KEY TAKEAWAYS
95% of AI pilot programs fail to deliver measurable P&L impact (MIT, 2025)
84% of AI project failures are leadership-driven, not technical (RAND Corporation, 2024)
Companies partnering with specialized vendors succeed 67% of the time vs. 33% for internal builds (MIT, 2025)
The gap between wanting to lead and being equipped to lead comes down to six things.
Every week, I sit across from an organization that wants to lead the AI charge.
And every week, I notice the same pattern: there is a real gap between the passionate desire to lead and the discipline required to actually lead in anything, not just AI.
The numbers confirm it. According to MIT’s GenAI Divide: State of AI in Business 2025 report, 95% of AI pilot programs fail to deliver measurable impact on a company’s bottom line. The RAND Corporation found that over 80% of AI projects fail, and that 84% of those failures are leadership-driven, not technical. Let that sit for a moment. The technology is not the problem. Leadership is.
Meanwhile, MIT’s research also found that companies who partner with specialized vendors succeed roughly 67% of the time, while internal builds succeed only about a third as often. The difference is not budget or talent. It is approach.
Recently we had an experience that reminded me why so many organizations stall before they ever get started. So I want to share six things that separate the companies who move from those who get stuck.
1. You Are the Subject Matter Expert in Your Own Problems
This is where most organizations get it twisted. They walk into a room expecting the solutions expert to also diagnose the problem. That is not how it works.
You live inside your business every single day. You see the inefficiencies. You feel the bottlenecks. You know where the gaps are. No one from the outside can identify your problems better than you can. What a solutions expert brings is the ability to build the answer, but only after you have done the work of clearly defining the question.
The companies that win? They walk in and say, “Here is the problem. Here is what it is costing us. Build us the solution.” The ones that struggle sit back and say, “Tell us what you can do for us.” That is not leadership. That is delegation without direction.
2. Once You Name the Problem, the Entire Conversation Changes
Here is what actually happens on most discovery calls. The expert is sitting in the room, ready to solve. They are not there to guess. They are waiting to hear your problem so they can build the answer. But instead of naming the problem, the conversation turns into a loop.
“What is your biggest challenge right now?”
“Well, let me give you some background on our philosophy…”
“Understood. But what is the specific problem?”
“We have been looking into a few things and wanted to understand the details of your platform…”“Of course. But what problem are we solving?”
Then someone sees a demo or a use case from a completely different industry and says, “Oh, we could do something like that, but cheaper, and with a different approach.” Now we are down a rabbit hole chasing a random concept instead of solving the actual issue.
This is the most frustrating part for any expert sitting in that room. Because the moment you walk into a conversation and clearly state your problem, the remaining 99% of that call gets dedicated to building a solution around your specific situation. A bespoke approach that touches every pain point versus a standard, off-the-shelf system that not only are you implementing blindly, but the people you are trying to impact with it do not even understand it. The very people the solution is supposed to help are crying for support, and the tool that was chosen to help them was never built with their actual problems in mind.
Now, to be fair, some organizations are asking the right questions. They will say, “We are using this tool and we are not sure if it is enough or if we need to go deeper.” That is a great starting point. But even the right questions cannot lead to the right solution if the core problem was never clearly identified. Without that, there is no foundation to build on.
And here is the hard truth from the expert’s side of the table: when you sit in a room and dissect an organization, you can often see exactly where the problem is. But you will lose 100 out of 100 times if you walk in and simply tell people why their diaper stinks. People do not buy solutions that are pushed on them. They buy solutions they have owned. The breakthrough happens when the organization is willing to say, “Here is where we have soiled ourselves, and here is what we are struggling with.” That is when real progress starts.
3. Studying Something for Years Does Not Mean You Have Solved It
Here is something I see more often than I should. A leader acknowledges a gap in their AI expertise. Good. That is self-awareness. But underneath that leader are team members who have been researching AI for years, sometimes the better part of a decade, and still have not built a single practical solution for the organization they serve.
At some point, you have to ask yourself: if you have been studying something this long and still have not produced an answer, is the issue really a lack of information? Or is it that you have been listening for what you want to hear instead of following the guidance of experts who have already built and delivered these solutions?
Some people have researched everything except how to identify their actual problem and seal their lips long enough to hear the solution. That is not preparation. That is avoidance dressed up as diligence.
It is natural to ask, “What have you done for others in this space?” But every organization has different problems. And when someone does not even know their own problem and they are asking to be sold a solution, that is not a knowledge gap. That is a self-awareness gap. You cannot fix what you do not know is broken.
While we can identify problems inside an organization, real buy-in comes from the person living in the problem being willing to name it.
4. Leaders Do Not Wait for a Playbook. They Create One.
You cannot want to lead the charge in something without having the courage to create the solution to the problem. That is what leadership is. Leaders do something that has not already been done. Whether that is performing the same task better than everyone else or building something entirely new, leadership requires action, not observation.
It is easy to say in theory, “If we do not understand and adopt AI, we fall behind.” That is true. But that statement alone does not move anything forward. True custom solutions are created around closing the gaps on the specific problems a company faces, not around buzzwords or fear of being left behind.
5. Get Comfortable Being Uncomfortable
I recently spent time with friends who have a one-year-old. This kid is hilarious. Every time he fills his diaper, usually right after dinner, he just stands there smiling. Completely content in the mess. When they grab him to change the diaper, he is upset. He is comfortable in that state of filth and does not want it disrupted.
I think about that when I talk to businesses.
Some organizations have been sitting in the same soiled problems for so long that the discomfort has become familiar. They will not name the problem because deep down they understand what comes next: a solution that requires change, and change is uncomfortable. So they stay in it. They smile. They resist the cleanup.
But to lead a charge, you have to be willing to have the hard conversations. You have to be willing to disrupt the norm and the comfort. You have to be willing to say out loud what the problem is, what that problem is truly costing you, and then have the courage to act on the solution.
6. Know the Difference Between Checking a Box and Solving the Problem
There is a major difference between grabbing an out-of-the-box tool and investing in a custom-tailored solution that addresses leadership support, talent development, operational efficiency, and long-term scalability all at once. One makes you feel like you did something. The other actually changes the trajectory of your organization.
Too many companies choose what is easy now and end up paying for it later, either in rework, in lost time, or in the people they were trying to help still struggling because the solution was never built for them in the first place.
The Bottom Line
Wanting to lead and being equipped to lead are two very different things. The companies that actually get ahead do not just have ambition. They have clarity on the problem, they know the cost of leaving it unsolved, they have the discipline to listen, and they have the courage to choose the right solution even when it demands more of them.
The data backs it up. 95% of AI pilots fail. 84% of those failures are leadership failures. And companies that partner with specialized vendors succeed at double the rate of those who try to build it alone.
If your organization is navigating a major shift right now, start with these six things. Everything else gets easier from there.
Frequently Asked Questions
Why do most AI implementations fail?
According to MIT’s 2025 research, 95% of AI pilot programs fail to deliver measurable impact on a company’s P&L. The RAND Corporation found that 84% of those failures are leadership-driven, not technical. The most common reasons include an inability to clearly define the problem being solved, analysis paralysis, and choosing off-the-shelf tools that do not address the organization’s specific challenges.
Should my company build AI internally or partner with a vendor?
MIT’s research shows that companies partnering with specialized AI vendors succeed approximately 67% of the time, while internal builds succeed only about 33% of the time. The key advantage of working with a specialized partner is their ability to build bespoke solutions around your specific problems rather than forcing your organization into a generic framework.
How do I know if my company is ready for AI?
Readiness starts with one question: can you clearly articulate the specific problem you need AI to solve and what that problem is costing you? If you can, you are ready to have a productive conversation with a solutions expert. If you cannot, start there. The technology is not the barrier. Clarity is.
What is the difference between off-the-shelf AI and a custom AI solution?
An off-the-shelf AI tool is a pre-built product designed for general use cases. A custom AI solution is built around your organization’s specific problems, workflows, leadership needs, and growth objectives. The difference is the gap between checking a box and actually transforming how your business operates. Off-the-shelf solutions often fail because the people they are designed to help do not understand them, and they were never built with those people’s actual problems in mind.
Sources
1. MIT NANDA, The GenAI Divide: State of AI in Business 2025. Based on 150 interviews, 350 employee surveys, and analysis of 300 public AI deployments.
2. RAND Corporation, AI Project Failure Analysis, 2024. Research on enterprise AI implementation failure rates and root causes.
3. S&P Global Market Intelligence, 2025 Enterprise AI Survey. Survey of 1,000+ enterprises across North America and Europe.
