Developing a Successful Generative AI Strategy

Welcome to Leading Disruption, a weekly letter about disruptive leadership in a transforming world. Every week, we’ll discover how the best leaders set strategy, build culture, and manage uncertainty, to drive disruptive, transformative growth. For more insights like these, join my private email list.

Let me set a scene for you: You’re working in an organization that has decided to take a pass on generative AI. It’s too early. It’s too expensive. Your workforce isn’t ready. Your customer base isn’t asking for it. The technology isn’t developed enough. In fact, it’s changing so fast and so often, that it feels impossible to plan around it even if you wanted to.

So, you ignore generative AI and carry on with your company’s strategy … until you realize you missed your window. You lost your competitive advantage. With no plan in place, you are desperately trying to catch up in an industry that is rapidly passing you by.

Now, let’s imagine another scenario, in which you also are hesitant to adopt generative AI tools into your organization. In this case, however, you still strategize with artificial intelligence in mind. You continue to ideate with your teams about how these tools might come into play and how they might impact your business. You consider what circumstances have to change for your company to be ready to dive in, and you prepare for those opportunities. When they eventually come (note that I didn’t say “if”), your organization will be ready.

Which scenario is destined to be more successful? I think we all know the answer.

It is far riskier to have no plan—and no strategy—than having a plan that intentionally sets out when and how you will and won’t use generative AI.??

The Three Steps to Building a Generative AI Strategy

First, it’s important to have a clear understanding of strategy. A strategy is what you will do or won’t do in order to achieve your organization’s goals and objectives. Your strategy comprises a set of deliberate choices based on what you assume to be true about your market. A generative AI strategy is simply how you will employ generative AI to achieve those overall business objectives.

1. Understand the Use Cases

In considering a generative AI strategy, you need to understand what new capabilities generative AI offers. What can you do with this technology to give your organization the competitive edge—to earn more revenue or to operate more efficiently, for instance? Here’s one way of looking at potential use cases:

These “use cases” can offer creative content, they can provide personalization or automation, they can improve research and discovery, or they can assist in education. Once you understand the different ways artificial intelligence can be used, you can begin to understand its potential impact—not just across your organization but across the entire industry. These use cases may guide new expectations from your customers and partners as well as new capabilities from your competitors.

It’s worth repeating that even if you decide not to move forward with generative AI for many sensible reasons right now, it’s crucial to be knowledgeable about these use cases and how they could potentially be used in your organization. If you don’t know what this emerging technology can do, you can’t even begin to understand how it could impact your strategy.

2. Assess Your Current Strategy

One of the biggest mistakes leaders make when trying to create a generative AI strategy is “use case bonanza,” when you try to develop all of them at the same time. Instead of trying to plan for every single potential use case within your organization, you need to first assess your current strategy.?

When you developed it, you made choices based on what you assumed would be true—about how the market would develop, your customers' needs, and your product offering.

Ask yourself: What did you assume to be true? What would have to be true for your strategy to be successful? Then ask yourself: Now that you know about the capabilities of generative AI, do those assumptions still hold true?

Can you still create great sales pitches at scale? Are your customers getting the level of personalization they have now come to expect? Can you automate some of your HR or legal processes to get a leg up on your competition??

I can't emphasize this enough: You must be able to look at a generative AI strategy in the context of your existing business strategy. If you try to develop these initiatives in a vacuum, without a sense of how your overall strategy is going to be affected, you're going to be spinning in circles.?

Moreover, you need to explain to the top executives of your organization why this matters, especially if they are your peers. I’ve gone through countless digital transformation and disruptive innovation strategies with C-suite executives and their boards, and the truth is, they only care about the few objectives that will impact the business. If you can't put generative AI in the context of those top strategic goals, they won't care. If you can’t show how generative AI is going to support or challenge those goals, they won’t be interested.?

3. Prioritize Your Generative AI Initiatives

This final step may seem the most intimidating, but it’s truly all about balancing demand and risk.?

Look at the demand for a particular generative AI initiative against the risk—it could be the regulatory risk, the reputational risk, or even just the risk that you're giving wrong advice. Here’s an example: there’s a lot of demand for the low-hanging fruit of generative content, like marketing and copyediting. Conveniently, the risks associated with it are relatively low.?

Conversely, although there’s high demand for medical diagnoses and legal advice, the risks associated with those outcomes are equally high. Consumers would much rather work with a trained pathologist or lawyer than a chatbot. In that case, you may come up with ways to provide a highly demanded service with lower risk, like by offering summaries of legal documents.?

Once you can grasp the balancing of demand and risk, you can lay out your investment timeline. What’s worth doing now versus what can you save to tackle later? A great example might be your organization’s AI tech stack. It’s currently a profoundly involved process to build one out, but it looks to be a lot easier to tackle in a few months. Perhaps it’s more worthwhile to wait to explore this until then.

When you are ready to develop a plan, I believe it’s wise to create a 90-day plan as well as a long-term, 18-month roadmap. The 90-day plan identifies some immediate, easy wins and experiments to try,and it provides a quick turnaround to decide what you need to do as soon as possible. It helps inform you of policies, trainings, and guidelines that need to be in place up front so that you can do the deep thinking necessary to produce a strategic roadmap based on your use case determinations and your assessment of your organization’s current strategy.?

The insights gained during those 90 days can then be used to develop that long-term plan, which I call the “six-quarter walk.” That plan lays out in detail how your generative AI strategy will unfold to support your business strategy. The beauty of this approach is that it’s centered on the “what needs to be true” question and allows for agile quarterly updates as part of a rolling 18-month plan.

Of course, I get it. It can feel as though it’s impossible to create a sound generative AI strategy when the technology is constantly changing. It’s not the right time. The risk is too high. The demand isn’t there. In following these three steps—and committing to that 18-month plan that offers plenty of opportunities for revisions—you’ll likely discover that your strategy is less dependent on these external factors after all. It’s truly the most important thing you and your organization can do.

Your Turn

How are you and your organization preparing a strategy around generative AI? How are you prioritizing what initiatives come first, second, third … or not at all?