McKinsey Leader: When CEOs Hand AI to Analytics Teams, the Strategy Is Already Lost
Most companies are spending more on AI and getting less return. Only 39% of organizations report earnings impact from their AI investments, according to McKinsey research. The problem isn't the technology - it's who owns the decision.
Asutosh Padhi, senior partner and global leader of firm strategy at McKinsey, has a simple test for whether a company's AI strategy will succeed: watch who answers when you ask the CEO how it's going. If the CEO turns to the chief analytics officer, "you know that it's game over," he said.
The gap between AI spending and business results reflects a broader leadership failure. About 90% of enterprise AI efforts have not produced measurable business value, according to recent studies Padhi cited. CFOs and CEOs consistently tell him the same thing: IT spending keeps rising, but returns remain unclear.
The Fragmentation Problem
Most companies operate with siloed data across multiple ERP systems, past acquisitions, and fragmented sources. This infrastructure problem compounds the leadership problem. Teams run 40 or 50 isolated pilots that rarely scale or connect to the broader business.
The conventional wisdom - start with easy wins - backfires. Simple use cases happen "on the side that no one is really paying attention to," Padhi said. Even when they succeed, they don't shift the organization.
Start With the Hardest Problem
The counterintuitive approach: tackle your toughest business problem first. When you choose something that will move enterprise value, everyone pays attention. The organizational commitment and change management resources follow naturally.
Once that works, scaling becomes possible. The goal is what McKinsey calls an AI management operating system - an architecture running from the CEO to the front line that embeds AI into redesigned workflows and accelerates decision-making.
Companies that build this system can compress product introduction cycles by 70% or more, Padhi said.
What Leaders Need Now
The bar for executive leadership has risen. Three qualities now matter most: technological fluency, speed, and human judgment.
Technology fluency. "If you don't learn technology, and if you think that you're going to outsource that, it's game over," Padhi said. Leaders who delegate AI strategy to technical staff create distance between themselves and the core competitive advantage.
Speed. Waiting two years to evaluate AI means competitors build compounding advantages in the meantime. The cost of delay compounds faster than the cost of a misstep.
Human judgment. As information becomes more available, what you do with it becomes harder. Empathy, kindness, and judgment will command a premium because technology alone won't answer the questions that matter.
For executives focused on AI strategy, AI for Executives & Strategy resources address these ownership and implementation challenges directly. The AI Learning Path for CEOs covers the strategic decisions that determine whether AI investment produces measurable returns.
Your membership also unlocks:
