TL;DR: Most organizations see the “AI wave” coming but fail to ride it because they treat innovation as an extension of their legacy business. To survive, incumbents must decouple their Second Horizon execution from their primary operations. By leveraging the Six-Stage AI Maturity Model to assess readiness and the Opportunity-to-Value Framework (OVF) to enforce outcome-led delivery, leaders can shift from tech-heavy “demonstrations” to scalable Decision Velocity.
During periods of business disruption, the organizations that survive – and even flourish – are rarely the ones that saw the change coming first. They are the ones that had the discipline to act on what they saw. That distinction is everything. Because in my experience, most organizations see the wave. Very few know how to ride it.
What separates the survivors is a deliberate, structured approach … what I call the two-horizon strategy.
What do I mean by that?
These companies operate within their existing business model – real revenue, real customers, and real infrastructure – their current paradigm and their first horizon. This is something they need to continue to protect. But, these companies also need to recognize that there is a new paradigm emerging. This was exactly the mistake Blockbusters made – dismissing streaming. But recognizing this second horizon is one thing, what they do about this is quite another.
Nokia understood smartphones were coming. They even had prototypes. But their entire organization – culture, incentives, sales, engineering – was built around the first horizon. The result was that they lost the smartphone market entirely, despite being the dominant mobile phone company in the world.
There is a tension between these two horizons which is exactly why many incumbents get into trouble. Many treat this second horizon as an extension of the first, and they optimize themselves into a corner. Take Kodak for example, they invented the digital camera in 1975. Their engineer Steven Sasson built the first prototype in-house. Leadership saw it, understood it, and then buried it — because it threatened film. For decades they treated digital as an extension of their existing business, asking “how do we make digital printing profitable?” rather than “how do we rebuild around an image-first, filmless world?” They optimized the first horizon so aggressively that by the time they took the second seriously, it was too late. Kodak filed for bankruptcy in 2012. The irony? The very technology that killed them was invented in their own labs.
The companies that navigate these transitions well are the ones that explicitly fund and resource the second horizon separately – not as an extension of the first. Let me illustrate this by referencing the world I have inhabited for most of my career: geospatial. The old value proposition of this industry, actionable insight, is today a major bottleneck. The decision-velocity world – now my main area of focus – does not need a better location layer, it needs a fundamentally different architecture where location is an embedded input, not a delivered product.
I have written more on this shift in two recent pieces — if the geospatial angle interests you:
- Beyond the Map — Geospatial Industry Leaders are Exploring the Emerging World Model Era
- World Models — How GIS Can Realize its Geospatial Nervous System Vision
This is the gap I have spent the last several years trying to solve. Recognizing the second horizon is necessary but insufficient; the real challenge is execution. Most organizations simply do not have the architecture, the process discipline, or the commercial frameworks to run two horizons at the same time without one slowly strangling the other. That is why I built the Opportunity-to-Value Framework (OVF). It works hand in hand with a second framework I have developed: the Six-Stage AI Maturity Model that helps organizations understand where they actually are on the path from data silos to anticipatory intelligence.
Together, these two frameworks give organizations both a map of where they are going and the operating system to get there. Not theoretical models, but things I have built, broken, rebuilt, and actually used, in the kind of complex, data-intensive environments where the cost of getting this wrong is very real.
Execution is where second horizons go to die
Unless they have been living under a rock, most organizations realize that AI is reshaping the landscape and, as I have discussed, building a parallel second horizon is vital for incumbents survival and ongoing success. In this section we will discuss execution of the second horizon.
Why do most AI projects fail?
Most AI projects fail because they are designed as technology demonstrations – narrowly scoped, disconnected from real business outcomes, and never architected to scale across the organization. For most organizations, AI adoption has been technology-led rather than outcome-led, and that distinction explains most of the failure. The exceptions tend to prove the rule: the organizations scaling AI successfully started with a business problem, not a technology.
Closing this gap requires two things working together. Firstly, knowing where you are, where you want to get to and having a disciplined path to get there. Before any organization commits serious resources to AI, they need an honest answer to two deceptively simple questions: where are we today, and where do we actually need to get to? Without that baseline, AI initiatives launch blind: no shared understanding of readiness, no clarity on data gaps, no agreement on what success even looks like. Most organizations skip this step. And this is the goal of the Six-Stage AI Maturity Model.
That is usually the first mistake.
The second mistake is having no structured, end-to-end delivery mechanism built around one core principle: start with the business problem, not the technology. And that is why I built the Opportunity-to-Value Framework. What makes it different is three scaling value gates embedded throughout the process, each one forcing a critical question before the project moves forward. Does this actually solve a high-value business problem? Can we technically execute this at scale? And critically – will people actually use it?
These are the questions that most AI projects never ask until it is too late.
Together, these two frameworks replace the ad-hoc, technology-first approach that kills most AI initiatives with a disciplined, outcome-led path, one designed to scale from day one, not patched together for scale after the fact.
These are not theoretical models. They are frameworks I have built, broken, rebuilt, and actually deployed in complex, data-intensive environments where the cost of getting this wrong is very real. Most organizations do not need another consultant telling them AI is important. They need someone who can walk in, tell them honestly where they stand, and build the engine that gets them to where they need to be — before the window closes.
The second horizon does not wait. And neither, in my experience, do the competitors who are already building it.
Matt Sheehan
Matt Sheehan is a senior executive and AI strategist with over 25 years of experience leading complex organizations through technology-driven transformation. He specializes in moving AI from experimentation to enterprise-scale value — closing the gap between technology capability and measurable business outcomes. Matt has built the frameworks, the teams, and the delivery systems that turn AI pilots into scalable operating models, with a particular focus on decision velocity: compressing the distance between insight and action in environments where the cost of being slow is real.