Skill 寫法圖鑑

fin-earnings-summary

asgard-ai-platform | 金融與投資 | 原文 | MIT

摘要法說會的財務數字、財測和分析師 Q&A 重點

中文翻譯

---
name: "fin-earnings-summary"
description: "摘要並分析法說會,包含財務亮點、管理層談話、財測(guidance)與分析師 Q&A 的重點。當使用者需要消化法說會逐字稿、擷取關鍵財務數據與前瞻財測、比較實際結果與市場共識預期,或準備投資人會議時使用。就算使用者只說「幫我摘要這場法說會」、「管理層對下一季怎麼說」、「有超出預期嗎」或「法說會的重點」也要用。"
metadata:
  category: "WP-08 財務投資"
  tags: ["finance", "earnings", "investor-relations", "analysis"]
---

# 法說會摘要與分析

## 框架

```
IRON LAW: Separate Facts from Spin

Management presentations emphasize positives and downplay negatives.
Your job is to extract: (1) hard numbers, (2) forward guidance with
specificity level, and (3) what management DIDN'T say (notable omissions).

"Strong momentum" without numbers is spin.
"Revenue grew 15% to NT$2.3B" is a fact.
```

### 法說會結構

| 段落 | 時長 | 要擷取什麼 |
|---------|----------|----------------|
| **準備好的談話(Prepared Remarks)** | 15-20 分鐘 | 營收、EPS、毛利率、關鍵指標,並比較實際 vs 財測 vs 市場共識 |
| **管理層評論** | 10-15 分鐘 | 前瞻財測、策略性計畫、市場展望 |
| **Q&A** | 20-30 分鐘 | 分析師在擔心什麼(從他們的問題可以看出來)、管理層的坦率程度 |

### 要擷取的關鍵數據點

**財務亮點(對比市場共識)**
| 指標 | 實際 | 共識 | 超出/不及 | 年增 |
|--------|--------|----------|-----------|-----------|
| 營收 | ${X} | ${X} | 超出/不及 {%} | {%} |
| EPS | ${X} | ${X} | 超出/不及 {%} | {%} |
| 毛利率 | {%} | {%} | ±{pp} | ±{pp} |
| 營業利益率 | {%} | {%} | ±{pp} | ±{pp} |

**前瞻財測**
| 指標 | 財測 | 先前財測 | 變化 | 具體程度 |
|--------|---------|---------------|--------|------------|
| 下一季營收 | ${X}-${X} | ${X}-${X} | 上調/維持/下調 | 具體/籠統 |
| 全年 EPS | ${X}-${X} | ... | ... | ... |

**質化訊號**
- 新產品/新計畫的宣布
- 市場/需求的評論(轉強、穩定、轉弱)
- 競爭地位的更新
- 資本配置(庫藏股、股利、併購、資本支出)
- 語氣:有信心、謹慎、防禦、閃躲

### 分析步驟

**階段 1:擷取數字**
- 營收、EPS、毛利率:實際 vs 共識 vs 先前財測
- 每個項目是超出還是不及

**階段 2:評估財測**
- 財測是上調、維持還是下調?
- 財測有多具體?(區間 vs 籠統)
- 隱含的假設(成長率、毛利率走向)

**階段 3:讀 Q&A**
- 分析師在問什麼?(反映市場的擔憂)
- 管理層在哪裡迴避或給模糊的答案?(警訊)
- 準備好的談話中沒有的新資訊

**階段 4:對照市場反應**
- 法說後的股價變動
- 市場反應跟數字吻合嗎?(有時候超出預期卻被賣出 = 財測令人失望)

## 輸出格式

```markdown
# Earnings Summary: {Company} — {Quarter}

## Headline
{One sentence: beat/miss + most important takeaway}

## Financial Highlights
| Metric | Actual | Consensus | vs Consensus | YoY |
|--------|--------|----------|-------------|-----|
| Revenue | ... | ... | ... | ... |
| EPS | ... | ... | ... | ... |

## Forward Guidance
| Metric | New Guidance | Prior | Change |
|--------|------------|-------|--------|
| ... | ... | ... | Raised/Maintained/Lowered |

## Management Commentary (Key Points)
1. {key point with quote}
2. {key point}

## Q&A Highlights
- **Analyst concern**: {topic} → **Management response**: {summary}
- **Notable omission**: {what wasn't addressed}

## Market Reaction
- Post-earnings move: {+/-X%}
- Interpretation: {why the market reacted this way}

## Investment Implication
{What this means for the investment thesis — bullish/neutral/bearish signal}
```

## 常見陷阱

- **超出預期也可能是「低品質」的**:靠砍成本讓 EPS 超出預期,營收卻不及,是低品質的超出。營收超出預期、毛利率又擴張,才是高品質。要區分。
- **財測比實際數字更重要**:市場價格已經反映預期。一家公司第 3 季超出預期、卻下調第 4 季財測,股價常常會下跌。要專注在前瞻的訊號。
- **台灣法說會的特殊性**:許多台灣公司(尤其是台積電、聯發科)會辦英語法說會,其他則只有中文。就算你無法即時參加,公開資訊觀測站(MOPS)也有摘要說明。
- **市場共識預估可能過時**:如果預估很久沒更新,「超出預期」可能只是門檻設太低。要確認共識最近一次修正是什麼時候。
- **一季不構成趨勢**:不要對單一季度反應過度。看 3-4 季才能分辨真正的趨勢和雜訊。

## 參考資料

- 財務報表分析,見 data-financial-analysis skill
- 台灣股票資料來源,見 tw-stock-analysis skill

英文原文

---
name: "fin-earnings-summary"
description: "Summarize and analyze earnings calls (法說會) including financial highlights, management commentary, guidance, and analyst Q&A key takeaways. Use this skill when the user needs to digest an earnings call transcript, extract key financial data and forward guidance, compare actual results vs consensus estimates, or prepare for an investor meeting — even if they say 'summarize this earnings call', 'what did management say about next quarter', 'did they beat estimates', or 'key takeaways from the 法說會'."
metadata:
  category: "WP-08 財務投資"
  tags: ["finance", "earnings", "investor-relations", "analysis"]
---

# Earnings Call Summary & Analysis

## Framework

```
IRON LAW: Separate Facts from Spin

Management presentations emphasize positives and downplay negatives.
Your job is to extract: (1) hard numbers, (2) forward guidance with
specificity level, and (3) what management DIDN'T say (notable omissions).

"Strong momentum" without numbers is spin.
"Revenue grew 15% to NT$2.3B" is a fact.
```

### Earnings Call Structure

| Section | Duration | What to Extract |
|---------|----------|----------------|
| **Prepared Remarks** | 15-20 min | Revenue, EPS, margins, key metrics — actual vs guidance vs consensus |
| **Management Commentary** | 10-15 min | Forward guidance, strategic initiatives, market outlook |
| **Q&A** | 20-30 min | What analysts are worried about (revealed by their questions), management's candor level |

### Key Data Points to Extract

**Financial Highlights (vs Consensus)**
| Metric | Actual | Consensus | Beat/Miss | YoY Change |
|--------|--------|----------|-----------|-----------|
| Revenue | ${X} | ${X} | Beat/Miss by {%} | {%} |
| EPS | ${X} | ${X} | Beat/Miss by {%} | {%} |
| Gross Margin | {%} | {%} | ±{pp} | ±{pp} |
| Operating Margin | {%} | {%} | ±{pp} | ±{pp} |

**Forward Guidance**
| Metric | Guidance | Prior Guidance | Change | Specificity |
|--------|---------|---------------|--------|------------|
| Next Q Revenue | ${X}-${X} | ${X}-${X} | Raised/Maintained/Lowered | Specific/Vague |
| Full Year EPS | ${X}-${X} | ... | ... | ... |

**Qualitative Signals**
- New product/initiative announcements
- Market/demand commentary (strengthening, stable, weakening)
- Competitive positioning updates
- Capital allocation (buyback, dividend, M&A, capex)
- Tone: confident, cautious, defensive, evasive

### Analysis Steps

**Phase 1: Extract Numbers**
- Revenue, EPS, margins — actual vs consensus vs prior guidance
- Beat or miss on each line item

**Phase 2: Evaluate Guidance**
- Did they raise, maintain, or lower guidance?
- How specific is the guidance? (range vs vague)
- Implied assumptions (growth rate, margin trajectory)

**Phase 3: Read the Q&A**
- What are analysts asking? (reveals market concerns)
- Where does management deflect or give vague answers? (red flags)
- New information that wasn't in prepared remarks

**Phase 4: Compare to Market Reaction**
- Stock price move post-earnings
- Does the market reaction align with the numbers? (sometimes a beat is sold off = guidance disappointed)

## Output Format

```markdown
# Earnings Summary: {Company} — {Quarter}

## Headline
{One sentence: beat/miss + most important takeaway}

## Financial Highlights
| Metric | Actual | Consensus | vs Consensus | YoY |
|--------|--------|----------|-------------|-----|
| Revenue | ... | ... | ... | ... |
| EPS | ... | ... | ... | ... |

## Forward Guidance
| Metric | New Guidance | Prior | Change |
|--------|------------|-------|--------|
| ... | ... | ... | Raised/Maintained/Lowered |

## Management Commentary (Key Points)
1. {key point with quote}
2. {key point}

## Q&A Highlights
- **Analyst concern**: {topic} → **Management response**: {summary}
- **Notable omission**: {what wasn't addressed}

## Market Reaction
- Post-earnings move: {+/-X%}
- Interpretation: {why the market reacted this way}

## Investment Implication
{What this means for the investment thesis — bullish/neutral/bearish signal}
```

## Gotchas

- **Beats can be "low quality"**: Beating EPS by cutting costs while revenue misses is a low-quality beat. Revenue beats with margin expansion is high quality. Differentiate.
- **Guidance matters more than actuals**: The market prices in expectations. A company that beats Q3 but lowers Q4 guidance will often sell off. Focus on forward-looking signals.
- **Taiwan 法說會 specifics**: Many Taiwan companies (especially TSMC, MediaTek) hold English-language calls. Others are Chinese-only. MOPS has summary notes even if you can't attend live.
- **Consensus estimates can be stale**: If estimates haven't been updated recently, a "beat" may just mean the bar was set too low. Check when consensus was last revised.
- **One quarter doesn't make a trend**: Don't overreact to a single quarter. Look at 3-4 quarters to identify real trends vs noise.

## References

- For financial statement analysis, see the data-financial-analysis skill
- For Taiwan stock data sources, see the tw-stock-analysis skill