How to Create an AI Use Case

How to Create an AI Use Case

In “How to Create an AI Council,” I outlined how and why you should create your AI Council, which exists to communicate, create, assess, and oversee all your organization’s AI initiatives.

In “How to Create an AI Policy,” I unpacked the basics of what should be included in your organization’s AI policy, which is as a set of foundational guidelines that ensure your AI initiatives are implemented ethically, transparently, and in alignment with your organization’s goals.

Now, we’ll explore how to create an AI use case.

 

The AI landscape in early 2024

According to a recent McKinsey survey, AI adoption has surged to 72% in 2024, up from about 50% in previous years. More importantly, 65% of respondents report that their organizations are regularly using generative AI, nearly double the percentage from just ten months ago.

This isn’t just hype, either.

Organizations are seeing real benefits. The survey reports both cost decreases and revenue jumps in business units deploying AI technology. The largest share of respondents reported cost decreases in human resources, while meaningful revenue increases were most commonly seen in supply chain and inventory management.

What we’re seeing this year is a big shift where even small organizations are moving beyond the well-worn narrative that generative AI is only useful for specific, relatively low-value initiatives like content generation and content marketing.

The survey data confirms that more and more organizations are figuring out their AI use cases and those AI use cases are showing real results.

 

Defining an AI use case

An AI use case is a specific application of AI technology that’s used to solve a business problem or improve a process.

Your AI use case might involve the use of an LLM like Claude or ChatGPT, a martech solution like Jasper AI or Brand24, or even some sort of analytical AI, which you’d primarily use to extract insights, identify trends, and make data-driven decisions or predictions.

Regardless, what you’re looking for are areas where AI can add genuine value to your organization, typically through efficiency gains or performance lifts.

In Marketing Artificial Intelligence: AI, Marketing, and the Future of Business, Paul Roetzer and Mike Kaput write, “When you are getting started with AI and looking to build internal support, you will want to focus your investments on quick-win pilot projects with a narrowly defined scope and high probability of success.”

This approach allows you to learn, adapt, and scale your AI initiatives effectively. Which is precisely what you’ll need to do to remain competitive as we move through the next few years.

 

Identifying your AI use case

When identifying potential AI use cases, think about the tasks you carry out in your role—or the tasks your department is responsible for—and ask these four questions.

  1. Is this task data-driven?
  2. Is this task repetitive?
  3. Is this task predictive?
  4. Is this task generative?

If you can answer “yes” to one or more of these questions, you might have a good candidate for an AI use case.

 

Understanding the impact

 The next step is to systematically think through the AI use cases (tasks) you’ve identified. The Marketing AI Institute has created a template that I find particularly useful, which you can download here.

Using the Marketing AI Institute model, you’ll create a spreadsheet with the following headers:

  • Category: Classify the AI use case into a specific business function or department for organized analysis and implementation.
  • Task/Use Case: Clearly define the use case.
  • Interval: Specify how frequently the task is performed (e.g., daily, weekly, monthly) to understand its recurrence and impact.
  • Estimated Hours Per Month: Quantify the current time investment for the task to gauge potential time savings through AI implementation.
  • Existing Tech: Identify what technology, if any, you currently use to complete the task
  • Estimated Monthly Cost: Fill in the cost related to the tech you want to use

Working through each potential use case this way will allow you to consider both the value to intelligently automate all or a portion of the activity, with value being defined by potential time and money saved and the increased probability of achieving business goals, as well as the ability to intelligently automate the activity based on existing AI solutions that could be built with the right resources.

 

Choosing pilot projects

Once you’ve identified your use case, you’ll then launch your pilot project.

When selecting your pilot projects, keep these Marketing AI Institute guidelines in mind:

  1. Have a clear use case
  2. Assign an owner
  3. Know how you’ll measure impact (use SMART goals)
  4. Commit the time to properly onboard and train with the tech
  5. Choose a monthly payment plan until you prove value

Also remember the 30-90 Rule for AI Pilot Projects: Activate within 30 days and test over 90 days. Then, decide whether to keep or cancel based on the results.

 

Implementing AI use cases

Moving from identification to implementation requires careful planning. Ensure proper onboarding and training for your team. And remember, not every pilot project will succeed. That’s okay. The key is to experiment, learn, and adapt.

 

Aligning with your AI policy

As you develop your use cases, always ensure they align with the AI policy your council created. This alignment ensures that your AI initiatives remain ethical, transparent, and in line with your organization’s values and goals.

 

Bring your AI strategy to life

Creating AI use cases is where your AI strategy comes to life.

By following this structured approach—identifying potential use cases and carefully implementing pilot projects—you can start harnessing the power of AI in your organization.

Remember, the goal isn’t to implement AI everywhere all at once.

Start small, learn from your experiences, and scale your successes.

With the right approach, you can navigate the AI landscape and unlock significant value for your organization.

Need some GenAI implementation help? Check out Thrive’s free AI Solutions here.

Never miss an insight. We’ll email you when new articles are published.
ReLATED ARTICLES