Option generation sprint
The single highest-return hour in strategy. Teams routinely decide between two options when nine existed, and the eighth was the good one.
The printable canvas and the handoff into Claude Code are part of a paid plan. See what a plan includes. It needs the Claude desktop app on this machine, and your team and the prompt library already installed in that project — we cannot see your disk, so open it there.
What it is
A disciplined hour spent producing strategic options without evaluating any of them. It exists because teams reliably arrive at a decision with two options on the table, when nine were available and the good one was eighth. Generation and evaluation use different modes of thought and must not share a session.
- The restated diagnosis
- The challenge the options must answer, written in one line before anything else. Fuzzy questions produce fuzzy options.
- Forced coverage
- A grid — usually Ansoff — used to make sure you cover existing and new products, existing and new markets, rather than clustering around whatever you are already invested in.
- Imported analogues
- At least three options borrowed from a different industry. What would a subscription business, a franchise, or a marketplace do with these assets?
- Subtractive options
- What you could stop, exit or sell — plus doing nothing, argued properly. Growth-only option sets are half a set.
- The evaluation ban
- Nothing is judged until twelve exist. The moment someone says "that would never work", generation stops.
How you run it
- Restate the diagnosis at the topOptions are answers. A fuzzy question generates fuzzy options. Write the core challenge in one line first.
- Force coverage with a gridUse Ansoff or similar so you cover market and product moves, not just the ones you are already invested in.
- Import from elsewhereWhat would a subscription business, a franchise, a marketplace do with this asset? Three of your twelve should come from outside your sector.
- Ban evaluation until you have twelveThe moment someone says "that would never work", generation stops. The value is entirely in not filtering early.
- Only then, cluster and cutGroup, kill the duplicates, and carry a shortlist into scenario testing or a scoring tool.
The prompt
Run this tool in your own Claude
The short prompt starts your partner against the library on your disk. The long one carries everything with it and needs nothing installed.
Your playbook
It lands in the earliest stage this tool suits. Move it on the Playbook page.
You’ll need
- A stated diagnosis — what problem the options solve
- Business context captured
- A willingness not to evaluate anything for sixty minutes
You’ll end up with
- Twelve or more distinct options, unevaluated
- At least three from outside your industry
- A shortlist you would defend