An AI decision making framework, to help leaders in the AI age

Most AI projects fail at strategy - not technology.

The AI Value Canvas is a free tool for deciding whether an AI idea actually deserves your time - think of it as a toolkit an AI Architect or Strategist might use - before any budget or code gets committed. Download it, keep it, use it on the next idea - and the one after that.

Download the Canvas (free Word doc) ↓ See how it works
Just your name and email - editable Word document, space expands as you write.
The problem

Two numbers worth sitting with

Different research, different methods - same story. Most organizations invest heavily in AI and end up with little to show for it.

95%
of enterprise generative AI pilots fail to deliver ROIMIT Sloan / Project NANDA, 2025
74%
of companies struggle to scale AI value, even after a working pilotBoston Consulting Group, 2024
#1
reason for failure: teams skip the strategy and jump straight to buildingCommon pattern across both studies

Traditional AI development starts with technical feasibility - can we build it, do we have the data. What gets skipped is the set of questions that actually determine whether it survives contact with reality: do users want it, can we build it responsibly, does it make economic sense. The Canvas exists to force those questions first.

The framework

Nine elements, four domains

Worked through in order - left to right, top to bottom - so every AI idea gets grounded in real user need before anyone discusses a model.

🟡 Customer Understanding

Do users actually want this?

  • Customer / User Segments
  • Jobs to Be Done
  • Pains & Gains

🟢 AI Opportunity & Solution

Can we solve this with AI - not just rules?

  • AI Opportunity Fit
  • Solution Concept / Interaction Flow

🔵 Data, Model & Risk

Can we build and maintain this responsibly?

  • Data & Model Considerations
  • Risks & Guardrails

🟠 Value & Cost

Does this create more value than it costs?

  • Value & Metrics of Success
  • Cost Drivers
Customer → Solution → Data & Risk → Value & Cost
Eric Wong
Who made this

Eric Wong

18 years in IT, product development & corporate leadership

Eric has spent his career at the intersection of business innovation, digital transformation, and technology delivery - leading large transformation initiatives and cross-functional teams inside large organizations before turning to training and advisory work.

The AI Value Canvas grew out of a simple observation: the AI projects that failed weren't held back by weak technology. They were held back by skipped strategy teams that never properly answered whether users wanted the thing, whether AI was even the right tool, or whether the economics held up.

Technology Digital Transformation Artificial Intelligence Corporate Innovation Agile Leadership
What you get

A working tool, not a brochure

The download is the Canvas itself - the classic one-page grid, laid out as an editable Word document. All four domains and nine elements, in the same color-coded layout you'd fill in on a wall. Type directly into any box in Word, Google Docs, or Pages and it expands as you write, so you're never boxed in.

  • All 4 domains and 9 elements, in the familiar one-page canvas layout
  • Open and edit in Word, Google Docs, or Pages - no special software
  • Space for your project name, owner, and horizon
  • Short prompts under each box so you know what a strong answer looks like
Customer
Understanding
AI Opportunity
& Solution
Data, Model
& Risk
Value & Cost
AI VALUE CANVAS · one-page editable Word document
Free - just your name and email

Download the AI Value Canvas

Keep it. Use it on your next AI idea, and the one after that - it gets more useful the more places you've applied it.

That didn't go through - check your name and email and try again.
No spam. Just the Canvas, and occasionally news about the AI Value Canvas.
Step 2 - after you've filled it in

Get your Canvas rated

Run your completed Canvas through the AVC Star Rater - a free AI tool that reviews what you've written and tells you where it's strong and where it still needs work.

Rate my Canvas →