Purpose
Transform raw operational data, status notes, or project details into a concise narrative that leads with the bottom line, makes the implication explicit, and ends with a clear decision or action ask — so executives understand what matters and what they need to do in under 90 seconds.
When to Use
Invoke this skill when the user asks to:
- Turn a status update or data dump into an executive summary
- Prepare a slide narrative or talking points for a leadership review
- Write a board, steering committee, or QBR update
- Reframe an operational update so it answers "so what" for senior leaders
- Distill a long project report into a one-paragraph or one-slide brief
- Surface a decision or resource ask buried inside operational details
Inputs Required
Before starting, collect:
- Raw input: The data, status notes, metrics, or operational details to convert (paste it in — bullet form is fine)
- Audience: Who reads this — CEO, board, VP, steering committee, cross-functional execs
- Decision or ask (if any): Is there a decision needed, a resource ask, or is this informational only?
- Headline reality: In one phrase, is the news good, bad, mixed, or on-track?
- Time horizon: What period this covers (this week, this quarter, project-to-date)
- Constraints to flag: Risks, blockers, dependencies, or trade-offs leadership should know
- Format target: One paragraph / one slide / email / verbal talking points
If the decision/ask is missing, default to an informational framing but flag any implicit decision the data points toward.
Instructions
Step 1: Extract the Bottom Line
Read the raw input and answer one question: If the executive remembered only one sentence, what should it be?
- Write a single declarative headline — outcome first, not activity ("Pipeline is 18% below target, driven by enterprise stall" — not "We held 40 sales calls").
- Lead with the verdict, not the journey. Status (on-track / at-risk / off-track) belongs in the first six words.
- Cut all process language: meetings held, work in progress, things "being looked into."
Step 2: Make the "So What" Explicit
Executives don't want data — they want the implication of the data.
- For every key metric or fact, ask: "So what does this mean for the business?"
- Translate operational facts into business consequences: revenue, cost, risk, timeline, customer impact, strategic goal.
- Example: "Onboarding takes 14 days" → "So what" → "Slow onboarding is delaying $400K in Q3 recognized revenue."
- If a fact has no "so what," it doesn't belong in an executive narrative — drop it.
Step 3: Structure with BLUF (Bottom Line Up Front)
Order the narrative so the conclusion comes first:
- Headline (1 sentence): The bottom line from Step 1.
- So What (1–2 sentences): The business implication and why it matters now.
- Evidence (2–4 supporting points): The minimum facts that back the headline — quantified, not exhaustive.
- The Ask (1–2 sentences): The decision needed, resource requested, or "no action needed — informational" if applicable.
Never bury the ask at the bottom of a paragraph of context. The reader should see it without hunting.
Step 4: Quantify and Tighten
- Replace vague qualifiers ("significantly," "a lot," "soon") with numbers, percentages, or dates.
- Compress: every sentence must earn its place. Aim to cut 40–60% of the raw input's word count.
- Use comparison anchors — "vs. plan," "vs. last quarter," "vs. target" — so numbers carry meaning.
- Strip jargon, acronyms the audience may not know, and internal team shorthand.
Step 5: Surface Risks and Trade-Offs Honestly
- State risks and blockers plainly — executives distrust updates with no bad news.
- For each risk: name it, state the impact, and state what's being done or what's needed. ("Vendor delay puts launch at risk for 2 weeks; mitigation requires approving overtime budget of $25K.")
- Frame trade-offs as choices, not complaints — give the executive something to decide on.
Step 6: Provide Action Recommendations
- If there's a decision, recommend a position — don't just present options neutrally. Executives value a point of view.
- Offer 1–3 ranked options when a decision is open, each with a one-line consequence.
- Make the recommended action specific enough to act on: who, what, by when.
Output Format
- Structure: BLUF order — Headline → So What → Evidence → Risks/Trade-offs → The Ask/Recommendation
- Length: 120–250 words for the core narrative; talking-points or slide variants stay under 100 words
- Format: Tight prose for the headline and so-what; scannable bullets for evidence and risks; explicit labeled ask
- Sections: Headline (bold) · So What · Key Evidence (3–4 bullets) · Risks & Trade-offs · Recommendation / Ask
- Tone: Calm, direct, confident — chief-of-staff voice. No hedging, no filler, no apologizing. Bad news stated as plainly as good news.
Quality Checklist
Before delivering output, verify:
- [ ] The headline states the outcome and status in the first sentence — no activity-first framing
- [ ] Every retained fact has an explicit or obvious "so what" — no orphan data points
- [ ] The ask or "no action needed" is clearly labeled and findable in under 5 seconds
- [ ] Vague qualifiers are replaced with numbers, dates, or comparison anchors
- [ ] At least one risk or trade-off is stated honestly (if any exist in the input)
- [ ] Core narrative is under 250 words and cuts 40%+ of the raw input's length
Notes
- The most common failure in executive updates is leading with activity ("here's what we did") instead of outcome ("here's where we are"). Always invert it.
- When the news is bad, resist softening — executives respond better to a clear problem with a recommended fix than to a buried, hedged update.
- If the input contains no decision and no risk, the narrative may be very short. Short is a feature, not a failure — don't pad to fill space.
- For verbal talking points, write the headline as the first thing spoken aloud, then pause for reaction before evidence.
How to use this skill
This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.
Installing in Claude Code:
~/.claude/skills/<folder-name>/SKILL.md— auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)<your-project>/.claude/skills/<folder-name>/SKILL.md— auto-loads only when working in that project (recommended for project-specific skills)
Restart Claude Code or open a new session to register.
Using in Claude.ai, ChatGPT, Gemini, or other AI tools:
- Drag this file into a new chat, or copy and paste its contents
- Send your task or request as your next message — the AI will apply the skill where it fits
For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills