Skip to content

florianbonnet14/ThePowerOfAnalytics_ClaudeSkills / executive-summary-writer

executive-summary-writer

Verified

Transform analysis findings into clear, actionable executive summaries using the MAIN framework (Motive, Answer, Impact, Next steps) and Pyramidal Principle. Use when completing any analysis that needs stakeholder buy-in, preparing for leadership meetings, documenting investigation findings, writing business reviews, creating post-mortems, proposing initiatives based on data, or communicating insights to executives.

references/main-framework.md

View on GitHub
# MAIN Framework Reference

Complete guide to the MAIN framework for executive summaries.

## Framework Overview

MAIN is an acronym for:
- **M**otive - Why this matters
- **A**nswer - What we found and recommend
- **I**mpact - Quantified business implications
- **N**ext steps - Specific actions with owners

## M - Motive

**Purpose:** Set context and explain why this matters

### What to Include

- The business context or problem being addressed
- Why this analysis was conducted now
- Recent trends or changes that triggered investigation
- Who is affected and how

### Structure

```
[1-2 sentences on business context]
[1 sentence on what triggered this analysis]
[1 sentence on scope/focus]
```

### Good Examples

✓ "Our Q3 conversion rate dropped 15% vs. Q2, falling short of our target and threatening our annual growth goals. This analysis investigates the root causes to inform immediate corrective actions."

✓ "As we prepare to launch our premium tier next month, understanding current customer segmentation and willingness to pay is critical. This analysis examines upgrade behavior from our past year of data."

✓ "Following customer feedback about slow app performance, we analyzed user behavior across 50,000 sessions in the past 30 days to identify specific pain points and their impact on retention."

### Bad Examples

✗ "We did an analysis." (No context or motivation)
✗ "Here are some interesting findings about users." (Vague purpose)
✗ "The CEO asked us to look at this." (Doesn't explain business context)

**Length:** 2-4 sentences, 50-100 words maximum

---

## A - Answer

**Purpose:** State your main findings and recommendations upfront

### What to Include

- The 1-sentence answer to your investigation question
- Your recommendation (what should be done)
- The expected impact if recommendation is followed

### Structure

```
[1 sentence: What we found]
[1 sentence: What we recommend]
[1 sentence: Expected outcome]
```

### The "So What?" Test

Your Answer section should pass this test:
- Can someone act on this without reading further? ✓
- Is it clear what success looks like? ✓
- Would a busy executive understand the takeaway? ✓

### Good Examples

✓ "The conversion drop is driven by new customers (Month 1) converting 30% less than historical cohorts, while returning customers perform normally. We recommend improving the onboarding experience by adding a guided tutorial and simplifying initial setup. If successful, this should recover 10 percentage points within 6 weeks."

✓ "65% of our users fall into the 'Experimenter' segment who want templates and inspiration rather than blank-slate creation tools. We should shift product development toward template marketplace and curation features. This could increase activation by 25% and improve Week 1 retention by 15 percentage points."

✓ "Slow loading times (>3 seconds) affect only 12% of sessions but account for 40% of churn. Prioritizing performance optimization for our image rendering pipeline should reduce churn by 5-7 percentage points (~$2M annual revenue impact)."

### Bad Examples

✗ "We found some interesting patterns." (No specifics)
✗ "Users behave differently based on many factors." (Too vague)
✗ "The data shows conversion varies." (So what? What do we do?)

**Length:** 3-5 sentences, 75-125 words maximum

---

## I - Impact

**Purpose:** Quantify business implications

### What to Include

- Business metrics affected (revenue, users, costs)
- Magnitude of impact (with numbers)
- Timeline (when impact will be felt)
- Risks if no action taken

### Structure

```
[Quantified positive impact if action taken]
[Quantified negative impact if no action taken]
[Timeline and confidence level]
```

### Impact Categories

**Financial Impact:**
- Revenue increase/decrease: "$500K annual revenue at risk"
- Cost savings: "Could save $200K in ops costs"
- LTV impact: "15% LTV improvement = $2.5M over 12 months"

**Customer Impact:**
- User growth: "Could acquire 10K additional users per month"
- Retention: "5 percentage point retention improvement"
- Satisfaction: "NPS increase from 40 to 55"

**Operational Impact:**
- Efficiency: "Reduce fulfillment time by 30%"
- Quality: "Cut error rate from 5% to 2%"
- Team capacity: "Frees up 20 engineering hours per week"

### Good Examples

✓ "If we don't address this, we'll miss our Q4 target by $3M (15% shortfall). Implementing the recommended changes should recover $2.5M of this gap within 8 weeks, based on similar improvements we made in Q1 2023. Risk: If underlying product-market fit has shifted, recovery may take longer."

✓ "This represents 40,000 potential customers per month who abandon during onboarding. At our standard 15% conversion rate, that's 6,000 lost conversions monthly ($1.8M annual revenue). Quick wins could capture 25% of this opportunity (1,500 conversions, $450K) within one quarter."

✓ "Current manual process costs 200 engineering hours monthly. Automation would save 80% of this time (160 hours = $16K/month, $192K annually) and reduce errors from 8% to <1%. Payback period: 3 months."

### Bad Examples

✗ "This is important." (No quantification)
✗ "Could help revenue." (How much? When?)
✗ "Customers will be happier." (Measure how?)

### Pro Tips

- Always include numbers with units ($, %, absolute values)
- Show both upside and downside scenarios
- Be honest about uncertainty ("estimate: $1-2M" or "confidence: medium")
- Connect to company-level goals/OKRs when possible

**Length:** 3-5 sentences, 75-150 words

---

## N - Next Steps

**Purpose:** Define clear actions with owners and timelines

### What to Include

- Specific actions to be taken (not vague intentions)
- Owner for each action (name, not team)
- Timeline/ETA for each action
- Success criteria or milestones

### Structure

```
1. [Action] - Owner: [Name] - ETA: [Date] - Success: [Metric]
2. [Action] - Owner: [Name] - ETA: [Date] - Success: [Metric]
3. [Action] - Owner: [Name] - ETA: [Date] - Success: [Metric]
```

### Action Clarity Spectrum

**❌ Too Vague:**
- "Improve onboarding"
- "Talk to customers"
- "Fix the problem"
- "Monitor the situation"

**✓ Specific:**
- "Design and A/B test 3 new onboarding flows"
- "Conduct 15 customer interviews with Month-1 churned users"
- "Implement lazy loading for images to reduce load time to <2s"
- "Set up daily dashboard with alert if conversion drops >5%"

### Good Examples

✓ Next Steps:
1. Design new onboarding tutorial - Owner: Jane (Design) - ETA: Nov 30
   Success: Tutorial designs approved by stakeholders
   
2. Build and deploy tutorial v1 - Owner: Mike (Eng) - ETA: Dec 15
   Success: Tutorial live to 10% of users
   
3. Run A/B test for 2 weeks - Owner: Sarah (PM) - ETA: Dec 31
   Success: Statistical significance achieved, decision on rollout
   
4. Analyze results and recommend rollout - Owner: Sarah - ETA: Jan 5
   Success: Exec team has data to decide on full rollout
   
5. If successful, rollout to 100% - Owner: Mike - ETA: Jan 15
   Success: All new users see tutorial, monitor Week 1 retention

### Bad Examples

✗ Next Steps:
- Someone should look into this
- We'll fix it soon
- TBD
- Product team to handle

### Pro Tips

- Limit to 3-7 actions (more = analysis paralysis)
- First action should start within 1 week
- Include decision gates ("if test successful, then...")
- Add review meeting date at end

**Length:** 3-7 action items, 100-200 words total

---

## The Pyramidal Principle

**Core Concept:** Answer first, then support with evidence

### Structure

```
Level 1: Executive Summary (MAIN)
    ↓ (If stakeholder wants more detail)
Level 2: Key Supporting Insights (3-5 points)
    ↓ (If stakeholder wants even more detail)
Level 3: Detailed Analysis & Data
```

### Visual

```
         [ANSWER]
        /    |    \
    [Key] [Key] [Key]
     /|\    /|\   /|\
   [D][D][D] etc...

Where:
- Top = Main conclusion
- Middle = Supporting insights
- Bottom = Detailed data/charts
```

### Why It Works

1. **Respects stakeholder time** - Get answer immediately
2. **Enables selective drill-down** - Dig deeper only where interested
3. **Prevents burial of lede** - Most important info comes first
4. **Supports busy executives** - Can stop reading at any level

### How to Apply

**Page 1: Executive Summary**
- Use MAIN framework
- Should be self-contained
- Reader can stop here and act

**Pages 2-4: Key Insights**
- 3-5 main findings that support your Answer
- Each insight gets its own section
- Include 1-2 charts per insight
- Keep it high-level still

**Pages 5+: Supporting Analysis**
- Detailed data tables
- Additional segmentation
- Methodology explanations
- Alternative analyses considered
- Appendices

---

## Writing Style Guidelines

### Use Active Voice
❌ "It was determined that..."
✓ "We determined that..."

❌ "An improvement of 15% was observed"
✓ "We observed a 15% improvement"

### Use Specific Numbers
❌ "Many customers"
✓ "40% of customers (12,000 users)"

❌ "Significantly better"
✓ "25% better (from 40% to 50%)"

### Use Present/Future Tense for Recommendations
❌ "We would recommend..."
✓ "We recommend..."

❌ "This could potentially improve..."
✓ "This will improve..."

### Avoid Hedging Language
❌ "We think maybe this might help possibly"
✓ "We recommend this because [evidence]"

**Note:** Being direct doesn't mean ignoring uncertainty. State confidence explicitly: "We recommend X (confidence: medium, assuming Y holds true)."

### Keep Sentences Short
❌ "The analysis that we conducted over the past three weeks across multiple customer segments and various metrics including retention, conversion, and engagement, which involved extensive data collection and statistical validation, clearly demonstrates that..."

✓ "Our three-week analysis shows: retention dropped 15%, specifically in the SMB segment."

### Use Headers and White Space
Break up dense text:
- Headers every 3-4 paragraphs
- Bullet points for lists
- Bold key numbers/findings
- White space between sections
receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementationStars: 322kUpdated Claude CodeCodexVerifiedInstall
requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirementsStars: 322kUpdated Claude CodeCodexVerifiedInstall
using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallbackStars: 322kUpdated Claude CodeCodexVerifiedInstall
verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions alwaysStars: 322kUpdated Claude CodeCodexVerifiedInstall