Executives across industries are pouring unprecedented capital into data platforms, analytics and artificial intelligence. The promise is compelling. Better insights. Faster decisions. Measurable growth. However, the results are often familiar and frustrating. Major AI programs perform poorly. Productivity growth has stalled. The quality of decisions has improved on paper but not in practice.
The problem is rarely the technology itself. More commonly, it is the system into which the technology was introduced.
AI cannot fix execution gaps. It amplifies them. When culture, decision-making authority, and day-to-day work processes are not aligned, advanced technology exposes previously hidden or manageable weaknesses. In many organizations, the faster insights arrive, the more clearly the organization’s limitations are revealed.
Most operating models still reflect earlier eras. Information spreads slowly. Power is centralized. Decisions are often escalated by default. These structures once provided stability. Today, they quietly undermine speed and accountability.
AI thrives on clarity. It requires timely decisions, clear ownership, and trust in the data. When these conditions do not exist, performance deteriorates rapidly.
The price of stagnation
The operating model determines how work is done. It controls who makes decisions, how information flows, how teams coordinate, and how success is measured. While strategies evolve and technology advances, operating models tend to change minimally. Over time, layers accumulate. Abnormalities multiply. Responsibilities are blurred.
The friction was subtle at first. Then they got back together.
AI tools can present insights in real time, but decision-making authority remains vague. The analysis emphasizes opportunity, but incentives still reward risk aversion. Pay lip service to collaboration, while processes reinforce functional silos. Rather than accelerating execution, technology increases pressure.
In these environments, AI becomes a stress test. It doesn’t create dysfunction, but it makes existing dysfunction more visible. When trust is weak, data can be questioned. Without clear responsibilities, insights can become stagnant. If leaders hesitate to transfer authority, decision-making will hit a wall.
Why execution failed
Execution failure is rarely due to lack of ambition or investment. This occurs because the operating model was never designed to support the behaviors required for sustained performance.
Three faults occurred repeatedly.
The first involves decision-making authority. Artificial intelligence enables faster, more distributed decision-making. However, many organizations still rely on centralized approvals. Insights spread faster than leaders can process them, creating delays that negate the value of speed.
The second breakdown is procedural. New tools are layered onto legacy workflows. Employees adapt by working around the system rather than through the system. Complexity increases. Friction becomes normalized.
The third breakdown is cultural. Data challenges intuition. Automation disrupts established roles. Without norms that support learning, accountability, and adaptation, insights are viewed as advisory rather than actionable.
Under stable conditions, these gaps can exist. Under pressure from advanced analytics and automation, they have become structural liabilities.
Growth is structural, not technological
Sustained growth doesn’t just come from technology. It comes from alignment. Structure, behavior and responsibility must complement each other.
Organizations that derive real value from AI approach the challenge differently. They don’t just focus on tools. They examine how decisions are made and where decisions stall. They clarify ownership of the results. They redesign workflows so insights translate directly into action. As procedures change, cultural expectations are reinforced.
This is not about replacing judgment with algorithms. It is about ensuring that judgments are made at the right level, at the right time, with the right information.
When operating models are aligned, AI can improve focus and accelerate learning. If this is not the case, AI adds noise and amplifies risk.
strategic blind spots
Operating models are often viewed as internal mechanisms. Strategy and technology take precedence. If anything, the structure will be adjusted later. This sequence is costly.
The operating model determines which strategies can be executed and which technologies can actually be delivered. They are not passive infrastructure. They positively impact performance.
In an environment where advantage depends on speed and follow-up, the question is no longer whether to invest in AI. The more relevant question is whether the organization aims to act on the information revealed by AI.
For many businesses, the answer is disturbing.
Rethink how work gets done
Revisiting the operating model does not require dismantling the organization. It requires facing reality. Where does decision-making slow down? Where the responsibility disappears. Where incentives conflict with stated priorities.
This means examining decision bottlenecks rather than reporting relationships. This means tying rewards to outcomes rather than activities. This means designing workflows around value creation rather than functional convenience. It also means addressing cultural norms that quietly undermine ownership.
Technology will continue to advance. Artificial intelligence will become faster, easier to use and more deeply integrated into daily work. Organizations that don’t change their operating model will move faster without moving forward.
Those who put more effort into coordination will experience things differently. AI won’t feel like a gamble. It feels like leverage.
Not because technology has changed, but because organizations have changed.
The views expressed in Fortune opinion pieces are solely those of the author and do not necessarily reflect the views and beliefs of: wealth.