The Organizational Reckoning: Why AI Demands a New Kind of Company

By José Manuel Abrams - MBA

Chief Data Culturist - www.dataculturehivemind.com

The AI Thinker

The Sears, Roebuck and Company Moment — Why Incumbents Miss the Shift

Sears, Roebuck and Company, also known as Sears, had everything. A national distribution network, deep customer loyalty, a catalog that reached millions of homes, and decades of retail dominance. They had the resources, the reach, and the relationships to adapt to a changing world. And yet, Sears, found itself increasingly irrelevant as a competitor built an entirely new model around it.

What happened? It wasn’t a lack of money or ambition. It was a structural problem. Sears was organized around assumptions that no longer held. Their hierarchies, their decision-making methods, their team structures — all of it was optimized for a world that was quietly disappearing. When the shift came, they tried to adjust by adding new capabilities onto old foundations. They bolted e-commerce onto retail operations. They hired digital marketers without changing how decisions got made. They invested in technology without rethinking the organizational logic underneath it.

Amazon, meanwhile, wasn’t adapting. It was building. From day one, technology wasn’t a department — it was the operating system of the entire company. Every function, every role, every decision was designed around data flows, systems thinking, and digital infrastructure. Amazon didn’t try to become a better version of the old retail model. It reimagined what retail could look like when technology was baked into its DNA from the very beginning. The result wasn’t just a more efficient version of what existed. It was something categorically different.

We are at that same inflection point today — except the technology is AI, and the stakes are even higher. Companies that treat AI as a tool to plug into existing structures will find themselves in the same position as Sears did facing Amazon. The window to restructure is open, but it won’t stay open long.

Why AI Is Different This Time

Each wave of technology has prompted predictions of organizational transformation, and every time, most companies have managed to absorb it without truly changing. They added IT departments. They hired data analysts. They built digital teams. The org chart shifted at the edges, but the core logic — business on one side, technology on the other — remained intact.

AI breaks that logic entirely. Previous technologies were tools that humans used to do their jobs. AI is increasingly capable of doing parts of those jobs itself — and more importantly, it does so in ways that are deeply dependent on data quality, system architecture, and information governance.

A leader who doesn’t understand how data moves through an organization, where it comes from, and how reliable it is, cannot successfully lead in an AI-powered environment. They will make decisions based on AI outputs without the foundation to evaluate them. They will approve proposals without seeing the implementation risks. They will be surprised by failures that someone with a data background would have anticipated.

This is not a technology problem. It is an organizational design problem.

For twenty-five years, I’ve watched companies build walls between their business units and their data and technology teams. Business leaders made a strategy. Data and IT teams executed. The assumption was that you didn’t need to understand the plumbing to make good decisions — you just needed the outputs. AI has made that assumption dangerous. Because now the plumbing is the strategy.

Amazon understood this from the start. Data wasn’t a support function — it was central to every decision, every product, every customer interaction. That organizational philosophy enabled them to move faster, adapt more quickly, and outpace incumbents who still treated technology as a back-office function.

The leaders who will thrive in this environment are those who can speak both languages fluently — who understand business objectives and the data architecture required to achieve them. They are rare today. Organizations need to deliberately start building them.

Restructuring for Survival — What Companies Actually Need to Do

So what does restructuring actually look like? It’s not about hiring a Chief AI Officer and calling it done. It’s not about sending executives to a two-day AI workshop. It’s about rethinking the fundamental organizational logic that governs how decisions get made, how teams are composed, and what skills are valued at every level of the company.

Here are the shifts that matter most:

Dissolve the wall between business and data. The model in which data teams serve business units as internal vendors is no longer sufficient. Business leaders need sufficient fluency in data and systems to meaningfully participate in decisions about AI implementation. Data professionals need enough industry context to design solutions that actually solve the right problems. This requires deliberate rotation, cross-functional team design, and recruitment standards that reward hybrid skill sets.

Rethink what leadership competence means. For decades, the path to executive leadership ran through domain expertise — the best salesperson became sales VP, the best marketer became CMO. AI requires a new definition of leadership competence, one that includes the ability to evaluate AI-generated insights, understand data governance implications, and anticipate how system design affects organizational outcomes. This needs to be reflected in how companies develop and promote talent.

Build for adaptability, not efficiency. The organizational structures of the twentieth century were optimized for efficiency — clear roles, defined processes, minimal redundancy. AI-era organizations need to optimize for adaptability. That means flatter structures, faster decision cycles, and teams that can reorganize around problems rather than being locked into fixed functions.

Invest in data foundations before AI applications. One of the most common mistakes I see is companies rushing to deploy AI tools on top of poor data infrastructure. The AI amplifies whatever lies beneath it — including the gaps, inconsistencies, and governance failures. Before asking what AI can do for your organization, ask whether your data is ready to support it.

In Closing

The companies that will define the next decade are not necessarily the ones with the most advanced AI tools. They are the ones that restructure themselves to think, decide, and operate in fundamentally new ways. That requires leaders who understand both the business and the technology. It requires dissolving old boundaries and building new ones in better places. And it requires the willingness to stop bolting new capabilities onto old structures and start imagining what the organization needs to look like from the ground up.

Sears had the catalog, the customers, and the reach. They just couldn’t reimagine what came next. Amazon didn’t have those advantages — but it had the organizational vision to build for a new world rather than defend the old one. The question every organization needs to answer right now is: are we the next Sears, or are we building our Amazon?

Jose Manuel Abrams has spent twenty-five years working in data management and organizational transformation. This blog reflects their perspective on what AI means for the future of organizational design. 🔗 Visit the blog here https://www.dataculturehivemind.com

Disclaimer: The opinions expressed on this blog are solely those of the author and do not reflect the views, positions, or opinions of their employer.