Irina Sedenko
Research Director, Data & Analytics
Info-Tech Research Group

Mark Tauschek
Distinguished Analyst & VP Fellowships
Info-Tech Research Group

IT leaders are being inundated with demand for AI use cases. Since the release of ChatGPT 3.5 in November 2022, generative AI has hit the mainstream. With this rapidly increasing awareness has come immediate demand from business units and departments in organizations of all sizes across all industries. This is a golden opportunity for IT to deliver value by helping organizational leaders understand the “art of the possible.” IT leaders that can help identify and select the AI use cases that can offer value to the organization quickly will be recognized as innovators.

The challenge will be developing a systematic approach to identifying, evaluating, and selecting the pilot use cases that can deliver measurable value, be implemented quickly, and scale to production if proven valuable. You’ll need an approach to catalog the longlist of AI use cases, determine organizational value and readiness, narrow the longlist to a shortlist, and select one or two pilot use cases that can be implemented and demonstrate value quickly.

IT leaders must take advantage of this opportunity to lead and work with the organization to deliver measurable value and demonstrate that it can move beyond a trusted advisor to a recognized innovator.

Arzoo Wadhvaniya
Research Specialist
Info-Tech Research Group

Andrew Sharp
Research Director
Info-Tech Research Group

You’ve been asked to identify and select practical, valuable, and right-sized pilot use cases for AI on behalf of your organization. There’s a large pool of ideas, and you must:

• Get input from the business on potential use cases.
• Align use cases to your business and IT strategy.
• Select the best use cases from among the many available options.

It’s difficult to identify a pilot because:

• You don’t have a methodology or a process in place to arrive at a shortlist.
• You haven’t established criteria necessary to select use cases that would add value to your business.
• You’re unsure of the readiness of your organization to pilot AI technologies.

Follow Info-Tech’s methodology and fit-for-purpose tools to:

• Give AI in your organization a purpose aligned to business value.
• Develop an iterative, structured, and reusable process to identify, rationalize, and prioritize potential AI use cases.
• Select a single pilot project.

Carefully consider value when choosing AI use cases based on corporate priorities. Consider creating a specialized team to structure and accelerate scaling.

Ask yourself: Do we track the use cases in the pipeline against clear objectives and key results?

Adopt a consistent execution model, with agile build and validation cycles for AI use cases. The development of minimum viable products (MVPs), which add features and users as they are scaled and incorporated into the operating model, begins with prototypes that gather early end-user input. Use cases and associated operating models are eventually implemented throughout the entire organization.

Ask yourself: Do we have a systematic approach that prioritizes scaling of use cases based on value?

Set up an enablement role to make sure that as new AI applications are developed, they are accessible to the teams across the organization that require them. The enabling function may in some cases assign specialized teams to develop any lacking capabilities.

Ask yourself: Do we have identified roles that can continue to identify missing features after deployment?

There are many potential use cases, but you may only have time, skills, and funding to run one pilot. Where can you best drive value?

Your Challenge:

AI has great potential but also carries great risks. How do you balance value and risk to deliver AI pilot projects to your organization?

The capabilities, costs, and challenges of AI, as well as the approaches to AI, are evolving rapidly. How do you manage risks?

Educate key stakeholders on AI technology fundamentals with support from Info-Tech.

Assess use cases in the context of your organization’s risk appetite.

Involve cross-disciplinary stakeholders in brainstorming and decision making.

Align to organizational goals and clearly define the “why” – the purpose of the pilot.

Keep the pilot right-sized and focus on making something work.

Evaluate your readiness to pilot and scale.

Be ready to walk away from pilot projects that aren’t delivering expected value.

Info-Tech’s Approach

1. Gather a longlist.
Identify potential use cases with input from business and IT stakeholders.

2. Cut to a shortlist.
Quickly sort by value and readiness to identify a shortlist of five or six ideas.

3. Select a pilot.
Conduct a focused value-readiness exercise to identify one or two pilot use cases.

Implement a pilot AI use case.

The Discussion List tab helps the working group sort use cases into different quadrants.

The Summary tab provides the average value-readiness score and ranks the use cases accordingly.

The Scoring Scales tab provides criteria to rank the use cases.

• Create a working group and identify AI champions from each area of the business.
• Identify key elements of value and AI readiness for your organization.
• Identify a longlist of AI use cases.
• Work with AI champions to initially narrow down the longlist in terms of suitability, readiness, and value of use cases.

• Conduct a voting exercise with the working group to prioritize high-value, high-readiness use cases for AI.

• Do a deep dive on the value and readiness of use cases on the shortlist.
• Summarize your findings and select a pilot project.

• A longlist of documented AI use cases.
• An identified working group to support the selection process.

• A shortlist of five to six high-value, high-readiness use cases for AI.

• Ideas developed, socialized, and prioritized.
• First use case selected and approved.

Call #1: Review Info-Tech’s methodology, discuss prerequisites for this project, identify roles in the working group, and review approaches and tools for building a longlist.

Call #2: Discuss how to run a voting exercise to cut your longlist to a shortlist.

Call #3: Develop value and readiness scoring scales. Conduct a detailed assessment of your shortlist. Select a pilot project.