The reading version
One email shouldn't take 45 minutes. It does because the email is three jobs at once: a status summary, a political document, and a tone-matching exercise with whatever voice your org expects.
The trick that works is handing the model *two* things, not one. Your messy notes give it the facts. Your last email gives it the tone. With both, the draft lands close enough that you're editing for judgment — not rewriting from scratch.
What only you know
AI handles structure, diplomacy, and specificity. It cannot know your stakeholders, their sensitivities, or the political context. Your job is the twenty-second scan before sending: is this the right thing to flag, to the right audience, at the right moment?
- Shape raw updates into sections
- Group wins by theme
- Enforce word count
- Draft the headline
- Decide what the single headline is
- Rank the blockers
- Name the risk you're worried about
- Sign your name to it
