Becoming an AI Augmented Enterprise

How to Leverage AI Strategy

Previously in this series, Ahmad Zaidi, co-founder and CEO of AI start-up TransforML, and Gui Loureiro, Regional CEO Walmart Canada, Central America, Chile and Mexico and co-author of Reinventing the Leader, and I have discussed artificial intelligence (AI) and entry level hiring as well as AI and its augmentation superpowers for individuals. Readers seemed to find these AI topics helpful. So, in this Playing to Win/Practitioner Insights piece, we are broadening our scope to the enterprise with Becoming an AI Augmented Enterprise: How to Leverage AI Strategy. As always, you can find all the previous PTW/PI here.

AI Enterprises vs. AI-Augmented Enterprises

Of course, there are AI enterprises – that is, companies who product is AI, such as Anthropic, Open AI or Mistral. Others, such as Apple, Google, or Microsoft, sell AI as a product line. Still others, such as Nvidia or Databricks, supply inputs to AI enterprises.

But that combined universe of companies is small compared to the universe of companies that are already or will soon be using AI to enhance their business. In this respect, what we might call AI Strategy is analogous to Functional Strategy, which has been discussed previously in this series. The task of every functional strategy is to help the overall enterprise win bigger and better with its overall strategy, including opening up ways to win that wouldn’t have been possible without functional strategy innovation.

For example, Four Seasons’ unique human resources functional strategy underpins the company’s legendary and differentiated guest service in the luxury hotel space. At the other end of the price spectrum, Costco’s unique human resources functional strategy makes possible its unparalleled mass merchandiser shopper experience. P&G’s sales strategy, featuring large multifunctional teams co-located at its retail partners’ headquarters, helps it achieve the most powerful go-to-market positioning in its industry.

Think about it as fractal. When you look inside of enterprise strategy, functional strategy mirrors it at the next leval. That is, just as the enterprise must, the function needs to make choices across its Strategy Choice Cascade (SCC). They aren’t the same choices but rather a set of choices across the same five boxes that helps the enterprise win bigger and better than otherwise. Similarly, AI can augment the enterprise strategy of any company.

It can enhance the Must-have Capabilities (MHC) and/or Enabling Management Systems (EMS) for the enterprise’s current Where-to-Play/How-to-Win WTP/HTW choice. For example, AI could help its factory run more efficiently or enhance the effectiveness of its recruiting process. But it can also alter the enterprise’s HTW as with Netflix, which not just delivered movies of your choice but helped you choose with AI. It can open up new enterprise WTP opportunities, such as adding AI-driven advice to the product/service. The key is for AI strategy choices to be precise about what enterprise augmentation they seek to enable.

AI Augmented Enterprise Strategy

To maximally augment the enterprise, a company needs to make AI centric choices across the entire SCC, as follows:

Winning Aspiration

We have previously argued that a key question on AI is whether a company makes its primary goal to use AI to substitute for humans or to augment humans. We believe that a generically better aspiration is to strive for augmentation versus substitution. That has been the pattern for winners with previous technological disruptions. Earlier technology waves show that moving first against such a winning aspiration adds significant distance between leaders and laggards. For example, in the computerization wave of the late 80s to early 90s, organizations that invested saw results compound over time, with the long-term productivity impact of those investments being up to five times larger when measured over a 5–7 year horizon. Similarly, the Cloud wave of the past decade was associated with a 2.3-6.9% increase in sales.

We think it is already happening with AI. For example, Netflix’s strategy to use AI to augment customers’ ability to choose movies more suited for them has won big.

Where-to-Play/How-to-Win

Where can AI enhance the enterprise’s ability to deliver its offering in a way that would enhance its ability to win?

For example, with a previous technological innovation – the Internet – Thomson Reuters Westlaw business, which was the dominant provider of legal casebooks in America, realized that the Internet could transform delivery of its service. For a century, Westlaw’s enterprise strategy was to enable litigators to search for precedents by equipping them with books that compiled all legal judgments coming out of US courts, combined with a numbering system that made searching easier. The company employed over 1000 full-time lawyers to write ‘headnote’ summaries of each case and categorize the subject of the case using its proprietary numbering system.

But the client workflow was cumbersome. The litigators would provide instructions to law librarians who would go to the firm’s library, which would be full of Westlaw casebooks, and pull out what they thought were the most relevant cases and photocopy those cases to provide to the litigators. Sometimes it would be exactly what the litigator needed, but oftentimes there was rework required.

With the coming of the Internet and highspeed access in law firms, Westlaw realized that it could put its entire library online, enabling litigators to search on their own from their desktop, allowing for greater speed and precision. The high-cost real estate required for law libraries disappeared as did the jobs of some law librarians – the inevitable substitution. But Westlaw still needed all the summarizing lawyers, plus lots of new specialists to manage a giant server farm underground in Minnesota and its precious database. But the key to its spectacular success was that it augmented the capabilities of the litigators that it served by building its own capability in on-line knowledge distribution.

At the AI strategy choice level, the company needs to determine in what areas it can invest in AI initiatives that will provide value for the enterprise in augmenting the WTP/HTW and/or MHC and EMS of the enterprise. For example, Amazon AI suggests books and Netflix AI movies, enhancing the user experience for the customers of each, delivered as a critical element of enterprise strategy. That is the WTP choice of the AI strategy – to enable that enterprise HTW. However, that AI WTP choice needs to have an AI HTW attached to it. If any competitor can do it and do it as well, it isn’t a powerful augmenter of the enterprise strategy.

Thinking back to Westlaw, its Internet WTP leveraged its existing enterprise HTW – which involved having the best (physical) database of case law and a proprietary numbering system that was taught at every law school, plus the greatest scale as the market leader – to make it difficult for competitors to follow with as attractive an online offering. In contrast, selling pet food on the Internet was a WTP without a matching HTW.

John Deere’s application of AI to precision agriculture provides a modern example. Deere has long competed on helping farmers increase productivity per acre, and AI allows it to augment that value proposition. Through its See & Spray system, cameras and machine learning models embedded in Deere sprayers can distinguish crops from weeds in real time and apply herbicide only where needed. For farmers, this reduces chemical usage dramatically while maintaining or improving yields, directly strengthening the core economics of farming operations.

But the power of this AI WTP – See & Spray – comes from Deere’s existing enterprise HTW. The company already operates a vast installed base of connected tractors, planters, and sprayers across millions of acres, generating enormous proprietary datasets about crops, soil conditions, and field performance. By integrating AI directly into its equipment and training models on this proprietary data, Deere makes its precision capabilities difficult for competitors to replicate. In this way, AI augments Deere’s existing strategic advantage rather than creating a standalone capability. This illustrates the broader principle: the most powerful AI investments are those that reinforce the enterprise’s existing WTP and HTW rather than attempting to create advantage from scratch.

Must-Have Capabilities

To deliver on the promise of AI augmentation, a company needs to make certain that its people are augmented with AI. An example of an organization that has already invested in this significantly is JPMorgan Chase. The firm has deployed internal generative-AI tools that allow bankers, analysts, and other professionals to query large volumes of internal documents and data using natural language. In investment banking, for example, AI tools help analysts review financial disclosures, extract key information from documents, and prepare draft materials for client work. Importantly, the AI does not replace the professional judgment of the banker or analyst; instead, it reduces the time spent on manual information gathering and document review.

Other financial institutions are already working on augmenting people with AI in similar ways. But the key to unlocking value isn’t in building AI tools and handing them over to people. Rather, in our experience, the real value unlock comes from complementing them with change management to build capabilities of people to capture value from AI.

Enabling Management Systems

Even when enterprises make the right choices about where AI should augment the enterprise and invest in augmenting their people, many initiatives stall because the surrounding management systems remain unchanged. AI tools alone do not transform an organization; they must be embedded in systems that allow people to rely on them and incorporate them into everyday work.

To make such embedding successful, companies must put in place trust-producing foundations. Given the inherent unpredictability of AI systems, one of the most effective mechanisms is logging both the inputs given to them and the outputs they produce, and exposing these to business teams so they can review results, provide feedback, and help improve performance over time. This creates a rapid learning loop where models and workflows can be iterated and optimized in real operating conditions. Traceability is equally important: when outputs are linked to underlying sources, prompts, or decision logs, users can understand how conclusions were generated and develop confidence in using the system. In essense, it helps train the (AI model) trainers.

But trust alone is not enough. Enterprises must also redesign key management processes to take advantage of AI augmentation. For example, AI can help cascade strategic choices into aligned initiatives, owners, and KPIs across the organization. It can also enable machine-assisted follow-through and exception-based management, replacing status meetings with continuous monitoring that surfaces where performance is diverging and leadership attention is required. In this way, AI strengthens governance and execution by allowing leaders to focus less on information gathering and more on judgment and action.

Practitioner Insights

As every function needs a strategy to choose where to focus its energy and place its investments to create the most value for the enterprise, so does AI.

For existing companies, the natural – and arguably optimal – place to start is with augmenting the enterprise’s existing HTW in its current WTP, as with Netflix and John Deere.

For startups, the possibility exists for AI to enable an entirely new WTP/HTW enterprise strategy. That would be the case for wildly successful ‘robo-investor’ wealth management start-up Wealthsimple. It its case the AI strategy is the enterprise strategy.

There are already and will be many more of both types. But for the former, as strategy progresses, AI can also drive change in the enterprise WTP/HTW.

Back to the Westlaw Internet example, as Westlaw became a more accepted and valued part of the digital/software life of law firms, Westlaw’s WTP began to migrate from the purely legal practice of law firms to helping them with the business of law firms – with software offerings that helped them manage themselves as businesses. Arguably that would have never happened had Westlaw not made the transition from law books to online legal search.

While job #1 for existing companies is to augment their enterprise strategies with AI strategy to fend off the incursion of AI driven start-ups, we look forward to incumbent enterprises seizing the opportunity to raise their Winning Aspirations, alter their Where-to-Play choices, and reinvent their How-to-Wins powered by thoughtful and creative AI strategy.