Many organizations are caught in a conundrum in implementing their AI strategies.

Should CIOs and Digital Trailblazers seek widespread experimentation with genAI tools to increase AI literacy? Or, should they limit experimentation and focus on deploying to production and scaling adoption?

From McKinsey’s State of AI Report 2025, no more than 10% report scaling AI agents in any individual business function. Large enterprises ($5B or more) are far more likely to scale AI beyond pilots (49%) than businesses under $1B (27-32%).

Closing the gap between AI pilots and business value

I want to help leaders close the gap between moving beyond pilots and delivering business value. Additionally, if the AI winners are only large enterprises, it may mean the collapse of many hard-working owners, leaders, and employees at SMBs.

Before I reveal my approach, let me share two more data points.

Databricks 2026 State of AI Agents reports, “Companies that actively use AI governance put 12x more AI projects into production.” These companies likely adopt AI governance and strategy, as I recommended in developing a balanced AI governance strategy.

One more important data point from Databricks: “40% of the top AI use cases focus on customer experience and engagement.”

I find this surprising because more of the talk and early ROI around AI agents are focused on productivity and workflow efficiencies. If more businesses venture into AI in CX, I expect many more digital transformation opportunities. In fact, I predicted genAI would have its Uber/Airbnb moment in 2026, where a startup would disrupt an industry with an entirely new AI-first UX and business model.

Address the experiment chaos while supporting idea founders

So organizations need a funnel of smart ideas that they plan and pilot to get the winners into production. They need to have balanced objectives between efficiency and growth drivers. But of course, just getting AI into production is just the beginning of gaining adoption and delivering business value.

The problem is that CIOs need to get founders and agile teams to plan and pilot the most promising ideas. These ideas need a clear statement of business objectives, end-user value propositions, and success criteria to align everyone working on them. The process for submitting ideas needs to be easy for “founders”, require little upfront work, and collect enough information to help evaluate trade-offs and make priority decisions.

Experiments require a vetted AI vision statement

My solutions are simple but very effective. Can your organization afford the costs and risks of open experimentation, i.e., throwing tools at employees and letting them tinker? Equally troubling is governance that reaches the point of exhaustion, where legacy PMOs require business plans and multi-step stage gates before presenting an initiative for investment approval.

My simple solution goes back to the StarCIO Vision Statement Template I released in 2022 when Digital Trailblazer was published. Hundreds of companies are using it as part of their initiative intake process. This template solves three problems:

• Some companies had no real intake process, so only enhancement work and a handful of top-down strategic initiatives were getting prioritized. These companies used the StarCIO Vision Statement Template to create simple intake processes to get bottom-up ideas into their agile planning funnels.
• Other companies were drowning in the complexity of multi-slide PowerPoints or complex PMO tools for presenting ideas. Idea founders often had to run a separate, upfront planning process just to submit their idea. (Note: I wrote about how bad this problem is in chapters 9-10 of Digital Trailblazer.)
• A small number of companies struggled to manage their citizen development and data science programs. They deployed low-code/no-code and data visualization tools to business teams, but without a governance process to decide which ideas to develop. Some of these organizations ended up with significant technical debt because business teams shifted from too many spreadsheets to as many siloed dashboards and applications.

One-pagers drive alignment and quicker, easier decisions

StarCIO’s Vision Statement Template is one page. Anyone can fill it out without doing time-consuming upfront work. But it includes enough information to evaluate business value and determine whether the initiative should proceed to an agile planning phase.

Because our template is simple, many organizations have developed forms and low-code workflows to support idea intake using its structure. They then automate filling out the template and use it to evaluate ideas and communicate them to executives, teams, and stakeholders.

Announcing StarCIO’s AI Vision Statement Template

So, after four years, I’m ready to announce a major enhancement to the StarCIO’s Vision Statement Template. You can still use the original for initiatives. The new one is specific for reviewing AI experiments, products, and programs. Use it to evaluate LLMs for specific use cases, test genAI dev tools, experiment with AI agents on SaaS platforms, or develop AI agents.

The new template is still one page! It streamlines sections on Benefactors and Strategy. It then adds sections covering AI Value, Data Governance, Agentic and Automation Capabilities, and Risk Management.   

I’m going through a round of peer reviews and will then begin releasing it to Digital Trailblazers. Sign up below to get notified when it’s available.