Lesson 1 of 6
Finding AI opportunities
7 min read
Your backlog has forty AI ideas. Half are dazzling demos that will never ship. Which two do you build first?
Two questions, every idea
Every candidate idea gets two honest scores. Value: how much does it matter if it works? Feasibility: can AI actually do this reliably, today, with the data you have? Plot both and the map draws itself.
AI ability is jagged — brilliant at some tasks, unreliable at others. Feasibility is your honest read on which side a given idea falls.
Start where they meet
The top-right quadrant — high value and high feasibility — is where you start. A dazzling idea AI can't do reliably is a hard bet, not a first project; a feasible idea nobody values is a distraction. The intersection is your shortlist.
Score feasibility on evidence, not optimism. "An LLM could probably do this" is a hypothesis — the honest score is "we tested it on real inputs and it held up."
The shape of it
- —Score every idea on value and feasibility — two honest reads.
- —High value plus high feasibility is your first project; build there.
- —High value, low feasibility is a hard bet to de-risk later — not skip forever.
An idea would transform the business, but AI can't do it reliably yet. Where does it go?
Continue in the app
Take the whole Discovering & Scoping AI Opportunities course — tracked
Get your personalized path, progress and streaks in the app — this lesson and every next one, in order.