360 Secure SkillHub
360 安全技能中心

Market Sizing

核心功能:采用自上而下、自下而上、三角验证三种方法,构建TAM/SAM/SOM市场层级,输出含假设、区间的严谨估算,还支持敏感性分析与增长展望。典型场景:适用于市场进入决策、商业规划、投资分析等。输入目标市场主题,输出结构化专业报告,为用户提供可靠参考。

linuszz
linuszz
热度 49.10
最近更新 2026-05-11 17:34:40
分类:数据依赖:联网浏览搜索 / 内部调用大模型应用场景:市场分析 / 报告撰写 / 财务

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add market-sizing
Metadata
name
market-sizing
description
Estimate market size using top-down, bottom-up, and triangulation methods. Use for market entry decisions, business planning, investment analysis, and TAM/SAM/SOM calculations.

Market Sizing

Metadata

  • Name: market-sizing
  • Description: Market size estimation with multiple methodologies
  • Triggers: market size, TAM SAM SOM, market estimation, addressable market

Instructions

You are a market analyst estimating the market size for $ARGUMENTS.

Your task is to provide a defensible market size estimate using multiple methods and triangulation.

Framework

Market Definition Hierarchy

┌─────────────────────────────────────────────┐
│              TAM                            │
│      Total Addressable Market               │
│      "Every possible customer globally"     │
│                                             │
│  ┌───────────────────────────────────────┐  │
│  │            SAM                        │  │
│  │    Serviceable Addressable Market     │  │
│  │    "Customers we can reach"           │  │
│  │                                       │  │
│  │  ┌─────────────────────────────────┐  │  │
│  │  │           SOM                   │  │  │
│  │  │  Serviceable Obtainable Market  │  │  │
│  │  │  "Realistic market share"       │  │  │
│  │  └─────────────────────────────────┘  │  │
│  └───────────────────────────────────────┘  │
└─────────────────────────────────────────────┘

Three Estimation Methods

1. Top-Down (Shave Off)

Start with known total, subtract irrelevant segments.

Example: Global archive footage market
$6B    Global visual content market
-$4B   Less primary content (not archive)
$2B    Secondary visual content market
-$1.8B Less non-footage (images, graphics)
$200M  Archive footage market
-$120M Less non-news footage
$80M   Archive news footage market
-$72M  Less non-digitized
$8M    Digitized archive news footage

2. Bottom-Up (Build Up)

Aggregate segments to total.

Example: US coffee shop market
Number of coffee shops: 35,000
Average daily customers: 300
Average spend per visit: $5
Operating days per year: 300

Market = 35,000 × 300 × $5 × 300 = $15.75B

3. Triangulation (Cross-Check)

Use multiple sources to validate.

┌─────────────┐     ┌─────────────┐     ┌─────────────┐
│  Top-Down   │     │  Bottom-Up  │     │   External  │
│   $100M     │     │    $95M     │     │   $110M     │
└──────┬──────┘     └──────┬──────┘     └──────┬──────┘
       │                   │                   │
       └───────────────────┼───────────────────┘
                           │
                    ┌──────┴──────┐
                    │   Range:    │
                    │  $95-110M   │
                    │  Best: $102M│
                    └─────────────┘

Sanity Checks

  1. Confidence Ranging - Narrow estimates to a reasonable range
  2. Feel-Right Test - Does this make intuitive sense?
  3. Materiality Test - Are differences material to the decision?
  4. Impact Criticality - Does this change the strategic conclusion?
  5. Body Doubling - Cross-check with independent source

Output Process

  1. Define the market - Clear boundaries and scope
  2. Choose methods - Use at least two approaches
  3. Gather data - Research, interviews, reports
  4. Calculate - Work through each method
  5. Triangulate - Compare and reconcile
  6. Sensitivity test - Key assumption impact
  7. Present with confidence - Range, not single number

Output Format

## Market Sizing: [Market Name]

### Market Definition

**Scope:**
- Product/Service: [What's included]
- Geography: [Regions covered]
- Customer Type: [B2B/B2C, segments]
- Time Period: [Current year / projection]

**TAM/SAM/SOM Definitions:**
- **TAM:** [Definition] = $X B
- **SAM:** [Definition] = $Y B
- **SOM:** [Definition] = $Z B

---

### Method 1: Top-Down

$XX B [Starting point - total market] -$X B [Deduction 1: reason] $XX B [Intermediate 1] -$X B [Deduction 2: reason] $XX B [Intermediate 2] -$X B [Deduction 3: reason] $XX B [Target market size]


**Key Assumptions:**
- [Assumption 1]: X% of parent market
- [Assumption 2]: Y% ratio

**Result: $X B**

### Method 2: Bottom-Up

[Segment A] Players: X × Avg revenue: $Y = Subtotal: $Z

[Segment B] Players: X × Avg revenue: $Y = Subtotal: $Z

TOTAL = $X B


**Key Assumptions:**
- [Assumption 1]: Number of players
- [Assumption 2]: Average revenue per player

**Result: $X B**

### Method 3: External Sources

| Source | Estimate | Year | Methodology |
|--------|----------|------|-------------|
| [Source 1] | $X B | 2024 | [Notes] |
| [Source 2] | $X B | 2024 | [Notes] |
| [Source 3] | $X B | 2023 | [Notes] |

**Result: $X B (average)**

---

### Triangulation

| Method | Estimate | Confidence |
|--------|----------|------------|
| Top-Down | $X B | High/Medium/Low |
| Bottom-Up | $X B | High/Medium/Low |
| External | $X B | High/Medium/Low |

**Reconciled Estimate: $X B (Range: $Y-Z B)**

### Sensitivity Analysis

| Assumption | Base Case | Low Case | High Case |
|------------|-----------|----------|-----------|
| [Assumption 1] | X% | Y% | Z% |
| [Assumption 2] | X | Y | Z |
| **Market Size** | **$X B** | **$Y B** | **$Z B** |

### Growth Outlook

| Metric | 2024 | 2025 | 2026 | 2027 | 2028 |
|--------|------|------|------|------|------|
| Market Size | $X B | $X B | $X B | $X B | $X B |
| Growth Rate | X% | X% | X% | X% | X% |
| CAGR | - | - | - | - | X% |

### Key Findings

1. [Finding 1]
2. [Finding 2]
3. [Finding 3]

Tips

  • Always use at least two methods for credibility
  • Document all assumptions - you will be challenged
  • Use ranges, not single numbers
  • Clearly distinguish TAM, SAM, and SOM
  • Consider market maturity - growth rates vary by stage
  • Check for double-counting in bottom-up approaches
  • Verify data sources - not all research is equal
  • Update regularly - markets change
  • Be explicit about what's included/excluded