strategyplaybook.aiSqwyz
Sign in
← All 124 tools
Research · Diagnose, Explore, Choose, Plan, Review

Evidence log

The unglamorous one that keeps the whole engagement honest. Every material claim gets a source, a date and a confidence rating, so that in month four you can tell what you checked from what you assumed.

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.

Stage
01 Diagnose, 02 Explore, 03 Choose, 04 Plan, 05 Review
Works at
All levels
Maturity
Any
Time
Ongoing — ten minutes after each piece of work

What it is

A running record of every material claim the strategy rests on, with its source, its date and how strongly it is evidenced. It is a habit rather than a workshop, and it is what lets you tell, in month four, what you checked from what you assumed.

The claim
A specific statement about the world that could be true or false. Conclusions are not claims; the facts underneath them are.
Source
Publisher, title, date and a link that was actually opened. A citation nobody retrieved is not a citation.
Confidence
Verified primary source, credible secondary, single unverified mention, or our own assumption. Most strategy rests on the last two and rarely admits it.
Date
When the claim was established. Market facts decay, and a two-year-old share figure is not wrong but it is not current.
Status
Whether it has since been confirmed, contradicted or left untested.
Decision relevance
Which decisions depend on it. Lets you answer the only question that matters at a decision point: which of these, if false, would change what we are about to do?
The mistake people makeThe log has no value unless it is read before decisions. Kept diligently and never consulted, it is an expensive way to feel rigorous — build the review into the decision forum rather than into the project plan.
What it’s forThe strategy is accumulating claims and you need to know which are evidenced, by what, and how strongly.
What it’s not forNever — but it is a habit rather than a workshop. Keep it running alongside everything else.

How you run it

  1. Log the claim, not the conclusion"Two competitors took six points of share" is a claim. "We are losing" is a conclusion. Only claims can be sourced.
  2. Record the source properlyPublisher, title, date, and a link that was actually opened. A citation nobody has fetched is not a citation.
  3. Rate the confidence, honestlyVerified primary source, credible secondary, single unverified mention, or our own assumption. Most strategy rests on the last two and nobody admits it.
  4. Date everythingMarket facts decay. A share figure from two years ago is not wrong, but it is not current, and the difference matters at decision time.
  5. Review it before every major decisionThe question is always the same: which of these claims, if false, would change what we are about to do?

The prompt

Two ways to run it

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.

🔒 Prompt locked

Everything above is free — what this tool is, what it is for, what it is not for, and how to run it as a workshop. Three tool prompts a month are free with an account; beyond three, and for the printable canvas, it is the paid part.

See plans — from £19 a month

Your playbook

It lands in the earliest stage this tool suits. Move it on the Playbook page.

You’ll need

  • Every output produced so far
  • The sources behind them, including the ones that were only half-read

You’ll end up with

  • One row per material claim: the claim, the source, the date, the confidence
  • A visible list of the claims nobody has evidenced
  • A record of what has gone stale

Runs well with

Acquisition screen and synergy thesisAdjacency mapAfter action reviewAI operating model options