bring those 30 somethings or 20 somethings that are working at the leading edge of this into the room when you're trying to figure out how it impacts your technology, your school. Not technology. How it impacts your strategy, your direction, your goals. Again, this is not a technology discussion.

We have more than enough technology right now to do some very incredible things. The lag is in our imagination, reimagination of what's possible. You make an excellent point. Involve those 20 somethings, those AI natives in the discussions, because that'll help you figure out if a reimagination is possible.

It's an interesting situation because when you think about who's at the table when it comes to making key strategic and tactical decisions, it's usually the most experienced people or the smartest people. It's not necessarily the youngest people, but the reality is people in their 20s who live and breathe this stuff, who embrace it naturally, are the probably the ones who have the the most Or the best insight. They're the ones they're on the ground. They see what's happening.

They're using the tools. They're experimenting without even thinking about it. Is that a hard thing for senior leaders to accept that you do need people at the table who may not have a lot of experience, but they have a lot to say? I think intuitively it it makes sense that it would be harder for a senior executive to break out of their normal operating procedure.

Having said that, most of the very successful senior executives that I've met and worked with over the decades are already have already doing that. They're already very open to ideas and the wider team, and they've created virtual teams that are not just their peers. So if they're not already doing that, then they've got a challenge, they've got an extra challenge that has nothing to do with AI. But the other thing your comment brings into mind is besides the fact that it's the more AI native generation, which is living at the leading edge of this.

The other group that is living at the leading edge of this are the CIOs and CTOs and the people that actually did not write the book for on purpose, because they are becoming effectively very important strategic leaders of the organization. 10, 20 years ago, a CIO was focused on how to implement a cloud application and ensure the desktop apps are being updated. Copately and all that kind of stuff. A lot of back office stuff that's important, but not strategic, not truly strategic.

Now these CIOs and CTOs and chief AI officers and a lot of titles out there, they are in many cases driving, if not owning, the AI strategy of the organization, their critical component now. And of course, I'm not trying to push those guys off that, off that. stage that they've been given that's fantastic that technology is now core to corporate and enterprise strategy. But should the CIO of the organization actually be the main decider of how you're going to reimagine how your organization works, where it goes and how it's going to get there?

I think they should be a partner in that. They shouldn't be the owner of it. Another reason why I wrote the book was to make sure the non-CIOs and CTOs can co at least co-own for where they're driving this boat. Another issue for senior managers or another reality for senior managers is control.

They want to have strategic and tactical oversight for their organizations. You could argue that the whole return to work movement is about control versus efficiencies and productivity, but that's another story for another day. Don't get me started. So you juxtapose the control theme.

I oversee this organization. I make all I oversee all the key decisions, with the reality that. When it comes to AI, there's a certain sense of autonomous systems. So if you're a senior leader who isn't technical, is there a way to decide or test how much autonomy is too much too soon?

As a bit of an aside and to start, I always have to be careful myself, even that I'm not anthropomorphizing these systems overly. They aren't people. So let's be clear about that, although they're capable of doing many of the things that teams of people can do. They aren't people, so I just want to be clear.

And if I fall into the context of talking about them in the context of being an employee of the organization, that's not literally true. The reason I bring this up is you raise a really good topic, and there's a whole section of the book about governance and autonomy. And this is actually, you mentioned bolt-on versus built-in. And that's actually the section that concept is most relevant to.

Bolt-on governance after the fact onto autonomous systems, in which case it's producing issues and errors, but maybe you're trapping them, maybe you're correcting them before they get too large. Or are you building autonomous systems with governance built into the system itself? And mainly what that means. Is you're providing the appropriate context for this system.

So if the system in inherently understands what's right, what's wrong, what the goal is, so what moving toward the goal would look like, then it's far less likely that you're going to have an issue or a problem with this autonomy in the first place. We spent a lot of time talking about that in the book. And by the way, again, and back to my theme that this is not really a book about a gentic AI. Everything I just said is one hundred percent relevant to teams of human workers as well.

How much autonomy do you give them? How much governance do they need? The appropriate context. You could literally remove the words AI from the entire book and it would still be a valid book on corporate strategy.

But the big change is the speed and pace at which this river, the currents in the river are changing and surging now. That's why we really have to think about this and work in a new way than ever before. No discussion about AI cannot include comments. I know where you're going.

No discussion about AI can really ignore the realities of the impact on employment. It's been fascinating over the past few weeks or past few months to see organizations that stripped out hundreds or thousands of employees to suddenly rehire. People. Ford is a great example.

They got rid of all the gray hidden haired engineers because they wanted to embrace AI and all of a sudden they're hiring 300 of them back. AI is being used as an excuse. I think it's called AI washing, where organizations like Block hired way too many people and now they want to pare down the organization to make investors happy. So they're using AI as an excuse.

You're getting a lot of that happening. So what are your thoughts on the impact of a gentic AI on unemployment? And both you and I have young kids, so it's probably something we think about when it comes to not only our own careers, but the futures of our children. Yeah.

What are your thoughts? Honestly, even hearing the question send shivers down my spine. That's the very last chapter in the book is a reflection on that and what all of this means for the executive. Personally and what all this could mean for society at large.

It's a corporate strategy book, don't worry. It's not a philosophy book. But I do address this because I think we all care about, as you said, I've got kids and I've got a new grandson, even. Congratulations.

Thank you. I'm very concerned about the future we are building as government leaders, as corporate leaders, specifically on the question of employment. I don't know. Is the honest answer.

Nobody does. You can read one study which says if you try to extract all the corporate AI washing that's going on, just I was gonna do a layoff anyway, and let's use AI as the excuse. It's very hard to determine that so far there's been actually any actual job, net job loss due to AI. I'm reading a really interesting book right now by a guy named Brian Merchant called Blood on the Machine.

It's about the Industrial Revolution and the Luddite movement. And people who hear the word Luddite think, ⁓ anti-technology activists. It's actually a very interesting analysis of the change that happened, particularly in the cloth and cotton industry in the early eighteen hundreds. Although at the time this was truly a revolutionary technology which would have potentially and did have massive impact on employment in England in particular.

At the end of the day, in the longer run, employment in England didn't drop or unemployment didn't spike up to fifty percent. England has its own economic challenges now, but fundamentally the cotton gin and these things didn't cause that to happen in the long run. In the short run, there was a lot of disruption. I think that's fundamentally what's gonna h mo most likely to happen here is in the near term, and by that not the long near term, but over the next three to five years, there could be a lot of disruption.

To the way organizations work and employ people, governments are gonna have to find a way to work more rapidly and to think about how to help us bridge this gap. But in the long run, do I fundamentally believe that people, humans, will have an important place in organizations and enterprises and societies? I absolutely do. It's the next five, maybe ten years that that could be really rough.

But I'll tell you what, in the long run, the thing I'm more concerned about. Is not employment and having stuff for creative people to do. It's actually the effect on the ability for people to think. Because if we move to the place where we are basically offshoring to AI, all of the hard thinking and all we're doing is just asking questions.

That is absolutely, in my opinion, gonna atrophy our ability to engage in those kind of complex. thought processes and discussions and strategic thoughts as we're talking about. I was able to write this book because I had 30 years of real world experience behind me where I've been dealing with and thinking about and working through these hard issues without AI being the main driver, but many of the same issues. Now I wonder if I was a 20 something who's now entering the work, the corporate world, And or say a knowledge worker and who has these tools and this intelligence at their disposal, would I have been able to write this book myself?

I'm very concerned about that. And so we have to work through this challenge of retaining what makes us creative, what makes us intelligent, what gives a human being this into this inherent ability to produce interesting and valuable output. And not let AI take that away from us. Fascinating conversation, inspiring, terrifying, enlightening, all those things, but that's the reality of this ⁓ AI world in which we live.

Final question where can people learn more about you, your book, and your work? Probably the easiest place is the book's website, The River Doesn't Wait or RiverDosantWait dot com. Either will work. And you can also check me out on at Blaine Matthew on LinkedIn.

I'm easy to find there as well. Thanks, Blaine. And thanks to everyone for listening to another episode of Marketing Spark. If you found this conversation valuable, subscribe on Apple Podcasts, Spotify, or your favorite podcast app.

Drop a quick rating and share it on social media. And if you're a B2B or SaaS CEO struggling with stall growth and frustrated with marketing, we should talk. You can reach me by email, mark at markdevins.ca, connect with me on LinkedIn, or visit marketingspark.

co. Talk to you next time.