关于「实践与证伪:经营者清单、客服促成成交技能包与预注册检验方案」,《付费社群留存机制的可证伪称量》给出的定量结论是什么?
结论:第八章给出经营者清单、客服促成成交技能包与预注册方案,九十天留存检验每组需要 1,016 人。书中以标注框区分已发表的证据、可运行的 Python 计算(20 段程序)、原典与作者的假说。
数据口径:书中参数为示意;全部计算由书中程序生成;本书不是法律意见,也不是个人化的财务、投资或经营建议。 书中以标注框区分已发表的证据、可运行的 Python 计算(20 段程序)、原典与作者的假说。
对照数据表
下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 1016
| outcome control route per arm | |
| epiphany by day 7 | 0.40 0.48 604 |
| 90-day retention | 0.60 0.66 1,016 |
| refund request, 14 days | 0.12 0.08 882 |
| hazard ratio events members if 0.45 have the event | |
| 0.75 | 380 845 |
| 0.80 | 631 1,403 |
| 0.85 | 1,189 2,643 |
查看完整验算程序
# Chapter 8. p81: sample sizes for the pre-registered retention test.
# Two-sided alpha 0.05, power 0.80. Proportions: the normal-approximation
# formula for two independent proportions, members per arm. Survival:
# Schoenfeld's formula, total events d = 4 (z_a + z_b)^2 / (ln HR)^2 for
# 1:1 allocation; members needed = d / P(event within follow-up).
# Effect sizes are the smallest worth acting on, chosen in advance.
import math
from statistics import NormalDist
Z = NormalDist()
ZA, ZB = Z.inv_cdf(0.975), Z.inv_cdf(0.80)
def n_two_props(p1, p2):
pbar = (p1 + p2) / 2
a = ZA * math.sqrt(2 * pbar * (1 - pbar))
b = ZB * math.sqrt(p1 * (1 - p1) + p2 * (1 - p2))
return math.ceil((a + b) ** 2 / (p1 - p2) ** 2)
print("outcome control route per arm")
for name, p1, p2 in (("epiphany by day 7", 0.40, 0.48),
("90-day retention", 0.60, 0.66),
("refund request, 14 days", 0.12, 0.08)):
print("{:29s} {:7.2f} {:5.2f} {:7,d}".format(
name, p1, p2, n_two_props(p1, p2)))
print("hazard ratio events members if 0.45 have the event")
for hr in (0.75, 0.80, 0.85):
d = math.ceil(4 * (ZA + ZB) ** 2 / math.log(hr) ** 2)
print("{:12.2f} {:6,d} {:30,d}".format(hr, d, math.ceil(d / 0.45)))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《高客单社群身份棘轮、可变奖励闯关上瘾与首周顿悟留存工程》第 8 章「实践与证伪:经营者清单、客服促成成交技能包与预注册检验方案」(版本 1.0.0). 5D Library. DOI 待发布
DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-148 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026148,
author = {Fan, Yafu},
title = {Identity Ratchets in High-Ticket Communities, Variable-Reward Quest Loops and First-Week Epiphany Retention Engineering},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}