关于「转介绍最大化:只为真实购买付一次的推荐奖金」,《合规分销与佣金斜率》给出的定量结论是什么?

结论:第四章算出利润最大的推荐奖金 158.5 元,每位起点顾客带来 1.1543 位推荐买家;下单即付时奖金的 0.7376 被作假者领走,退货期满才付则无人作假。

数据口径:最优的线性佣金斜率由努力成本、风险厌恶与噪音决定,示意基准下是 0.5263;给上线按下线业绩分成的二级佣金,主要买到的是拉人头而不是真实零售;按本书采用的最保守读法,以下线业绩给上线计酬本身落在 2005 年《禁止传销条例》列举的行为之内,合规方案的二级斜率应取零,转介绍改由只为真实购买、退货期满才发放的单层顾客推荐奖励来争取。 书中参数为示意;全部计算由书中程序生成;本书是非个人化法律、税务或理财建议,法规只述大意,以上架前人工复核的现行文本为准。

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: 1.1543, 158.5

bonus share refer R referred profit/seed
0.00.0500 0.0400 0.0417 12.50
20.00.1686 0.1349 0.1559 43.65
50.00.3193 0.2554 0.3431 85.77
150.00.6505 0.5204 1.0851 162.77
158.50.6698 0.5358 1.1543 163.33
200.00.7496 0.5997 1.4979 149.79
300.00.8714 0.6971 2.3019 0.00
profit-maximising bonus 158.5 yuan, profit 163.33
largest bonus keeping 0.9 of it203.5 yuan, referred 1.5271
referrals gained by that step: +0.3229
override b220: same referrals, profit 150.97 (-0.0757)
override b250: same referrals, profit 132.41 (-0.1893)
查看完整验算程序
# Chapter 4. p41: the per-purchase referral bonus that maximises profit,
# and the largest bonus that keeps 0.9 of that profit.
# A customer contacts up to 2 friends; a contacted friend buys with
# probability PI = 0.4. Referring costs the customer t, exponential with
# mean MU = 60 yuan; 5 % refer whatever the bonus. Share who refer:
# F(b) = 1 - 0.95 * exp(-b * PI / MU). Every buyer may refer in turn,
# so R(b) = 2 * F(b) * PI and referred purchases per seed customer are
# R / (1 - R). Gross margin per purchase M = 300 yuan. The bonus is paid
# only on a real purchase. An override b2 to the referrer's referrer
# changes nobody's own decision here, so it only adds cost on buyers
# from depth 2 onward. All numbers illustrative.
import math

PI, MU, F0, FRIENDS, M = 0.4, 60.0, 0.05, 2, 300.0


def reach(b):
    f = 1.0 - (1.0 - F0) * math.exp(-b * PI / MU)
    r = FRIENDS * f * PI
    return f, r, r / (1.0 - r)


def profit(b, b2=0.0):
    f, r, tot = reach(b)
    deeper = tot - r
    return (M - b) * tot - b2 * deeper


bs = [k * 0.5 for k in range(0, 601)]
bstar = max(bs, key=profit)
pmax = profit(bstar)
print("bonus  share refer      R  referred  profit/seed")
for b in sorted((0.0, 20.0, 50.0, bstar, 150.0, 200.0, 300.0)):
    f, r, tot = reach(b)
    print("{:5.1f} {:12.4f} {:6.4f} {:9.4f} {:12.2f}".format(
        b, f, r, tot, profit(b)))
floor = [b for b in bs if profit(b) >= 0.9 * pmax]
bhi = max(floor)
print("profit-maximising bonus {:.1f} yuan, profit {:.2f}".format(
    bstar, pmax))
print("largest bonus keeping 0.9 of it: {:.1f} yuan, referred {:.4f}"
      .format(bhi, reach(bhi)[2]))
print("referrals gained by that step: {:+.4f}".format(
    reach(bhi)[2] / reach(bstar)[2] - 1.0))
for b2 in (20.0, 50.0):
    print("override b2 = {:.0f}: same referrals, profit {:.2f} "
          "({:+.4f})".format(b2, profit(bstar, b2),
                            profit(bstar, b2) / pmax - 1.0))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《合规分销激励相容、二级佣金斜率委托代理模型与反传销层级红线内转介绍最大化母典》第 4 章「转介绍最大化:只为真实购买付一次的推荐奖金」(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-279 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026279,
  author    = {Fan, Yafu},
  title     = {Incentive-Compatible Compliant Referral Programmes: Commission Slopes in a Principal-Agent Model, a Second-Tier Slope of Zero and Maximal Referrals within Anti-Pyramid Limits},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论