Stakeholder readout
What it does
A readout is where research meets a decision that is already on someone's calendar. The structure that holds up is answer-first, the discipline behind the pyramid principle and BLUF briefings: the finding leads, and everything justifying it sits beneath. Ordering slides by the decisions they inform is the single change most readout decks need, and it's the one that gets skipped when the deck is built the night before.
This skill builds the readout from files you give it, which makes it the version to reach for when your material lives in exported docs and folders of transcripts. Each slide headline is written as a full sentence stating the finding, with the quote or count behind it attached underneath. It also drafts the speaker notes for the parts you would otherwise wing. Findings that don't survive the cut come back in a list with the reason, because the question from the room is usually about the one you left out.
What you get back
- A slide-by-slide outline, each slide titled with the finding written as a sentence.
- The evidence line under every headline: the quote or the count it rests on, with the participant named.
- Speaker notes per slide, written to be read once before you present.
- An executive summary that stands on its own for the people who will never open the deck.
- The findings that were cut, and why each one was cut.
- Open questions, with what you'd need in order to answer them.
How to use it
Example Prompts:
"Build a readout for the leadership team from the findings in this doc."
"Rewrite these slide titles so each one states the finding."
"Cut this to eight slides for a 20-minute slot and tell me what got dropped."
What it won't do
It won't design the deck. You get the structure and the words, and the visual build stays in whatever your team already uses. It also has no view on your organisation. It can't know which exec already promised this feature, or which finding is politically expensive this quarter, so read the cut list before you present, because that's where those tend to land.
Questions
What's the difference between this and the stakeholder readout generator?
Where the material comes from. This Claude skill works on files you hand it and needs no account, so it covers research that never lived in a platform. The stakeholder readout generator is a Great Question skill. It runs through our MCP and builds the deck out of the sessions and highlights already in your repo, with every slide linked back to its source.
Does it produce a PowerPoint file, or just the text?
It writes the readout content: the structure, the finding on each slide, and the speaker notes underneath. That drops into whatever deck template your team already uses, so the formatting stays yours.
How much raw material does it need before a readout is worth building?
Enough that you can already say the finding out loud. This is a write-up skill, so given a pile of untagged transcripts it will tell you the analysis hasn't happened yet. Run a synthesis or tagging pass first and bring the findings here.
