How to answer “What’s your AI strategy?”

What’s your AI strategy?”

Imagine you’re John, CEO of a successful mid-sized company.

Your calendar is blocked out with meetings, your email is overflowing, and yes, your board is breathing down your neck asking about your “AI strategy.”

You’ve been tap dancing around the topic for months. But let’s face it—the ‘AI question’ isn’t going away, and you’re feeling the pressure.

Sound familiar?

If you’ve ever sat in your office with a sense of dread about how to approach AI in your company, you’re not alone.

Artificial Intelligence, the term that has blown previous hype cycles out of the water, has left many executives grappling with more questions than answers.

  • What is a viable AI strategy?
  • How do you move from buzzwords to tangible plans?

And most importantly,

  • How do you integrate AI without derailing your existing business model?

This article is for you.  In it we’ll describe the steps you need to explore to align AI initiatives with your core competencies, validate your approach through high-impact pilot projects, implement using a lean startup methodology, and scale responsibly for long-term success.

So, whether you’re about to walk into a board meeting, strategise with your R&D head, or simply contemplating the future of your business, let’s start chipping away at the AI question. No more tap dancing, just actionable insights.


Start here: Assess Core Competencies and Identify AI’s Business Benefits

The first step in navigating the murky waters of AI is to clear the fog.

Forget the buzzwords, the hype, and the FOMO (Fear of Missing Out).

AI is not a strategy; it’s a tool.

Much like how the internet became an integral part of business operations, AI is poised to become ubiquitous, but that doesn’t make it your business strategy.

Start by taking stock of what your company excels at—your core competencies.

These are the unique abilities and advantages that have made you competitive in your industry.

It’s crucial to ground your AI initiatives in these competencies. After all, there’s no point in bolting a jet engine onto a bicycle. The same principle applies to AI; it should enhance what you’re already good at, not divert you into a technological wild goose chase.

Involving multiple departments at this stage is critical. Whether it’s operations, marketing, customer service, or R&D, each brings a unique perspective on what makes the business tick.

Collaborative brainstorming sessions can help identify specific challenges or opportunities that AI could address.

This cross-functional approach not only helps pinpoint the most promising applications for AI but also ensures that everyone is a stakeholder in this new journey.

It’s much easier to implement change when there’s collective buy-in.

So, what should you look for?

Identify areas where AI can extend human capabilities rather than replace them.

Identify areas where AI can extend human capabilities rather than replace them.

If your core competency is customer service, think about AI-driven chat support that can handle routine queries, freeing up your human staff to deal with more complex issues that require a personal touch.

If your business thrives on data analytics, consider machine learning algorithms that can help you sift through data more efficiently, drawing insights that would be difficult or time-consuming for a human to extract.

By aligning AI initiatives with what your company already does well, you’re not just checking off a “technology box” to please stakeholders. Instead, you’re enhancing your existing capabilities, improving efficiency, and delivering greater value to customers—all of which leads to a sustainable competitive advantage.

The takeaway is this: Don’t approach AI as a strategy unto itself.

Start by understanding the core of your business and envisioning how AI can amplify these strengths. In the end, AI is not about replacing human intelligence or labour; it’s about augmenting it, making it possible for your business to achieve more than it ever could before.


Step 2: Engage cross-functional teams to identify high-impact, low-risk pilot projects

Once you have a firm grasp on your core competencies and how AI can enhance them, the next step is to transition from theory to practice.

However, many (most?) companies stumble at this point, either paralysed by the options or rushing headlong into a high-stakes project.

The key is to find the middle ground: high-impact, low-risk pilot projects that can validate your approach to integrating AI.

A cross-functional approach is invaluable here as well. In fact, it’s the lifeline of successful AI initiatives.

While your tech team may be enamoured with the intricacies of machine learning algorithms, your customer service team could offer insights into practical applications that will make an immediate difference. The merging of these viewpoints—technical feasibility with business utility—is where the magic happens.

Consider creating a small task force that includes not just tech-savvy personnel but also stakeholders from business units that will be directly impacted by the AI initiative.

Have this team identify a set of potential pilot projects, complete with projected outcomes, resource requirements, and KPIs (further reading: setting goals for your AI committee).

Each proposed project should align with the broader business goals and enhance the core competencies identified earlier.

Start small.

Why start small?  There are multiple benefits.

First, it allows for quick wins that can demonstrate the value of AI to stakeholders.

Second, it provides an opportunity for learning and adjustment without the overwhelming cost of a large-scale failure. If things don’t go as planned, you have an escape hatch: a less successful pilot project is much easier to pivot or shut down than a company-wide initiative.

Now, don’t let the term “pilot” make you think that these projects are inconsequential.

On the contrary, even a small-scale project can offer significant insights into what it will take to scale AI within your organisation, both in terms of technology and personnel.

Plus, it can often serve as a proof of concept that helps you secure more resources for future AI initiatives.

The goal at this stage is to build a bridge between theory and practice, and there’s no better way to do that than with high-impact, low-risk pilot projects.

Done right, these initiatives can offer quick wins, provide valuable lessons, and lay the foundation for scaling AI in a manner that’s both responsible and aligned with your core business objectives.


Step 3: Adopt a Lean Startup methodology for implementation

As you transition from conceptual frameworks and pilot projects to full-scale implementation, it’s essential to borrow from the playbook of startups—yes, even if you’re far from being one.

The lean startup methodology, built around the concept of “Build-Measure-Learn,” offers a robust framework for implementing AI initiatives.

Why lean startup? The AI landscape is dynamic and full of uncertainties.

What may have been a groundbreaking algorithm last year might be obsolete today.

Traditional planning cycles often fail in such fast-paced environments.

The lean approach mitigates this risk by emphasising iterative development and validated learning.

The “Build-Measure-Learn” loop starts with building a Minimum Viable Product (MVP)—in this case, a scaled-down version of your AI solution that solves a core problem.

The MVP doesn’t have to be perfect; it just has to be good enough to test your hypotheses. After the MVP is live, metrics and KPIs come into play. Measuring the impact helps you understand what’s working and what’s not.

Here’s where many organisations stumble: they don’t act on what they learn.

Whether your metrics indicate success or scream for change, adapting your AI project based on these learnings is vital. That could mean pivoting, making incremental changes, or, in some cases, doubling down on your initial strategy if the results are promising.

However, for the loop to be effective, the company culture must value learning and adaptation over rigid planning and execution.

This is especially important in AI projects where results may not be immediately evident.

A culture that penalises failure will invariably stifle innovation, making it virtually impossible to reap the benefits of AI. Instead, embrace failures as learning opportunities that bring you one step closer to a genuinely impactful AI initiative.

And remember, the Build-Measure-Learn approach isn’t a one-time cycle.

It’s a continuous process that helps your organisation adapt to new insights and technological shifts. This iterative process is particularly crucial in the world of AI, where algorithms can always be fine-tuned and the data landscape is ever-evolving.

The lean startup methodology offers a resilient and adaptable framework that aligns well with the uncertainties and potentials of AI. By incorporating this iterative approach, you not only hedge against the risks but also significantly amplify the gains that AI can bring to your core business competencies.


Step 4: Scaling AI initiatives and sustaining long-term growth

Once you’ve validated your AI initiatives through pilot projects and iterative cycles, the next challenge is to scale these successful experiments to have a broader impact across your organisation.

However, scaling isn’t just about expanding the scope of your projects. It’s also about institutionalising the insights, methodologies, and technologies you’ve developed, ensuring they become a permanent part of your operational DNA.

The first step in scaling is resource allocation.

Based on the outcomes of your pilot projects and the Build-Measure-Learn cycles, you should have a good sense of what resources—both human and technological—are required to scale successfully.

This is the time to secure those resources, whether it’s specialised talent, additional computing power, or even external partnerships.

Another crucial factor is knowledge transfer.

The learnings from your pilot projects should be systematically documented and disseminated.

This serves two purposes: First, it ensures that the entire organisation benefits from the insights gathered, and second, it provides a blueprint for future AI projects.

Having a well-documented methodology can significantly reduce the learning curve and the time-to-market for subsequent initiatives.

As you scale, don’t lose sight of governance and ethics.

The complexities and impact of AI grow exponentially as you expand its footprint within your organisation.

Implementing rigorous governance protocols ensures that you maintain quality, data privacy, and ethical considerations. Remember, a poorly managed AI initiative can do more than just fail—it can erode customer trust and expose your company to legal risks.

Finally, sustain innovation.

AI is a rapidly evolving field, and what is cutting-edge today may be passé tomorrow.

Keep an ear to the ground for advancements in AI technologies and methodologies.

Create a culture that encourages continuous learning and experimentation. Periodically revisit your AI strategy to ensure it aligns with the latest trends and capabilities.

In essence, scaling AI is not a finish line but a new starting point. It demands continuous attention, adaptation, and refinement.

By focusing on resource allocation, knowledge transfer, governance, and sustained innovation, you can not only scale your AI initiatives effectively but also embed them as a lasting competitive advantage for your business.

Al Cattell
Al Cattell