关于「切出订阅层:刚需频率与计费周期之比」,《订阅叠砖、扣款挽回与年付拐点》给出的定量结论是什么?

结论:订阅层成立的条件,是每个计费周期内的刚需使用次数明显高于值回票价的门槛。在参差人群里,比值要到 2 左右,首次续费率才到七成;人群整齐时,1.31 倍左右即可。即使使用频率很高,参差人群里仍有一部分人整个周期不用,首月流失有一个由构成决定的底。示意参数下,按月订阅层要到每月 6 次刚需使用才胜过买断;按季计费多收的钱,有相当一部分来自不怎么用的客户,这部分收入最脆弱。

数据口径:书中的订阅数、客单、流失率、挽回率、折现率与价格均为示意参数,全部计算由书中程序生成,不对应任何真实客户或成交数据;本书是非法律、会计、税务或投资建议,不承诺任何收入水平,「稳定」只指收入可预期并写明前提与失效条件。 第一至三章说明,稳态收入等于每月新增乘以客单再除以月流失率,新增定斜率,流失率定天花板;订阅者异质时总体留存随月龄上升,移位贝塔几何模型用几个月的队列数据就能外推;前两次续费是筛子,单次干预回报是后段的 1.40 倍,三年总增益却在长尾,示意参数下是 18,598 对 4,760;订阅层只在刚需使用次数明显高于值回票价的门槛时成立。

English version

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: 1.31

uses per cycle m renew u*1 u*=2 u*=3 (shape k = 1.5)
0.50.350 0.107 0.031
1.00.535 0.256 0.117
2.00.719 0.479 0.307
3.00.808 0.615 0.455
4.00.858 0.702 0.561
6.00.911 0.803 0.696
8.00.937 0.858 0.775
mean uses needed for a first-renewal rate of 0.70
u*1 k = 1.5: 1.85 k = 10: 1.28 ratio m/u*: 1.85
u*2 k = 1.5: 3.97 k = 10: 2.62 ratio m/u*: 1.98
u*3 k = 1.5: 6.08 k = 10: 3.92 ratio m/u*: 2.03
share with zero uses even at m8, k = 1.5 0.063
ratio m/u* for u*2 when k = 10 1.31
查看完整验算程序
# Chapter 3. p31: first-renewal rate as a function of must-have usage.
# Uses in one billing cycle are Poisson with a person-level rate that is
# Gamma-distributed (shape k), so uses are negative binomial with mean m.
# A subscriber renews only if the cycle held at least u* uses that
# justify the price. k, u* and the target rate 0.70 are illustrative.
import math


def p_at_least(u, m, k):
    if u <= 0:
        return 1.0
    p = k / (k + m)
    terms = []
    for x in range(u):
        lg = math.lgamma(x + k) - math.lgamma(k) - math.lgamma(x + 1)
        terms.append(math.exp(lg + k * math.log(p) + x * math.log(1 - p)))
    return 1.0 - math.fsum(terms)


def m_needed(u, k, target):
    lo, hi = 1e-6, 200.0
    for _ in range(100):
        mid = (lo + hi) / 2.0
        if p_at_least(u, mid, k) < target:
            lo = mid
        else:
            hi = mid
    return hi


K = 1.5
print("uses per cycle m   renew u*=1   u*=2   u*=3   (shape k = 1.5)")
for m in (0.5, 1.0, 2.0, 3.0, 4.0, 6.0, 8.0):
    print("%15.1f  %11.3f  %5.3f  %5.3f" % (
        m, p_at_least(1, m, K), p_at_least(2, m, K), p_at_least(3, m, K)))
print()
print("mean uses needed for a first-renewal rate of 0.70")
for u in (1, 2, 3):
    print("  u* = %d   k = 1.5: %5.2f   k = 10: %5.2f   ratio m/u*: %4.2f" % (
        u, m_needed(u, K, 0.70), m_needed(u, 10.0, 0.70),
        m_needed(u, K, 0.70) / u))
print("share with zero uses even at m = 8, k = 1.5    %.3f" % (
    1.0 - p_at_least(1, 8.0, K)))
print("ratio m/u* for u* = 2 when k = 10                %.2f" % (
    m_needed(2, 10.0, 0.70) / 2))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《月度经常性收入叠砖、扣款失败自动挽回与年付现金流锁定母典》第 3 章「切出订阅层:刚需频率与计费周期之比」(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-367 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026367,
  author    = {Fan, Yafu},
  title     = {Stacking Monthly Recurring Revenue, Involuntary-Churn Dunning Recovery and Upfront Annual Cash Lock-In},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论