关于「里程碑解锁排程:目标梯度、前置的穷举最优与顿悟离习惯有多远」,《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:第四章给出前置的里程碑排程,并指出中位数会员在第一周只走完习惯平台的 0.2722。
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: 0.2722
| days to 95% | k per day | day 7 | day 30 | day 90 | half-way day |
| 18 | 0.16643 | 0.6881 | 0.9932 | 1.0000 | 4.2 |
| 40 | 0.07489 | 0.4080 | 0.8943 | 0.9988 | 9.3 |
| 66 | 0.04539 | 0.2722 | 0.7438 | 0.9832 | 15.3 |
| 120 | 0.02496 | 0.1603 | 0.5271 | 0.8943 | 27.8 |
| 254 | 0.01179 | 0.0792 | 0.2980 | 0.6541 | 58.8 |
| median person: first week reaches 0.2722 of the plateau | |||||
| median person: three quarters of the plateau by day 31 |
查看完整验算程序
# Chapter 4. p42: how far a first week gets toward a habit.
# Model: the asymptotic automaticity curve A(t) = 1 - exp(-k t), the
# shape Lally and colleagues fitted. A participant "reaches" the habit
# when A = 0.95, so k = ln 20 / T. T = 66 days is their reported median;
# 18 and 254 days are the ends of their reported range. The table gives
# the share of the plateau reached by day 7, 30 and 90, and the day by
# which half of it is reached. Curve shape from the study; this
# rendering is illustrative.
import math
print("days to 95% k per day day 7 day 30 day 90 half-way day")
for t95 in (18, 40, 66, 120, 254):
k = math.log(20) / t95
row = [1 - math.exp(-k * d) for d in (7, 30, 90)]
half = math.log(2) / k
print("{:11d} {:9.5f} {:6.4f} {:6.4f} {:6.4f} {:12.1f}".format(
t95, k, row[0], row[1], row[2], half))
k66 = math.log(20) / 66
print("median person: first week reaches {:.4f} of the plateau".format(
1 - math.exp(-7 * k66)))
print("median person: three quarters of the plateau by day {:d}".format(
math.ceil(math.log(4) / k66)))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 4 章「里程碑解锁排程:目标梯度、前置的穷举最优与顿悟离习惯有多远」(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}
}