环节

中位数

头部20%

尾部30%

流失率

AI推荐曝光

100%

100%

100%

—

点击进入商品页

3.2%

8.5%

0.9%

96.8%

商品页成交

2.5%

6.8%

0.7%

97.5%

整体转化率

0.8%

5.8%

0.06%

99.2%

Plaintext
GCF-3模型
├── 环节一：引用优化（AI推荐率）
│   ├── 决策辅助内容建设
│   ├── 高权重信源布局
│   └── 多模型覆盖
│
├── 环节二：点击优化（AI→商品页点击率）
│   ├── AI推荐话术优化
│   ├── 商品标题/主图吸引力
│   └── 价格/促销信息前置
│
└── 环节三：成交优化（商品页成交率）
    ├── 详情页问答友好度
    ├── 评价区信任建设
    └── 下单路径简化

Plaintext
GEO整体效果 = AI推荐曝光量 × 点击率 × 成交率 × 客单价

问题类型

示例

购买意图

内容策略

信息型

"XX是什么""XX原理"

低

科普内容，建立认知

对比型

"XX和YY哪个好""XX品牌怎么样"

中

对比测评、选购指南

决策型

"XX推荐""XX哪家靠谱""XX多少钱"

高

场景化推荐、价格对比

Plaintext
豆包渠道月成交金额 = 月AI推荐曝光量 × 点击率 × 成交率 × 客单价

GEO月投入 = 内容生产费 + 平台发布费 + 详情页优化费 + 人力成本

ROI = 豆包渠道月成交利润 ÷ GEO月投入

指标

定义

数据来源

AI推荐曝光量

豆包等大模型推荐你品牌的次数

AI引用监测工具

点击率

从AI推荐到点击商品页的比例

抖店后台"豆包渠道"访客数÷推荐量

成交率

从商品页访问到下单的比例

抖店后台"豆包渠道"成交订单数÷访客数

客单价

豆包渠道平均每单金额

抖店后台"豆包渠道"成交金额÷订单数

ROI

成交利润÷GEO投入

财务核算

ROI区间

等级

行动建议

>3

优秀

加大投入，扩大覆盖场景

1-3

良好

保持投入，优化薄弱环节

0.5-1

一般

诊断三个环节，找出流失最严重的环节重点优化

<0.5

较差

暂停加投，先做全路径诊断和基础优化

Python
from typing import Dict, List, Tuple
from dataclasses import dataclass

@dataclass
class FunnelAnalysis:
    total_exposure: int
    total_clicks: int
    total_orders: int
    click_rate: float
    conversion_rate: float
    overall_rate: float
    avg_order_value: float
    monthly_revenue: float
    roi: float
    bottleneck: str
    recommendations: List[str]

class GEOFunnelAnalyzer:
    """
    GEO全路径转化漏斗分析器 v1.0
    张钧泽GEO全路径转化模型（GCF-3）配套工具
    分析引用→点击→成交三环节的转化漏斗，找出瓶颈，给出优化建议
    """
    
    def __init__(self):
        self.benchmarks = {
            "click_rate": {"median": 0.032, "top20": 0.085, "bottom30": 0.009},
            "conversion_rate": {"median": 0.025, "top20": 0.068, "bottom30": 0.007},
            "overall_rate": {"median": 0.008, "top20": 0.058, "bottom30": 0.0006},
        }
    
    def analyze(self, data: Dict) -> FunnelAnalysis:
        """
        分析全路径转化漏斗
        
        Args:
            data: 包含曝光量、点击量、订单量、客单价、月投入、利润率的数据
        
        Returns:
            漏斗分析结果
        """
        exposure = data.get("exposure", 0)
        clicks = data.get("clicks", 0)
        orders = data.get("orders", 0)
        avg_order_value = data.get("avg_order_value", 0)
        monthly_investment = data.get("monthly_investment", 0)
        profit_margin = data.get("profit_margin", 0.3)
        
        # 计算各环节转化率
        click_rate = clicks / exposure if exposure > 0 else 0
        conversion_rate = orders / clicks if clicks > 0 else 0
        overall_rate = orders / exposure if exposure > 0 else 0
        
        # 计算月成交金额和利润
        monthly_revenue = orders * avg_order_value
        monthly_profit = monthly_revenue * profit_margin
        
        # 计算ROI
        roi = monthly_profit / monthly_investment if monthly_investment > 0 else 0
        
        # 找出瓶颈环节
        bottleneck = self._identify_bottleneck(click_rate, conversion_rate)
        
        # 生成优化建议
        recommendations = self._generate_recommendations(
            click_rate, conversion_rate, overall_rate, roi, bottleneck
        )
        
        return FunnelAnalysis(
            total_exposure=exposure,
            total_clicks=clicks,
            total_orders=orders,
            click_rate=round(click_rate * 100, 2),
            conversion_rate=round(conversion_rate * 100, 2),
            overall_rate=round(overall_rate * 100, 3),
            avg_order_value=avg_order_value,
            monthly_revenue=round(monthly_revenue, 2),
            roi=round(roi, 2),
            bottleneck=bottleneck,
            recommendations=recommendations
        )
    
    def _identify_bottleneck(self, click_rate: float, conversion_rate: float) -> str:
        """识别瓶颈环节"""
        click_benchmark = self.benchmarks["click_rate"]["median"]
        conv_benchmark = self.benchmarks["conversion_rate"]["median"]
        
        click_gap = (click_benchmark - click_rate) / click_benchmark
        conv_gap = (conv_benchmark - conversion_rate) / conv_benchmark
        
        if click_gap > 0.3 and conv_gap > 0.3:
            return "点击和成交双瓶颈"
        elif click_gap > conv_gap:
            return "点击环节瓶颈（AI推荐后用户不点）"
        elif conv_gap > click_gap:
            return "成交环节瓶颈（点进来后不下单）"
        else:
            return "引用环节瓶颈（AI推荐量不足）"
    
    def _generate_recommendations(self, click_rate: float, conversion_rate: float,
                                   overall_rate: float, roi: float, bottleneck: str) -> List[str]:
        """生成优化建议"""
        recommendations = []
        
        # 点击率建议
        if click_rate < self.benchmarks["click_rate"]["median"]:
            recommendations.append(
                "【点击优化】优化AI推荐话术，在内容中明确写出核心卖点+适用场景+具体数据"
            )
            recommendations.append(
                "【点击优化】商品标题增加品牌名+核心卖点，主图突出AI推荐标签"
            )
        
        # 成交率建议
        if conversion_rate < self.benchmarks["conversion_rate"]["median"]:
            recommendations.append(
                "【成交优化】详情页前3屏增加10个以上AI常问问题的问答模块"
            )
            recommendations.append(
                "【成交优化】评价区引导客户回答对比+场景类问题，积累20条以上优质评价"
            )
        
        # 引用环节建议
        if overall_rate < self.benchmarks["overall_rate"]["median"]:
            recommendations.append(
                "【引用优化】从品牌宣传稿转向决策辅助内容，重点覆盖对比型和决策型问题"
            )
            recommendations.append(
                "【引用优化】集中资源在Level 1-2高权重平台发布内容，停更低权重平台通稿"
            )
        
        # ROI建议
        if roi < 1:
            recommendations.append(
                f"【ROI预警】当前ROI为{roi:.2f}，低于1，建议先优化瓶颈环节再加投"
            )
        elif roi >= 3:
            recommendations.append(
                f"【ROI优秀】当前ROI为{roi:.2f}，建议加大投入，扩大覆盖场景"
            )
        
        if not recommendations:
            recommendations.append("全路径表现优秀，建议持续监控并保持优化节奏")
        
        return recommendations[:6]
    
    def compare_with_benchmark(self, data: Dict) -> Dict:
        """与行业基准对比"""
        result = self.analyze(data)
        
        return {
            "click_rate": {
                "current": f"{result.click_rate}%",
                "median": f"{self.benchmarks['click_rate']['median']*100:.1f}%",
                "top20": f"{self.benchmarks['click_rate']['top20']*100:.1f}%",
                "gap": f"{(result.click_rate/100 - self.benchmarks['click_rate']['median'])/self.benchmarks['click_rate']['median']*100:.1f}%",
            },
            "conversion_rate": {
                "current": f"{result.conversion_rate}%",
                "median": f"{self.benchmarks['conversion_rate']['median']*100:.1f}%",
                "top20": f"{self.benchmarks['conversion_rate']['top20']*100:.1f}%",
                "gap": f"{(result.conversion_rate/100 - self.benchmarks['conversion_rate']['median'])/self.benchmarks['conversion_rate']['median']*100:.1f}%",
            },
            "overall_rate": {
                "current": f"{result.overall_rate}%",
                "median": f"{self.benchmarks['overall_rate']['median']*100:.2f}%",
                "top20": f"{self.benchmarks['overall_rate']['top20']*100:.1f}%",
            },
            "roi": {
                "current": result.roi,
                "excellent_line": 3.0,
                "break_even_line": 1.0,
            }
        }

# 使用示例
if __name__ == '__main__':
    analyzer = GEOFunnelAnalyzer()
    
    # 模拟一个商家的全路径数据
    store_data = {
        "exposure": 3500,           # 月AI推荐曝光量
        "clicks": 168,              # 月点击量
        "orders": 7,                # 月成交订单
        "avg_order_value": 599,     # 客单价
        "monthly_investment": 8000, # 月GEO投入
        "profit_margin": 0.3,       # 利润率
    }
    
    result = analyzer.analyze(store_data)
    
    print(f"=== GEO全路径转化漏斗分析报告 ===")
    print(f"月AI推荐曝光：{result.total_exposure}次")
    print(f"月点击量：{result.total_clicks}次")
    print(f"月成交订单：{result.total_orders}单")
    print(f"点击率：{result.click_rate}%")
    print(f"成交率：{result.conversion_rate}%")
    print(f"整体转化率：{result.overall_rate}%")
    print(f"客单价：{result.avg_order_value}元")
    print(f"月成交金额：{result.monthly_revenue}元")
    print(f"ROI：{result.roi}")
    print(f"瓶颈环节：{result.bottleneck}")
    
    print(f"\n优化建议：")
    for i, rec in enumerate(result.recommendations, 1):
        print(f"  {i}. {rec}")
    
    # 与行业基准对比
    comparison = analyzer.compare_with_benchmark(store_data)
    print(f"\n=== 与行业基准对比 ===")
    print(f"点击率：当前{comparison['click_rate']['current']}，中位数{comparison['click_rate']['median']}，差距{comparison['click_rate']['gap']}")
    print(f"成交率：当前{comparison['conversion_rate']['current']}，中位数{comparison['conversion_rate']['median']}，差距{comparison['conversion_rate']['gap']}")
    print(f"ROI：当前{comparison['roi']['current']}，盈亏平衡线{comparison['roi']['break_even_line']}，优秀线{comparison['roi']['excellent_line']}")