AI Transformation: Strategy vs. Operations

I’ve been studying and discussing AI transformation lately, and I keep seeing two very different conversations.

The first conversation is strategic. It’s about how AI is reshaping competitive dynamics across entire industries: creating new coordination mechanisms, shifting control points, enabling companies to reshape the playing field itself. This is the conversation you see in strategy conferences, HBR articles, and boardrooms when consultants present their frameworks.

The second conversation is operational. It’s about how to build AI systems that work sustainably, where to draw delegation boundaries, how to structure accountability when things go wrong, and why the pilot that worked beautifully six months ago is now quietly degrading in production. This is the conversation you hear when you study the experiences of the people implementing AI.

Both conversations matter. But they’re happening in different rooms, using different vocabularies, and rarely connecting.

The strategic conversation and the operational conversation are still happening in different rooms.

I’ve been thinking about this disconnect while reading Sangeet Paul Choudary’s Reshuffle, which is the most sophisticated articulation of the strategic conversation I’ve encountered. Choudary concludes with a provocative claim: “You don’t need an AI strategy.” What you need instead, he argues, is a business strategy that leverages AI to reshape your competitive landscape. Stop asking “what can AI automate?” and start asking “how is AI changing the rules of competition?”

It’s a powerful reframe. Choudary shows convincingly that the biggest opportunity isn’t optimizing existing workflows—it’s recognizing how AI changes coordination mechanisms and control points across entire industries. His examples are compelling: TikTok solving the cold start problem through behavior graphs rather than social graphs, Shein coordinating fast-fashion production through algorithmic supply chain orchestration that replaces human management with real-time demand signals, and platforms like Uber Freight restructuring trucking by algorithmically controlling pricing, routing, and load assignments. These fundamentally reshape how industries coordinate rather than just making existing coordination faster.

Choudary’s ultimate strategic goal is what he calls “field reshaping”—not just competing better or adapting faster but dominating so completely that you force changes to the competitive landscape itself. His analogy: Tiger Woods won at golf so decisively that tournament organizers redesigned courses to contain his advantage. That’s the aspiration: reshape the competitive landscape so thoroughly that others must adapt to the rules you’ve established.

The book provides an essential lens for understanding where industries are heading. Every executive should engage with Choudary’s framework as a thought experiment: How might AI reshape the competitive dynamics of our industry? Where are the emerging coordination gaps and control points? Who will capture them? And can we position ourselves to be field reshapers rather than just anticipators?

But there’s a gap between understanding that strategic vision and having the confidence to execute it.

The AI Fluency Gap

Most CEOs and executive teams, particularly outside the technology sector, face a novel situation. They’re being asked to make transformational bets on technology that executives are still exploring and seeking to understand. Not even the engineers building the new foundation AI models claim to fully understand them yet. The “field reshaper” strategy requires intuition about AI’s capabilities, failure modes, and integration challenges—intuition that only comes from accumulated operational experience.

Betting the company on transforming your industry’s coordination layer while your organization is still building basic fluency with AI deployment creates legitimate governance challenges. The confidence to reshape the playing field comes from expertise that the industry is still developing.

This creates a dilemma. Choudary frames a spectrum from “reactive optimizer” (using AI to do old tasks faster) to “field reshaper” (using AI to change the rules of the game). The implicit message: reactive optimization is a trap. Don’t get stuck there.

But what if you’re starting there deliberately?

Perhaps you have a clear vision for reshaping your industry with AI but need to build up capabilities and resources first. Perhaps your vision is compelling but your board isn’t yet aligned. Perhaps you don’t yet have a clear vision, but you feel compelled to start building experience and capability so you’ll be ready when the fog clears. What’s the path forward?

This is the messy real world with constraints and limits that high-level strategic frameworks wave away. In the real world, we have to deal with the situation we’re in, not the one business school professors find novel and interesting.

The Strategy-Execution Gap

Influential business strategy frameworks tend to show destination, not journey. Choudary’s Reshuffle and Clayton Christensen’s disruptive innovation theory both identify what’s happening at the system level and provide frameworks for clear thinking about strategic positioning. But by focusing on end-state competitive positioning, they tend to skip over the steps of how you get there.

This isn’t new for Choudary—Reshuffle essentially applies the coordination economics framework from his earlier work (Platform Revolution, Platform Scale) to the AI era. Where those books examined how platforms orchestrate ecosystems through coordination mechanisms and network effects, Reshuffle positions AI as the new orchestration layer that enables “coordination without consensus” at unprecedented scale.

What Christensen’s later work somewhat reluctantly admitted was that awareness of the trap and leveraging current operational advantages in new directions can create escape routes. The incumbent that understands disruption theory and systematically builds resources through the existing business can engineer the degrees of freedom needed to pursue disruptive opportunities.

Choudary’s framework has similar characteristics. His “field reshaper” vision correctly identifies where the highest value lies. But executing that vision requires organizational capabilities that take time to develop: AI fluency built through operational experience, operating leverage to fund transformation without immediate ROI pressure, competitive breathing room for multi-year initiatives.

The practical question: If you understand where the field is being reshaped but need to build the capabilities to reshape it yourself, what’s the path forward?

Targeted Refactoring: The Path to Fluency

Reframe what Choudary calls “reactive optimization” as “targeted refactoring” when pursued with clear strategic intent.

Reactive optimization: “Let’s use AI to speed up claims review because that’s what we see other firms doing.”

Targeted refactoring: “Let’s systematically rebuild our claims review workflow with AI to learn how delegation boundaries work, where human judgment must remain in the loop, what our data infrastructure needs to support, and how to build organizational fluency with AI as a cognitive instrument not a magical black box.”

Targeted refactoring builds the experience and fluency necessary for more ambitious transformation. You learn where AI reliably operates and where it fails unpredictably. You develop expertise about delegation boundaries, accountability structures, and data quality requirements. This operational fluency creates option value—you’re building the experience Choudary’s field-reshaping strategy assumes you already have.

It also generates operating leverage to fund further transformation. Real efficiency gains from refactored workflows free up capital and organizational bandwidth. Choudary’s ambitious ecosystem strategies require resources. Targeted refactoring creates them.

Use the existing business to build the resources and credibility that create strategic degrees of freedom for more ambitious transformation.

The Walmart Pattern: Refactoring to Reshaping

The pre-AI era offers a perfect case study: Walmart’s response to Amazon. Choudary himself uses Walmart as an example of field reshaping in Reshuffle, focusing on how they leveraged barcode data in the 1970s-90s to restructure retail power dynamics and supplier relationships. Walmart’s response to Amazon follows the same pattern, showing not just that they’re field reshapers, but how they systematically build the capability to reshape fields when disrupted.

In the mid-2010s, Walmart faced an existential threat. Amazon’s asset-light, everything-store e-commerce model was eating their lunch. Walmart had 4,700+ expensive physical stores optimized for a business model that looked increasingly obsolete. Wall Street analysts were writing their obituary.

Walmart refused to accept that disruption was inevitable. They spent billions acquiring e-commerce capability through Jet.com and other acquisitions, built out fulfillment infrastructure, and learned how to integrate digital and physical operations. This was deliberate fluency-building through systematic practice.

Then came the insight: those 4,700+ physical stores weren’t a legacy burden. Refactored correctly, they became a competitive advantage. Walmart created BOPIS (Buy Online, Pick-up In-Store), used stores as local hubs for same-day delivery and grocery fulfillment, and solved the “last mile” problem that plagued Amazon. Store locations became control points.

Once Walmart built technical fluency and operating leverage through refactoring, they pursued exactly the kind of field-reshaping strategy Choudary advocates. They added pharmacies, health clinics, vision centers, financial services, insurance products—transforming from “competing with Amazon on product selection” to “becoming a coordination layer for customers’ life services.”

Walmart didn’t just survive Amazon’s disruption. They used their refactored physical footprint as a control point to change the game entirely. They’re no longer just a retailer. They’re positioning themselves as the high-touch, local orchestrator of essential services that digital-only competitors can’t easily replicate.

Targeted refactoring that builds technical fluency and generates operating leverage eventually enables field-reshaping strategies that were previously impossible. The same pattern applies to AI transformation, just at compressed timescales.

Where the Frameworks Converge

Choudary’s strategic lens and the targeted refactoring approach are genuinely complementary. His framework shows where AI is reshaping competitive systems: the coordination mechanisms changing, the control points emerging. Targeted refactoring addresses how you build the organizational capability to act on those insights. Where to draw delegation boundaries. How to structure sustainable collaboration. Which patterns of human-AI integration succeed versus fail.

For most organizations, the path forward requires both:

1. Understand the strategic landscape through Choudary’s framework. Where is your industry heading and how might AI change the game? What coordination gaps are emerging? Who might use AI to capture strategic control points? Even if you’re not positioned to reshape the field now, you should iterate on a concept of how the reshaping might unfold.

2. Build fluency through targeted refactoring using systematic delegation principles. Start with workflows that exceed human capability thresholds—scale (processing volume), accuracy (precision requirements), endurance (sustained attention), and complexity (pattern dimensions). Learn how to structure human-AI collaboration. Develop expertise about where AI reliably operates versus where it fails unpredictably.

3. Generate operating leverage from successful refactoring. Use the efficiency gains and freed capacity to fund more ambitious initiatives and expand your degrees of freedom.

4. Expand toward strategic transformation as your fluency and resources grow. The “field reshaper” strategy becomes viable when you’ve built the organizational capability to execute it confidently.

The Strategic Middle Path

Choudary’s framework naturally emphasizes the end-state positioning: the “field reshaper” as the ultimate strategic goal. This clarity about destination is valuable. But it can inadvertently suggest that intermediate positions are transitional states to move through quickly rather than legitimate strategic positions to occupy deliberately.

Most organizations will spend extended time in targeted refactoring. This isn’t compromise—it’s prudent governance. Fiduciary responsibility demands near-term results while investing for the long term. Governance requires demonstrating value at multiple milestones, not just at the end of a multi-year transformation. Coordination-layer opportunities may require coalition-building that takes time.

The field-reshaping opportunity is real. But for most organizations, the path there runs through targeted refactoring that builds AI fluency, generates leverage, and creates the option value necessary to pursue more ambitious transformation.

Choudary shows you the Tiger Woods vision: how AI can reshape entire competitive landscapes. But you don’t become Tiger Woods by watching a documentary about Tiger Woods. You become Tiger Woods by hitting thousands of balls on the range, logging countless hours on the course, building the muscle memory and pattern recognition that eventually creates dominance.

Targeted refactoring is the driving range.

---

This analysis builds on The Great Refactor and Refactoring Agents, which examine how AI is restructuring work around systematic delegation thresholds and sustainable human-AI collaboration patterns.